I have put a few notes and links up in various other more private / obscure places on the Geek, but this post serves as the formal "announcement" of a game design project / idea I have recently begun putting some serious effort into. This blog will be the place into which I'll start moving all development notes and, more importantly, playtesting info. I've got at least 1 member of the my GCL group interested in playtesting, but I need to recruit a few more willing testers. To that end, I need to introduce the game, I suppose.
What's this game all about, then?
Ties that Bind: 1850-70 is a railroad-building game set in the 1800s that follows the development of early railway lines in the United States, beginning in the Northeast and Midwest, expanding to the newly reunited Southern states, and culminating in the completion of the Transcontinental Railroad. It is generally a member of the broad 18xx family of games, which is "the collective term used to describe a set of railroad-themed stock market and tile laying games". This game follows the general 18xx game plan with various concessions to speed and ease of play.
18xx games are notoriously long and unfriendly to beginners to the genre. Their rulebooks are full of little exceptions and weird edge cases to remember, the games tend to run very long and have a lot of tedious math, and most of the titles open with an inscrutable and confusing auction round that is unlike anything else that follows and depends on players having internalized lots of information and tactics about the game that will follow. There have been a lot of attempts to streamline the genre for newbies, but most still run 3-ish hours and don't really do anything about the fiddly bits, edge case rules, overwhelming math, or confusing initial rounds.
My goal with this design is to make a "true" (more on that later) 18xx game that can be played in 90 minutes or less and that is streamlined to the point where some of the larger tactical considerations and broader strategic trends in this style of game can be appreciated and leveraged by players from the very first play without having to grok all manner of technical minutiae just to get past a "learning game" (or to do long division in their head!). I also want to create a game that's worth playing in its own right, of course.
What elements of 18xx are here?
If you're not familiar with the 18xx genre, here are a couple of essential elements that I've maintained in this game:
- The game is a train-themed game focused jointly on financial investments and railroad operations, with the performance of railroad companies impacting the value of the company stock held by players.
- The game progresses in alternating stock rounds and operating rounds, with companies acting at the direction of their majority stock-holder and keeping assets that are separate from player assets.
- The game has a 3-act / 3-phase development arc, in which the board state becomes increasingly complex and destinations increasingly valuable over time, and with implications for asset requirements.
- The game has significant decisions to be made in both the stock and operating rounds, and the opportunity to invest in or take over other companies, or to arrange to dump bad assets on others.
- The game plays on a hex map with linear track connecting locations, with railroad networks that grow both in length / size and in value, and the opportunity to block others out of various destinations.
- The operating rounds follow the familiar tracks-stations-routes-payments-trains order of actions from 18xx, although a number of these elements are implemented with variously significant changes.
- Companies choose either to pay out dividends or to withhold earnings, with implications and impacts on both company assets and stock market value, as well as player cash holdings and portfolio appreciation.
- Company actions (at the direction of their controlling players) determine the pace of the game, including the development of revenue opportunity on the board and the "rush" to buy up bigger and better trains.
- Player positions are differentiated at the start of the game by initial stock investments, but also by the acquisition of special cards which provide varying income, recoupment potential, and bonus abilities.
- Players compete in multiple ways: Better managing the growth of the companies they control, choosing better exit strategies from bad investments, deliberately sabotaging the assets of other players, etc.
So what makes this 18xx different?
A summary of some of the biggest changes and simplifications from 18xx "proper" that I've made toward my design goal:
- Instead of a convoluted initial auction or weird draft or even a fancy packet-assignment procedure for distributing initial "private company" / variable-player-powers, I just use a deal-2-choose-1 method.
- Rather than these private companies being something with multiple aspects to consider (resale value, ability, map-blocking, etc), here they are "characters" that only do one easily understood thing each.
- The game is played in strict turn order in both stock and operating rounds. In a typical 18xx, turns bounce around in the operating round as players act for their companies, which is an often awkward conceit.
- The most complex math required to play the game is doing small sums and performing single-digit multiplication. This is felt especially in the stock transactions and dividend payouts, which are usually tedious.
- All of the track is made up of Catan-style "road" pieces rather than 18xx's hex tiles. Searching for the right tiles, making sure they're available (or exist!), rotating them to fit--this all takes a lot of time.
- The trains aren't a deck of cards to cycle through and they don't "rust" in the typical way. Instead, each company has a supply of train "tokens" they can pay to increase, and occasionally is required to buy more quickly.
- Companies always only ever run 1 single unbranching route laid out with physical tokens. In all other 18xx, companies have to finagle multiple overlapping abstractly-traced routes, 1 for each train card they hold.
- Rather than trains determining phase changes, the development of networks on the map triggers the phases. Slow play in burning through the train deck can make 18xx games stagnate, but players always want to lay track.
- There is something along the lines of the weirdly termed "company credits" from the very first 18xx titles / rulebooks that makes it even more obvious which assets / currency belong to players and which belong to companies.
- The endgame triggers are not lengthy timers and are mostly approached monotonically (running out of a large pile of money is the typical end condition for new 18xx players, and certain play patterns can push it further away).
No, I know 18xx. What is really different?
Ok, so here are some of the more important rules-diffs and MacGuffins that diehard 18xx players might appreciate:
- The requirement to float a company is a single share, and the game is full-capitalization. Companies fly out onto the board from the very first turns with a range of possibilities for initial values / capital stocks.
- It's a simple 1-dimensional stock market with the possibility for double-bumps in operating rounds and double-"drops" in stock rounds if a president sells shares in a company (other players just drop it once).
- It's just dividend vs. withhold, but the company receives a 1x dividend payment so long as they hold at least 1 stock from their initial supply. There's no separate IPO / market / bank pool; it's all just the company.
- Private companies ("characters") are held by players at all times, and provide either income when a player's companies do given things, add value to their routes for various achievements, or allow once-per-game powerful effects.
- There is a late-game event (the Civil War) that destroys the track on nearly half the board, as well as the most lucrative offboard location. Companies have to pay out additional resources if they want to rebuild those routes.
- The board has a random setup for the starting location for half of the companies and all of the initial location values. Combined with the random selection of privates, this is intended to make up for any replay value lost in the shortened length.
- There's a "union station" mechanism that creates dits on the board whenever two or more railroad networks meet for the first time. These come in values that increase as the game goes on, and don't count against a company's route length.
Sounds vaguely ok. What's the current status?
I'm so happy you asked!
There is a fully fleshed-out version 0.000something rule book here that you can peruse.
It's an 18xx game at heart, so the rulebook is long and detailed. "Streamlined" is a relative term.
I've also been working on a fairly player-friendly Vassal module for the game.
Here's a little ditty I made up to show off the module and a few rounds of play:
In terms of development, I have only solo playtested (multiple hands--this isn't and won't ever be a solo game) it.
The game plays and it definitely has moments of interest, sufficient that I've been able to maintain my interest in the project long enough to go through quite a few rounds of iterative improvement and get rid of some really bad ideas.
The Vassal module also is 90%+ functional at this point, handling most of the aggravating fiddly stuff and the harder-to-remember rules automatically (pretty much everything is automated except the board play), so it's ready to go, too.
It's in a state where it's really in need of some full-table playtesting with actual human players who are not me, and who hopefully are at least vaguely familiar with this style of game (so you have a good point of reference to compare to).
That's where you come in!
If you're interested in trying out a few (likely asynchronous) games with me on Vassal, let me know below. If I get a few willing playtesters, I hope to start arranging some sessions this week. (You'll probably want to be familiar with Vassal already.)
At this point, I don't have any particular playtest questions that I'm trying to answer other than "does this thing work in general in a realistically competitive setting or does it just completely fail to function in ways I had not previously noticed?"
(Of course, since I've just given you the link to the rules and the module, you could obviously just go off and try it out with whomever you damn well please. It doesn't bother me if you do, but I'd really like to be involved in games at this early stage.)
So, that's where I am. Any takers?
Even if you aren't interested in playtesting it, if you want to read through the rules and point out grammatical errors or confusing bits or really stupid ideas I've yet to excise or just ask questions or make general comments, I'd greatly welcome that too.
Often Lumbering No-Nonsense Ludological Observations
Archive for Nate Straight
Introducing "Ties That Bind: 1850-70" -- Initial "design diary" and playtester call for a work-in-progress streamlined 18xx-like game design project
14 Jun 2021
- [+] Dice rolls
25 Jan 2021
So round 1 in my tagalong participation in John Owen's giveaway concept was pretty successful. Got two games sent out to folks who will hopefully give them more love than I have. I've got another one here to send off, for similar reasons as the last go around, but with a new twist for 2021.
Owing to the recent racist-adjacent comments from Daniele Tascini and the subsequent decision by Board&Dice to drop all future publications with him (if you want to, you can try to wade through the morass that is the thread on the topic here, but I've seen enough), I'm ridding myself of his games.
Fortunately, I only own one such title, and that's what I'm giving away: Sheepland.
It's complete and in good condition. The box is mostly air so may get dinged in transport--I'll try to stop that.
It's a cute enough little game, being sort of a speculation and investment game wrapped up in an absurd thematic setting. It works pretty well, but my kids (whom I mostly picked it up for) didn't really get it, so it's not going to see much play around here especially not with my new distaste for the designer.
If you don't care about such things (and, yes, I realize that there are likely numerous other designers who are racist that I own games from who simply haven't made that clear on social media) or can dissociate the game from it, you're welcome to have it. I'll cover the first $20 shipping to anywhere you like.
These are the conditions of the giveaway:
- You must actually want the game and have reasonably good hope to get it played. I don't want it taking up space on your shelf like it is mine. You must comment here why you want the game and who you hope to play it with. If there are multiple requests, I'll choose whose story I like best.
- You must either own or purchase a Black Lives Matter microbadge (I'll send you the GG for it if you don't want to spend your own) and commit to displaying it for a month. You must also state unequivocally in your request for the game that black lives matter. (You can feel however you like about Tascini's comments.)
So, who wants it? I'll pick the recipient on Feb 14th.
Black lives matter
EDIT: Thanks to a huge GG tip from a reader, I've got plenty of GG to go around. If you want to sport a BLM microbadge, even if you don't want the game, post below and I'll send you the requisite GG.
- [+] Dice rolls
So there's this slowly growing giveaway trend going on, kicked off by John "trawlerman" Owens in this geeklist. Kirk Roberts (aka "kirkroberts" ) and QwertyMartinQwiffiths have continued here.
I'm jumping into the fray as well with a couple of giveaways of my own. The rules are pretty simple:
1) You have to tell me why you want the game and promise to play it as soon as "the new normal" permits.
If you don't mind, comment below to indicate your interest in the game and explain why you're interested in it.
2) You have to send me your address so I can send you the game. Free to CONUS; I'll pay first $15 elsewhere.
In about two weeks time, I'll announce the recipients of this first lot and contact you for shipping details.
Magical Treehouse. Despite its low rating, I really think this is a game with an audience. I thought that audience was me, but it definitely isn't. It's a real-time drafting game (but it's super thinky so not particularly a frantic real-time experience) with a bit of a combo element, a bit of an engine element, and a bit of a spatial element.
The age range listed on the game page and the age range voted on by the community are batshit insane. This is in no way a game for 6 or 8 year olds (ask me how I know!). AEG appears to list 14+, and that's about right and should give you an idea of the relative complexity of the stuff going on here, of which there is a lot.
You will need a group that's willing to work through arcane symbols, weird game structures, and the "what the hell was THAT?!" feelings first-plays sometimes generate. If you don't have a group that's willing to push through a bit of murkiness, this game is definitely not going to do anything for you, so please don't ask for it.
The copy you'll receive is the AEG English edition with cute little meeples for the board instead of the little moon-shaped cardboard tokens from previous Japan-only editions. So far as I know, there aren't any other changes. The game is in excellent condition. Cards still snap and look great, but you might want to sleeve them.
It's a really beautiful production that has some really interesting ideas. I just don't think I'm going to be able to find a group willing to put the effort in to get through the learning curve and start to have fun with the drafting and card effects / combos / spells / etc. I think this game would appeal to microgame fans, despite being bigger.
Río de la Plata. This one I am part of the audience for, but it needs a high player count (would be best with its full 5) of like-minded players willing to commit a serious amount of time and repeat plays (it's a pretty simple game, but the rules are quite a bear). I haven't played it in almost 10 years and don't see that changing.
It's a weird constructive-interaction game that plays kind of like a game of Puerto Rico [sans role-selection] if every player were adding to a shared board instead of a player board. And then every few turns somebody gets to be the pirates and raze half the city for points. It's weird and wild and the attacks interject interesting instability.
It's kind of like a pseudo-cooperative game where there's a shared threat that has to be defended against, but players are still trying to win while they keep "the team" from losing... except that the shared threat ends up being controlled by the players themselves (each player ends up getting a turn at being the attacking pirates).
This is a big game (not in components, though... so don't expect a typical modern "look at all of this cool shit!" Euro--the recently Kickstartered re-boot Trinidad is closer to that). It'll take you at least 3-4 hours to get through your first game, I think, despite it being listed as a 90-minute affair. You'll need multiple plays to "get it".
It's a little slow, definitely not much to look at, but I really liked it the couple times I managed to play it waaaaaay back at a BGG.con in the way before-times. I would keep it if I had any inkling that it might get played again. It can be a pretty nasty game, in that you're going along engine-building and then everything burns down.
So, there you go. Tell me why you want one and who you plan to play it with.
- [+] Dice rolls
13 Aug 2014
back in Frankfurt!more of my stuff at www.dennisthebadger.com
So one of BGG's incomparably talented graphic artists, Dennis Bennett, has released to the public his incredibly cool "Badger Deck" recently via PrinterStudio.
The full Badger deck is 10 suits with ranks running 0 to 20, plus A-K-Q-J, W-F [Wizard-Fool], and five more lettered ranks [Castle, Monster, Princess, etc].
The deck also includes special artwork on some of the 'special' cards [plus one additional card in that suit] in one suit that allow it to be easily used for Tichu in particular.
Suffice to say, you can play essentially any ranks-and-suits game ever devised [even many that require doubled ranks, so long as there are fewer than 10 numbered ranks, since the 11-20 ranks are helpfully designed to look like the standard ranks], and do so with an incredibly beautifully rendered deck in a cutesy fantasy / chibi style.
Providentially, until this Friday, PrinterStudio is running a site-wide 20% off + free shipping sale, which will allow you to get the entire deck [Dennis has released it in two parts: "0-11 with specials" and "12-20 add on"] for right around $26 USD shipped [don't know about international shipping]. That's a sickeningly good deal and you should take advantage of it.
Here are the links:
Printer Studio Badger Deck - Part 1
Printer Studio Badger Deck - Part 2
The code for 20% off + free shipping is "20OFFSWIDE". I've been having trouble getting PrinterStudio to add both items to my cart simultaneously, so I had to check out twice, but you can use the code both times so it all works out the same. [Let me know if you find the trick to getting the cart to work, and I can update this blog post for others.]
Go and get this awesome deck before the code expires, tell your friends, and consider sending a bit of GG or $$ Dennis' way to thank him for this gift to the gaming community. There have been innumerable attempts to put together a "play any card game" style deck, but this is far and away the nicest looking and most well thought out effort I've seen.
Dennis did not compensate me in any way for this "promotion", nor did he request it.
For my regular readers who are depressed this isn't a "real" post: Look for an attempt at game-type classification by victory / scoring condition as well as a reader-requested post on systems theory and gaming in the months to come.
- [+] Dice rolls
Around the World in 80s Days
The year is 1980 and our hobby [at least on the continent I am writing from] consists largely of the sprawling wargames of SPI and The Avalon Hill Game Co, the mostly abstract and parlor style bookshelf line from 3M, a healthy smattering of negotiation classics like Junta or Machiavelli or Dune or Cosmic Encounter, and the first big wave of fantasy titles such as Magic Realm or Titan or of course Dungeons & Dragons. A faint rumbling of a "chugga chugga" or a "wheeeesh" can be heard from such titles as Railway Rivals, Boxcars / Rail Baron, 1829, or even Dispatcher. It wouldn't be until later into the 80s with the reimplementation of Railway Rivals as Dampfross and development of 1829 into 1830 that these titles would really gain popularity and influence.
Beyond that, none of these early train-themed games bear much resemblance to what we think of today as a "train game" [not even the seminal 1829, although 1830 does; the difference, I think, is in the very restrictive way in which 1829 introduces both new companies and new track into play]. Railway Rivals / Dampfross is more akin to Streetcar: A connection game followed by a race game. Rail Baron / Boxcars is more akin to Merchant of Venus: A roll-and-move game of buying infrastructure and making deliveries; sounds about right, and MoV is this close to being a train game, but Rail Baron not only has pre-set tracks but limited delivery contract / route options. Dispatcher is a minutia-laden game more akin to the logistics in a wargame.
Chugga Chugga Choo Choo
So if all of those games aren't exactly "train games", what exactly is a "train game"? And why does anyone care? Similar questions are currently being fielded by Jason Begy in his dissertation on train games for Concordia University. I'm not going to attempt to go into nearly as much detail in my research as what Jason is likely to end up with, but I want to share a bit of the story of Empire Builder and its important place in hobby gaming. Along the way, I hope to place it as a transition point from these earliest train games to something recognizable as "the modern train game", and obviously to describe a fair bit of what that term means. So let's begin at the beginning: What is a train game? Well, we'll get to that later. First, let's answer something easier:
What is a train?
The train is one of the more revolutionary inventions in history, and was a critical part of the various industrial revolutions that took place throughout the 18th and 19th centuries. Being able to produce widgets more effectively and efficiently [through the use of steam power, better machinery, increased understanding of materials and their properties, etc] is one thing, but to make widgets you need "stuff". You need lots of fuel to burn to power the machines. You need lots of raw materials to put into the machines. You need a way to transport the widgets to people who want to buy them. The efficiency of your economic engine is limited and defined by the amount of "stuff" you can move in and out, and the speed at which you can do it.
Trains had [have] a number of obvious advantages over other methods of moving "stuff" between producers, suppliers, and consumers: They were [for the time] faster than any other method of transport, and dramatically so [the Transcontinental Railroad in the US cut the travel time from the East to the West Coast from a matter of months down to 4-5 days]. They were [and still are] able to carry dramatically more payload than any other method of land transport. They require relatively little manpower and are relatively fuel efficient [especially when considering the tonnage of freight it is able to move]. Add it all up, and a train can carry more goods at distances further removed for a cost much lower than nearly any other mode of transport.
There is, of course, one piece of the equation which we are missing. Trains only run on track. Even though trains are fast and both fuel and cost efficient, train tracks are much less so. It took about 6-7 years to build the Transcontinental Railroad, and that was after decades of debate over where exactly it should go and who exactly should build it. The construction employed well into the tens of thousands of laborers. The total cost for the project was something around $100M in 1870 USD. That's just the main route. Every spur to stretch anywhere else in California but the Bay area, in the West but central Nevada and central Utah, and in the East but the Great Plains was yet another very expensive and time-consuming prospect.
This is the primary limitation on the effectiveness of trains. The infrastructure to support rail transport is expensive and slow to build, and increasingly so as it becomes less geographically centralized. For this reason, rail transport is most well suited to bulk transport between large centers of economic activity. Short runs and small loads are best left to other modes of transport. In the early history of rail transport [when UPS trucks and 18-wheelers were not a "thing"], much of this break-out traffic was handled by private local and regional railroads which connected smaller towns back to the major public transportation hubs. This was not generally very profitable, and inevitably these companies [and their track] were bought out by the larger railroads.
* A train moves "stuff" for the purpose of facilitating economic activity, and is an incredibly efficient and rapid means of economic transport.
* A railroad is an incredibly expensive and difficult bit of infrastructure, and is a limiting factor on the effectiveness of transport by train.
These, to me, are the defining elements of "trains" and the defining tensions involved in their deployment which a "train game" must capture.
There must be economic activity or growth which is furthered by the use of the trains or railroads in the game. There must be a tension between the "easy money" efficiency of running a train and the "hard labor" difficulty of building [or running] a railroad. There must be significant topological or geographic tradeoffs involved in the planning of rail connections or delivery routes. Niceties such as technological progression and industrial activity are welcome, but not essential. Side issues such as stock ownership and financial chicanery are an integral part of what "train game" means to many gamers, but this is mostly because these mechanisms in their full force just happen to have been applied to "train games" proper nearly to the exclusion of any other genre.
Enter Empire Builder.
First edition copy, photo credit Donald Dimitroff
The Little Engine that Could
Industry stalwart Mayfair Games has been around much longer than has its signature title CATAN. The company was founded in the early 80s for the purpose of publishing Empire Builder, an unassuming looking game co-designed by company founder Darwin Bromley [the other designer was Bill Fawcett]. Since then, the game has seen five revised editions, and over a dozen spin-offs in the "crayon rail" series. Its basic mechanisms for network expansion were copied directly in Funkenschlag [Power Grid's predecessor]. As a standard-bearer for network-building games in general and for Mayfair Games' entrepreneurial adventures as one of both the early "German game" importers and the early "train game" publishers, Empire Builder is a linchpin in the hobby.
If you have never played or researched Empire Builder, a brief overview is probably in order. The essence of the game is pretty simple: Take a natural resource map [with the cute little icons all over] for a country, overlay it with a hex grid made up of dots [the hex centers], then have players connect the dots [literally] with a grease pencil to indicate where they have built their tracks. The players use these tracks to run a train token back and forth between connected cities delivering various natural resources and industrial goods between their origin city and a randomly determined [through a contract card draw deck] destination, with payouts based on the distance between the cities involved. Repeat 'til someone has a sufficiently large network and sufficiently high cash assets.
This has a kind of intuitive common sense to it which I've written about before. The maps make sense. The kind of economic activity makes sense. The method of building track by drawing out the way you want to connect various cities makes sense. The means in which you make money by delivering goods from a fixed supply to a larger and variable set of demanders makes sense. Because of this, and best of all, even though it is simple it still feels like a real live red blooded train game [I'm looking at you, Ticket to Ride] and it captures the essential elements of the genre: Building track is expensive and takes a long time, but your train can zip back and forth across what track you've built pretty quickly; and there are real opportunity costs in how you plan both your network and deliveries.
The game also has one of the best early examples of what is now a pretty bog standard trope in train gaming [though 18xx largely eschews it]: Your income derived from economic activity during the game is for the large part of the game plowed right back into continuing the cycle of economic growth through more and more infrastructure [track]. It accomplishes this by forcing you to connect a certain number of larger cities, but more importantly by limiting how efficiently your wealth can grow if your network is not large enough to handle the vagaries of the random contract draw or to leverage the longer more valuable payouts. You simply will never reach the required total cash-on-hand victory condition without expending about as much [or more] cash over the course of the game to expand your network.
Like many modern economic engine games, there is a fairly well-defined tipping point when the focus of your game will shift from building the infrastructure / engine to churning out points with it. Along with Crude: The Oil Game, Outpost and Civilization, it's one of the earlier and more important examples of this basic game structure. Additionally, while it is not the first pickup and deliver game with players "carrying" goods on some kind of transport across the board to earn points [for one other easy example, see Alaska from a year earlier], it is definitely the earliest of any real impact and is pretty damn close to the archetype for that mechanism, what with movement points and carrying capacity and demand cards to fulfill. Merchant of Venus, Roads & Boats, Serenissima, and all of the other pu&d progenitors owe a large debt to Empire Builder.
More specifically, Empire Builder solidified a central element of pickup and deliver gameplay that has become part of the vast majority of games [especially train-themed games] which use the mechanism and which [importantly] is absent from all of the previous train-themed games mentioned in the intro: The idea that any given demand can be sourced from multiple supply locations, and [conversely] that any given supply can potentially be used to fulfill demands in multiple different locations. In prior games that had inklings of pu&d, delivery contracts were of the sort "go to that place, then come back to this other place". In Empire Builder, it is "Newcastle needs coal", but there are multiple places to get coal, and if you're not connected to any you'll have to decide which one works best with other potential deliveries.
That idea comes back in full force, of course, in Age of Steam and its family tree [parents and children]. While there are no trains which move around the board and demand is determined by the board topology itself and not a random contract draw, the essential tension remains of linking up a network various potential supply sources to various possible matching demands in such a way as to efficiently take advantage of as many as possible of the produced goods when they present themselves to be supplied to fulfill demands. If there is any other game series which is quintessentially the "modern train game", it is the AoS family, and while the supply/demand system is not a direct port in any sense, it is hard to imagine its having developed in the way it did without the earlier [possibly mediated] influence of Empire Builder.
I am less convinced / certain of it [i.e., would love counter-examples], but I am fairly sure that Empire Builder also represents one of the earliest instances of the concept [divorced from any physical pick-up-and-deliver action] of collecting a good[s] and using it to match up to a randomly drawn "contract" in order to gain a payout. This, of course, has developed through a little bit of mechanical accretion [turning in more than one good at a time in order to receive the payouts for the contracts in question] into a building block as basic to modern gaming as is worker placement or shifting turn order: set collection [esp. toward "contracts"]. Despite being abused and having become one of the most obvious design choices when you need to convert "doing stuff" into points, it is a fundamental design tool.
Mayfair booth at Origins 2007, photo credit Todd Eaton
It's Hard to Stop a Train
Empire Builder paved the way for Mayfair to become one of the most successful hobby publishers ["publisher marks" at least, as it hasn't been continuously owned/operated as the self-same company]. Importantly for the topic of this post, it also cemented them as one of the foremost publishers of train games. In addition to all of the Empire Builder spin-offs, Mayfair was responsible for the importing from Hartland Trefoil of two of the earlier 18xx titles [1835 and 1853], for the sprawling economic epic that is Silverton, for two independently / in-house developed 18xx games [1856: Railroading in Upper Canada from 1856 and 1870: Railroading across the Trans Mississippi from 1870], for a number of train-themed card games [Express, Freight Train, Station Master], and eventually [after a few corporate reorganizations] the functional [if not perfect] long-awaited reprint of the classic 1830: Railways & Robber Barons.
Add to that a number of Martin Wallace economic titles [Automobile, Aeroplanes, and of course Steam] and dozens of games licensed from KOSMOS [Catan, duh], and they are for all of their price-fixing and generally uninspired art and production direction a force for good in the hobby. In large part, no doubt, because of the head start which Empire Builder serendipitously gave them. What would Mayfair have been like without Empire Builder's success? What would your gaming life have been like without Mayfair's landing of Catan? They're not a Lookout Games [well, actually...] or an alea or a Hans im Glück that dominate the BGG hotness with new flashy releases every year from the most popular designers, but they remain a hobby standby seemingly in large part on the backs of tried and true designs like Empire Builder and Catan.
There's Room for Many a' More
I'll wrap up by trying to connect the breadcrumbs I've scattered throughout, primarily as it pertains to this thing I've called the "modern train game". It's a theme and a setting that seems to fascinate gamers regardless of their general gaming preferences. Titles like Rolling Freight, Snowdonia, Russian Railroads pop up every Essen and score high on the buzz lists / hotness tracker. Are these train games? Does it matter? Well, in all likelihood, no... but that would make this a waste of a blog post. I think it does matter that we define "train game" to mean something more than just "it has pictures of trains on the box", because there are implicit and conditioned things that are expected of something purporting to be a "train game". I'm going to toss in my two cents on what these are, without answering silly "Is ____ a train game?" questions.
Does a train game require a map? Surprisingly perhaps, I don't think so, but I do think it requires some kind of implied distance structure or other topology [which does not have to be geographical / physical; it could well be an abstract "network"]. I can imagine a free-form action-cost-as-distance game structure such as Container's being used in a game that is otherwise a train game. I can imagine some kind of "card adjacency" [not physical, but card-power / -value related] mechanism utilizing ranks and suits to denote distances traveled. And I can imagine a way to build an infrastructure that interacts with these systems without having "physical presence".
Does a train game require trains? Again, maybe a little less surprising, my answer is no. The basic concepts of very expensive or time-consuming infrastructure leading to very efficient growth all in the framework of a production / supply-and-demand economy are manifest clearly in actual trains and railroads, but also in various other settings. I consider the board play of Guatemala Café and Hacienda to be in many ways like (I wouldn't go so far as to say either is) a train game. Nearer to the Platonic ideal, Medieval Merchant and Power Grid seem very nearly to be train games. Then there's Samarkand which is also close. Merchant of Venus has at least one foot in the door.
Does a train game require network-building? Yes, or at the very least network modification [MoV's factories and spaceports, for instance, though again it's an edge-case example]. This is why at least some kind of distance structure or implied topology is necessary. Connectivity [and the building thereof] is central to what it means to have a "train" or a railroad. It is not merely a source of income [as in, say, Railroad Barons], but a source of mobility. That concept of mobility and connectedness and its impact on economic activity and growth is a sine qua non of train gaming, and a differentiating factor from mere economic engine games or mere financial simulations.
Does a train game require goods / pickup-and-deliver? No, although it does require some kind of implicit "from here to there" economic activity that is dependent on the connectivity just mentioned [for example, 18xx's counting of train routes which must pass through a station and then can "reach" to a number of connected cities of various economic value]. So, Martin Wallace's quip in the rulebook to First Train to Nuremberg that "with all good train games, there comes a time when you actually have to move things" is I think a little unfair [*wink*]. There are many ways to simulate economic activity and the "movement" of goods back and forth which do not require any movement.
Does a train game require stock-holding? No. Just no.
So, there you have it. For me, the modern train game is something at the intersection of network-building and economic-engine [to put it simply]. More than that, it is about the tension between hard-to-put-together and iteratively-expanded infrastructure [of uniform materials, so no combo-building super-power card-games here; rails and ties, there's your winning combo] and the rapid growth which it can facilitate. Because the infrastructure is expensive, it cannot reach everywhere in the game and significant opportunity-cost style decisions must be considered in the making of connections. The game pace needs to be long enough for the durability of the infrastructure to be felt [we are not merely blazing trails; a railroad is a permanent landscape fixture] and its use should be a large part of the game.
It also helps if the graphic design was done in MS Paint.
- [+] Dice rolls
13 Mar 2013
I actually will have something substantive to say about the project, but will withhold such commentary--to be updated here--until it ends, since the way the project's end is handled will likely greatly impact my impression of it. So far, I remain hopeful.
- [+] Dice rolls
Hobby Game Designer Compendium, Part 2--Plains, Trains, & Automobiles; or, The League of Progressively Meaner Gentlemen
02 Feb 2013
Folks on BGG often ask questions like "What is the best Reiner Knizia game for me to get if I like auction games but want something thematic?" or "Is Through The Ages representative of Vlaada Chvátil's style? Could I like Galaxy Trucker even if I hated TTA?" or "What is it that makes Stefan Feld's games so wildly popular among Euro gamers?"
There is an implication, probably accurate, behind all of these questions that these and other designers have something uniquely theirs that distinguishes their work in much the same way that content creators are distinguished in other media: architecture, art, comics, dance, music, writing, whathaveyou. It is usually something obvious yet nameless.
We as gamers lack the type of robust classification and criticism system that other artistic media have [though we're making progress toward it], so it is difficult to explain what makes, for example, Knizia and Sackson and Colovini and Schacht's games similar, let alone what differentiates them from the Kramer / Breese / Dorra or the Georges / Feld / Dorn crowds.
I hope in this [two-part] post to catalog some of the more popular and stylistically consistent designers that I am familiar with and to try to give form to nebulous concepts such as "a Wallace game".
There will be two parts to each entry: 1) A 10-word or less description of the designer's style that highlights their most defining traits; 2) A longer narrative of the mechanisms they are known for.
I'll also try to reach outside of each individual designer's own catalog to identify some games that might be seen as influences on or extensions of their design style as I've identified and described it.
A man with a face any mother could love, Alan Moon designs games about:
Teaching interesting life-lessons about decision-making... oh, and choo-choos.
Just barely edging out brass-knuckled Martin "Railways of the Industrial Age" Wallace, Alan Moon has the distinction of being the only designer whose top four ranked games are all the same game, and whose top ten ranked games actually consist of only five games. In fact, it is rumored that Alan Moon has only had one original game idea in his lifetime, and that that game is still being developed.
Alan Moon is also one of precious few designers [and here I'm actually being serious] whose work actually demonstrates any indication that he, you know, plays games instead of just thinks about them a lot. His games are for people who grew up enjoying games they weren't supposed to without knowing that they were breaking an unwritten geek code, for people who at one point kind of thought Rack-O was the pinnacle of strategy and secretly still think Monopoly "isn't that bad".
His games are interesting, yet easy to learn; intense, yet easy to play; interactive, yet easy to enjoy in a low-key manner. They're games you can play with your grandma and still [both] have fun and who the hell doesn't need more games like that? Among the two basic game ideas that Alan Moon has had in his lifetime [whoops, I underestimated] are competitive set-claiming and bizarrely difficult bidding... hey, that sounds kind of like a traditional card game! Yes, that is correct; every Alan Moon game is really Rummy, Bridge, or Poker in disguise [actually... that's not all not true].
In Ticket to Ride, the players draw cards into their hand, and then meld them when they have a seven of the same rank... no, wait... when they have five pink [purple? lilac? fuchsia?] trains. They then string these melds together [literally] in an attempt to get four concealed pungs... no, wait... a series of train routes that connect pairs of cities on cards they hold in secret. They then... no, wait... that's about it. Let's play! It is actually my contention that this isn't a Rummy game at all since the family is built on the draw-and-discard model of iteratively refining your hand toward a meld, and Ticket To Ride is just spam-drawing. Of course, that leaves me with concluding that Rummikub [and its public domain version, Vatican] are also not actually Rummy games, but I'm ok with that. Regardless, Ticket To Ride is almost as simple to learn and play as Rummy, and shares much of the same gameplay interest: Being efficient in your card draws / selections from the "discard pile"; satisficing with what you've drawn instead of holding out for "the big one"; trying to get that third Queen and discovering that the bastard across the table [i.e. grandpa] is holding the other two that you don't have. It's just fun.
In Das Amulett, the players signal and revise their intentions through a 2-level bidding system in which they compete for the right to name trum-... no wait, for the right to improve their hand [in the first level] and determine what resources will be won at auction [in the second level]. Ok, so this game is really nothing like Bridge, but it representative of a particular type of auction game that is distinctively Alan Moon's, being equal parts brinksmanship exercise and battleground, and that type of auction strikes me as being incredibly similar to the bidding mechanisms and card play found in traditional trick-taking card games, and particularly the more involved examples. Das Amulett, and others like Wongar or New England or Elfenwizards, are games of bidding not only for valuable lots of assets but for control over some aspect of the shape of the game or round or turn, whether it is preferential turn order, or control over what the next auction will be, or some other desirable status or privilege. Richard Breese also does a lot of design in this area in the "Key" series, but I would peg it still as a distinguishing feature of Alan Moon's bidding [and bluffing, see below] games, particularly those co-designed with Aaron Weissblum.
In Oasis and San Marco and quite a few others, the players make offers [in various forms] to each other, usually with an element of bluff concerning exactly what was offered or why. Both in making and in receiving these offers, there is a high degree of double-think that I find is also characteristic of traditional card gaming, and particularly of "beating" card games like Poker or Durak. In Oasis [following in the steps of Nicht die Bohne!, I suspect], players take it in turn to make an "offer" of resource cards; after the offers are made, each player in turn order will pick one of the other player's offers and receive their turn order button for next round. The players low in turn order will get sloppy seconds this turn, but hopefully will have offered enough to get priority of choice for the next. Oh yeah, they have to make these offers randomly. In San Marco, players have unique roles to play: One player serves as the divider / offerer, and one or two others serve as the chooser [yes, the same method choosy moms choose to let choosy kids choose peanut butter sandwich halves]. in splitting up the resources available each turn, the offerer must be reasonably equitable [or the chooser will leave her with nothing], but also conceal her desired lot [so it isn't taken out from under her].
And so on.
Another element common in Moon's designs [and appropriate to this perspective] is gambling or press-your-luck. In Ticket to Ride, as a basic example, you always have the option either of selecting a card whose identity / color you know [even if it's not ideal] or of picking random cards from the top of the deck and living with the potluck. Additionally, near the end game [or before], you have the choice to take a risk and plumb the depths of the destination ticket stack in hopes of something useful you already have connected.
In games like Diamant / Diamant, Clocktowers / Capitol, and Andromeda, the element of gambling is considerably more pronounced and overt. These games all share elements of "holding out" on closing off an open opportunity [or "putting in" for more shots at a probabilistic gain, alternatively] for as long as possible in hopes of getting a bigger and better return. His games ask you to make good choices and live with the consequences of poor ones.
Lest you think Alan Moon is all fun and games, he has quite a few relatively heavy games in his catalog, including a few wargames and some heavier train games [though by no means approaching the end of that genre's weight spectrum, or even the midpoint really], and a number of his games in the genres above [San Marco, in particular] can be quite thinky.
Games that share traits with Moon's designs include Aladdin's Dragons, Cleopatra and the Society of Architects, Show Manager, and Thebes.
Sid Sackson (1920-2002)
A man who opened arms and minds, Sid Sackson designed games about:
Simplicity and strategic depth whenas rarely the twain did meet.
Though he did not have as many published games as Knizia, as consistent and prestigious a career as Kramer, or as prolific a ludological writing career as David Parlett, it seems near impossible to overestimate the significance of Sid Sackson's love for designing, analyzing, collecting, playing, and cataloging board games.
When Sackson's design career began around the 1960s, hobby gaming was chock full of monster wargames [led by Avalon Hill's early catalog] and near mindless family games [all of the ones we love to hate now: Life, Yahtzee, Battleship, Risk, Aggravation, etc were designed around this time, building on the legacy of Monopoly and Clue]. Sackson's designs were something different: Short, snappy, mathematically elegant, strategically demanding, focused on indirect player interaction, and structured in ways that kept everyone involved.
In short, there is good reason why he is largely considered the "original" Euro game designer. He had compatriots in this effort and design thrust, most notably including Robert Abbott and Alex Randolph. [Bruce Whitehill of www.thebiggamehunter.com has some very nice write-ups on both Sackson and Randolph that each discuss quite a bit about this early hobby game design community.] But, it was Sackson who would have the biggest impact through his incredibly diverse, yet stylistically consistent output. His games take minutes to learn but provide great depth and variety.
In Bazaar, there are exactly three things you can do on your turn: 1) Roll the die and get a gem of that color; 2) Make a trade of gems from either side of a set of 10 equations [randomly chosen each game from a larger pool of available trade equations] to receive whatever is on the other side; and 3) Turn in a set of gems that matches a card to claim it for points [you can always do this, whereas you either pick #1 or #2, but not both]. The fun of the game is that the goal is to eventually claim the set-collection goal cards with as few surplus gems as possible; this seems like it would not be possible when your options all consist of getting more gems, but [aha!] some of the trades let you trade down from a large assortment of cruddy gems to the 1 or 2 that you actually [hopefully] need. It sounds easy, but managing the 20 different directions of trades and figuring out how to piece them together to get exactly the set of gems you want and no others is incredibly difficult. The game isn't for everyone, as it's essentially one giant puzzle with interaction consisting of a race to claim as many goal cards as possible, but it's quite challenging and always presents something new.
In Monad, there are also essentially three things you can do on your turn: 1) Trade a set [pseudo-restricted] of 2 cards of the same value for 1 card of the next higher value; 2) Buy a card of any value by paying a set of other cards that sum up to its value; 3) Turn in a set of 4, 5, or 6 different colored cards of the lowest value and claim a higher valued card. You can do any of these as many times as you like and are able to on your turn. The goal is to start with just a few of the piddliest valued cards and maneuver so as to get a certain number of the highest valued cards. This is not easy, especially because of the restrictions to option #1 [which aren't really relevant in a summary overview] and the funkiness of the values / distributions. This is not quite as much a pure puzzler as Bazaar, as there is not only a race element but a setting-up-your-opponents element for interaction. When you trade in or pay out cards, you have to return them to a common supply, and this supply is incredibly limited; whenever you throw stuff back in, you're making it available for the next player. There's a lot of tempo control and maneuvering so as to keep good cards away from the opponents; some small rules for bonus moves and wild cards just make it that much harder.
In Acquire, there are also three possible things you can do on your turn: 1) Play a tile to the board from your hand [you must do this]; 2) Buy up to 3 shares of stock in the growing chains of tiles on the board [you may do this]; 3) If the tile you played merges 2 or more chains of tiles already on the board, decide what to do with your shares in those chains--this option also has three options: 1) Keep 'em, 2) Trade 'em 2 for 1, or 3) Sell 'em. The heart of the game is in trying to gain majority holdings in the stocks that you think will have the largest presence on the board at the end of the game; you get bonuses for having this majority, and also can get mid-game bonuses for majorities if you own shares in a smaller chain that is getting bought out by a bigger one. From these simple rules comes a multi-faceted game of positioning, pushing and shoving, and stealing from the work of other players. This game turned 50 years old last year, and at the time was still in the top-100 [it is #109 at the time of this post]. I personally consider it among the top 3 or 4 hobby games ever designed, for its elegance, replayability, depth, and interactivity. It is a perfect game if ever there was one.
And so on.
One of Sackson's wonderful abilities was to take the oldest and often stalest of gaming paraphrenalia and breathe new life into them, from Dominoes to Dice, from Checkers to Cards. In his [sadly now dated] A Gamut of Games, he shares quite novel designs of his own and other provenance for all of these, in addition to pen-and-pencil, poker chips, vocabulary, and very simple homemade components. It is my belief, and one I think is borne out in reality, that Sackson probably could have made an interesting game out of pretty nigh any components and any restrictions.
This is a skill not to be underestimated. There are precious few modern designers who show this foundational understanding of what a "game" is. Among them, I would place the likes of Reiner Knizia, David Parlett, and Wolfgang Kramer. You could empty your purse / wallet / pocket in front of one of these gentlemen, give them an hour or two to dig through all of the stuff, and very likely you could return and be playing a game with the detritus of your daily life that, if not particularly deep, was entertaining and provided meaningful choice to players.
Sackson shows the importance of a [mathematically] functioning game system behind whatever cute mechanisms or compelling story is tossed into the box with the rules. Call it dry, if you like, but for all of the games in the world of hobby gaming that are simply broken, uninteresting, too easy to do well at, but hide all of this behind a veneer of exciting chrome or fun twists on worker placement or action selection, there are some hundred plus Sackson games that need none of this to provide a compelling game experience.
Games that share traits with Sackson's designs include Ingenious, Chinatown, Lost Cities, Backgammon, and the Series: GIPF Project games.
A man who embodies Eurogaming, Michael Schacht designs games about:
Small choice spectra intersecting with deeper and wider decision analysis.
More than any other designer [even Sackson], Michael Schacht is enamored with limiting player choice to a very small pool of options. One would be hard-pressed to quickly identify a Schacht title in which a player has more than, say, ten or twelve possible play options to choose from on their turn [including, in large part, trivialities such as "should I put this tile in location A, B, C, D, etc?"], and many have considerably fewer.
It is not that his games are simple or light; they are very often quite difficult to play well, in fact. For this reason, his games play very quickly and he is sort of a darling of the "under an hour, but with some depth" crowd. He gets a lot of play out of relatively few mechanisms by imposing fairly detailed limitations on and higher level interactions between them. He never provides a shortage of things to think about in considering your meager choices.
He also likes to toy around with shared spaces and/or resources to the point where the end result of one player's turn defines the starting game state that affects the next player's turn in a fundamental sense, and to a much greater extent than otherwise similar games where players merely bump heads or send out feelers. Players don't just "interact"; they flirt, touch, and dance. Much of it is coy, subtle, even unintended, but that's what makes it fun, of course.
In Hansa, the players share control of a single trading vessel sailing around Scandinavia and surrounding areas. Players are very limited in both the number of times they may move the ship [1 gold per, from a pool of around 6-8 in an average turn], and the direction in which it may go [routes on the board are all one-way]. The game consists of managing two interacting resources: trading posts and barrels. You can pick up barrels which you can later spend [as supplies] to build trading posts in a city; later still, you can spend those trading posts to sell off matched sets of barrels [for points]. You can only do one action per turn per city, so you have to setup a series of moves for future turns in order to really make any progress. Of course, the trouble is that you don't get to decide where those turns will begin; the player before you in turn order will simply say "hey, it's your turn" after they've finished moving the ship, and you have to pick up from that point. There are only so many ways to play out your turn, but each one has radical consequences for your opponent and for your flexibility in future rounds.
In Web of Power / China, the players build up networks of cities through regions on the board and compete simultaneously for political control of diplomatic pairings between regions. Turns follow the "3-2-1 rule": You can play up to 3 cards [your entire hand] to play 2 pieces in 1 region. This is a highly restrictive rule, and players have to conserve resources and manage their small hand very tightly in order to get the most leverage out of their limited options. The two resources [houses and ambassadors] are, of course, interconnected; the number of ambassadors that can be played in any given region depends on the number of houses that the current majority holder has there. This makes it very likely that by competing for majority control of the house scoring, you will open up your opponent for taking control of the political side of the region on the very next turn. Additionally, the first player to build into a region is only allowed to play 1 piece [a house] rather than the standard allotment of 2; of course this means the next player can easily add 2 houses in the newly opened region and immediately attain majority.
In Coloretto / Zooloretto / Aquaretto, the players [theme notwithstanding] attempt to collect a large number of 3 or 4 categories of items [from a pool of about twice as many] while collecting none at all of the remaining categories. A player's turn consists either of adding an item [randomly chosen, but freely assigned] to a standing offer [one per player, but shared publicly] or taking one of the offers and dropping out of the round. This iterates over a dozen or so rounds, and that's essentially the entirety of the game. The balancing act here is that players will of necessity be sharing wants, so when you try to fill up an offer just with stuff you want to get and nothing you want to avoid you will often be helping somebody else out as much as yourself [and they'll snatch all the goodies right out from under you]. There is also an element of cutting one's losses if a round of offers does not seem to be developing well. Zooloretto, Aquaretto, and a Coloretto variant add a restriction that within your chosen categories there is a limit on how much you can collect without penalty, so that feast and famine are bad.
And so on.
Schacht is also known for his mini-expansions and even mini-games that are available primarily on his publishing website, Spiele aus Timbuktu, but occasionally see publication by major publishers as promos or expansions. Because his games are so small, mechanically, they lend themselves well to variation and expansion. These aren't the kind of expansions that you collect and use all at once for "the full game" experience; they're more "variations on a theme".
A few other designers [including some on this list] handle expansions / variants similarly, but they often fall prey to the trap of simply making an entirely new game out of a minor change to a basic concept. Schacht has a few spin-off series [the --oretto series from above, also Valdora and Africana], but he usually tacks his new ideas and twists onto existing games in his catalog rather than developing them out as a new property.
This is nice, because there's less cost to the player interested in seeing where the designer has taken a favorite game. Schacht comes across, in fact, as one of the more "player-friendly" designers active in the hobby. He is generally active in all of his games' forums, and he regularly posts about new [free!] content for his more popular designs. If you are a print-and-play fan, you'll find much to love in Schacht's catalog.
Games that share traits with Schacht's designs include Biblios, Diamonds Club, Guatemala Café, and Portobello Market.
A man, a myth, a legend, and a lifetime achiever, Klaus Teuber designs games about:
Ramshackle settlements, family feuds, deserts, droughts, and pistols at dawn.
Klaus Teuber designed Catan. The single most-rated [>36000], most-owned [>45000] game in BGG's catalog. The game that for many represents the entire hobby of gaming. The nearly 20 year old game that really, when it gets right down to it, bears absolutely no resemblance to anything past or anything that has come since. The game that was a influential on the hobby as something like Magic: The Gathering or Dungeons & Dragons. It is, put simply, a pivotal game. And, it remains important.
Catan is so much of a phenomenon that people seem to forget that Teuber has done anything else [especially when so much of his future work has been tied, and understandably so, to the Catan license]. Yet, Teuber has also been successful in other gaming ventures. He has won 4 Spiel des Jahres awards, more than any designer side from Kramer. None of the other 3 SdJ titles [Catan is the 4th] from Teuber are even remotely related to Catan. Even the Catan-themed spin-offs of Candamir and Elasund and the "historical" titles are far removed.
Then, there are the games in the "Anno ____" series, based on the PC video game series, including large multi-player games and 2-player face-offs. And the games in the [family=127]Entdecker[/family] series, including two spin-offs and two versions of the original. And the two version of Löwenherz. And a ton of other games that, mostly, have never been exported from Germany. Teuber is not a one-hit wonder by any stretch of the imagination. Still, a large number of his games share some of the same traits as are familiar from Settlers: Exploration, trade, negotiation, meager means, and competitive land parceling.
In Drunter und Drüber, one of Teuber's SdJ winning titles other than Catan, the players attempt to clear an otherwise functional town of everything except outhouses and one preferred type of building per player [which they are dealt secretly at the beginning of the game and keep secret until the endgame]. They do this by laying tiles progressively from each corner of the board, extending lines of river, rails, roads, etc until all of the relevant transportation departments [longshoremen, railmen, highwaymen?] have exercise all of the powers of eminent domain that they can muster. Whenever eminent domain would conflict with the town's love of outhouses, a vote is held; the players negotiate among themselves and vote on whether the construction is allowed to bulldoze through their precious latrines. Of course, you would want to vote against construction that starts to move toward your preferred buildings, but no one has the same preferences. Yes, it doesn't really make much sense at all, but in this game from way back in 1991, you can begin to see a lot of the features that would wind their way into Teuber's later, better designs.
In Löwenherz, one of those better designs, the players put to test the maxim that good fences make good neighbors. The game board is a nameless and modular medieval landscape peppered with mines, castles, hills, and forests. Players begin with castles also scattered across the board, and lay down knights stretching out from their castles to claim off land through force, and fences between squares of the board grid to claim land through more neighborly means. Of course, the problem with fences is that they're so easily knocked down! Once a plot of land has been completely walled off and claimed by a single player, they can extend out to weaker neighboring provinces using the power of their knights; though, if the neighbor has a stronger army in their territory, they can of course block this assault and instead return kindness for kindness. The game functions through a deck of three-part action-cards: Part A is a "collect money" [used to build knights] action; Part B is a "build knights" action; Part C is a "build fences" action. Only one player [from 4, ideally] each turn can do each action, and players declare intentions in turn. If two players want the same action, they have to negotiate with each other or, failing friendly resolution, "duel" in a simultaneous bid. Oh, and points are zero-sum after initial awarding: If someone takes over your territory, you lose the points and they get them! This is a very mean game.
In Elasund: The First City, one of the slightly-more-like-Catan-than-not spin-offs in the series [though still a good stone's throw away], the players build up [surprise!] Elasund, the first city of Catan! This is considerably more complicated a process, it turns out, than what the ancestral game would have you believe; no measly 3 ore and 2 wheat will suffice here. Instead, what happens is that players first lay claim to their preferred parcels of land, then put together enough resources and land claims to build one of the many buildings needed for the city, and eventually harvest additional resources and power from those buildings to develop things elsewhere. Oh yeah... and you can build your shit right over the top of someone else's building or land claim! All you have to do is have better shit than what they had! If you help build the church for the city, you have even more preferential claim to previously built land! Of course, the robber element from Settlers shows up here, and it's just as mean. This and Lowenherz are delightfully nasty games of intimidation and pulling the rug out from under each other.
And so on.
One of the frustrating things about many of Teuber's designs is the meanness of the game systems, however, and not the players... "mean" here in the sense of ungenerous and miserly, not mere nastiness. It is appropriate that his designs often have desert terrain, as dry spells and droughts are common occurrences. This is because his clever mechanisms of die-rolling and negotiation for meting out resources are fickle and subject to wild variation. It is quite possible to go many rounds without actively doing anything on your turn; this can turn some players off and is strange as a feature for the man who is considered to have veered gaming away from Monopoly, et al.
In fact, I would argue that Settlers has considerably more in common with Monopoly than it does not, and probably more in common with that maligned title than it has with most modern Euro games. Settlers is a snowball, rich-get-richer game. It is a game of total unfairness. It is a game of kingmaking and spite. It is remarkable that it ever caught on if what Euro gaming is about is keeping everyone involved, playing nice, and having fair outcomes not influnced by luck. Settlers is anything but that! And, I would argue, the best Euro games are anything but that.
Strangely, as I hinted at in the intro, Settlers has never really been copied in any meaningful sense. Certainly not in the sense that Caylus, Age of Steam, Puerto Rico, or any of the other "big" Euros have; even Carcassonne has been copied and ported more successfully. There are, to be sure, resource-collecting hex-map games like Keythedral and Roads & Boats or trading games like Bohnanza or Genoa, but the combination of mechanisms and interplayer dynamics that Catan offers is, still, quite unique.
Games that share traits with Teuber's games include In the Shadow of the Emperor, Saludos Amigos!, Santiago, and Gonzaga.
A man among boys, Francis Tresham designs games about:
Beating the snot out of each other in unconventional ways
Tresham is among a very elite cadre of designers like Sackson, Greenwood, or Teuber whose output and game-related activity has had radical and lasting impact on the shape of the entire hobby that stretches beyond the mere substance of their designs. Standing like two monstrous guardians on either side of a narrow strait through which one passes into a seemingly endless and ever expanding swell of train and civilization games are his category creating and inestimably influential 1829/1830 and Civilization.
1829 was the first serious and economically oriented railroad building game. 1829 was also probably the originator of all train games to follow wherein the routes and track on which trains run are built by the players rather than predefined or predrawn on the board [perhaps competing with Railway Rivals released a year earlier which would inform the crayon rail series and later become Dampfross, and aside from simplistic tile-laying connection games like Main Line or Rivers, Roads & Rails where "track-building" is itself the sole goal of the game and there is no board].
Civilization was among the first [if not the first; After the Holocaust: The Nuclear Devastation of America – Recovery and Reunification and Empires of the Middle Ages are in contention] games to combine warfare with economic development, and almost certainly the first to do so under the now familiar theme of "beginning at the beginning" with a stone age [or worse] "civilization" and progressing it through the earliest stages of human development until [at least] the earliest inklings of modern political society that we find in classical antiquity. In Civilization is also embedded perhaps the earliest instance of the "economic engine"-building that would weave its way through Outpost into Puerto Rico and then explode.
Tresham's games are remarkable for their ingenuity, as well as their epic scope. Despite their length and high thematic immersion, Tresham's games [one exception noted later] are unlike the monster games of today that are nearly always full of luck at key decision points. The Series: 18xx line that 1829 and 1830 birthed are perhaps the pinnacle of luckless, high-strategy economic games. While Civ technically has a mild element of the unknown in the assignment of calamity cards, their position in the deck is known at setup and theoretically fully trackable from beginning to end.
His recent game Revolution: The Dutch Revolt has a similar thematic underpinning as the more traditional wargame, Here I Stand, but does not share the same dice- or card-driven combat structure; instead, it has a luckless area control system. Only his almost completely unknown titles Shocks & Scares and Spanish Main have significant chance elements. Despite the lack of "dramatic" chance events, his games retain a sense of narrative and, more importantly and [perhaps to some] unexpectedly, a high level of direct player interaction.
In 1830, the players interact in two novel ways: By sharing ownership [and, so, both risk and return] of fledgling train companies, and by using the control [or abdication thereof] of those companies as a means of attacking or diminishing or otherwise inhibiting the financial position of their opponents. Although stock ownership mechanisms are nothing new [Acquire, for instance, predates the 18xx by a few decades], the particular type of stock ownership mechanism used in 1829 / 1830 is still woefully unfamiliar and underused. Players' stock portfolios are not merely investments [to be bought low and sold high], but are a means of effecting control of the organisms that operate on the game board. Players themselves do not perform actions on the boards; companies do. And only the player who owns the most shares in the company whose turn it is gets to say what that company does. [Among recent games other than the 18xx and certain other train games, only Imperial--to my knowledge--shares this structure.] Obviously, the player serving as president of a given company will do whatever is in her own [as a player] best interests, and not necessarily what is in the best interest of minority shareholders or even the corporation itself.
Besides acting as a means of control over the course of the game [and as an investment that can appreciate in value and often pay regular dividends], stocks are also a method of directly impacting another player's financial position. There are a couple of ways that this can happen, but the nastiest is to drain a company that you own of its productive assets [or simply allow them to go to waste], then sell off your entire controlling interest in the company [at currently super-inflated market value, of course] and pass off the hulking mass of inoperability to a minority shareholder in line for the presidency. In the most extreme of cases, that new president will be on the hook to salvage the company using personal assets, not be able to pay to do so, and go bankrupt [which ends the game]. Even in less extreme scenarios, the new president is left holding shares that are now worthless [your sale instantly corrected for the too-high valuation in the open-market]. The game is a dance, then, between wanting to share in the benefits of buying in to the most profitable companies and needing to avoid this type of liability as a minority shareholder if the president suddenly decides personal gain is more important than corporate performance. Few games pit player incentives and liabilities against each other in as direct a way.
In Civilization, the players largely get on fairly peaceably; the game even [like a modern area majority game] allows units from multiple players to live in the same territory simultaneously... to a point. Beyond that threshold, combat takes place in what is among the simplest and most direct combat resolution systems in all of gaming: the player with fewer units removes one first, then the player with more units removes one, then the player with fewer, and so on until the number of units in the territory is again below the threshold [I guess they just forget they had been fighting after that, since they immediately go back to peaceable relations]. Combat is not the focus of the game, obviously, but is instead a means to an end. Trading and development is the focus of the game. Units eventually transform into cities; for every city, players gain increasingly more valuable resources; using those resources, players trade [a la Pit or Catan] and then purchase technological developments [worth VP toward winning the game]. Besides combat and trading, the most inventive method of interaction here are calamity cards "hiding" [at preset positions] in each deck of resource cards. If you draw a calamity card, you can attempt to offload it in a trade [claiming or implying it to be something else]; if you're stuck holding it at the end of the trading round, you suffer the consequences.
As in the 18xx, the goal of negative direct interaction is to assail an opponent's economic position and opportunity; this takes place in Civ through combat or trading of calamities, and the effect is to reduce the collection of resources by an opponent. Part of the turn cycle are two versions of a now familiar "feed your population" type of mechanism; in the first, a player must have unit tokens in their supply equal to twice the number of cities they have on the board; in the second, a player must have two units on the board for every city. If either condition is not met at the appropriate time, the player will lose cities and the ability to draw the most valuable resources. This is a difficult enough balancing act to deal with in isolation; it is made increasingly more difficult by well-timed combat or well-traded calamities. Knocking off a few opposing units through combat could be enough to eliminate one of your opponent's cities, denying their ability to draw a much needed resource to complete a valuable set. Destroying a dozen by calamity can be devastating. While the game's basic flow is a typical "build up your infrastructure and then leverage it for victory points" chain of collecting, trading, and spending resources, the very high marginal cost of infrastructure coupled with its fragility makes the game quite dramatic.
In Revolution, the players take the role of 1 of 5 different competing interest groups [religious or political factions] involved in the economic, political, and religious struggle in the late 16th and early 17th centuries for the area that is now the Netherlands. As in the similarly [but more grandly] themed Here I Stand, these factions have nearly completely asymmetric goals and resources. The game is an area majority game writ large. The essence of the game is quite similar to Civilization's combat mechanism, where the reaching of a threshold of population in a region triggers a removal of units one by one, starting with the weakest faction, until the threshold is once more unmet. Where it differs from Civ is in having a much more detailed system for extorting controlled territories, creating new units, moving existing units, messing around with other players' units, and employing alternative methods of assaulting an opponent's control of a region. The uniquely novel twist here is that each faction in the game has comparatively more or less control over the ability to use any of the multitude of ways of influencing a region or city. Some factions get cheaper armies; some get more armies; some can hire mercenaries; some have objectives that are easier to defend; some harder; etc.
And so on.
Besides the general "epic" scope of his games and the emphasis on luckless direct interaction, the overarching mechanical theme of Tresham's work is the trade-off between active and potential assets, and in particular on the transformation between the two states. In all of the above games, your economic resources exist in "raw" form or embedded in some productive asset. The basic resource is "money" in each game. The productive assets in 18xx are shares and trains; in Civ the assets are units and cities; in Revolution the assets are military units and various other influence.
The quirk that is uniquely Tresham's is the rate and frequency at which assets cycle back and forth between these two states. In 18xx, the game progresses through "stock rounds" in which money is converted into shares [later in the game, vice-versa] and "operating rounds" in which shares pay out dividends and money that was paid for shares is converted into trains [which drive dividend payouts] in an endless cycle. In Civ, every round has phases in which reserve assets are transformed into spendable cash and to units on the board, then from units on the board into cities [which, in turn, increase both the resource productivity of the civilization and the rate at which reserve assets become cash].
In both games, productive assets often get liquidated back to their raw form. In 18xx, trains can be sold between companies to move large sums of cash around, and shares are often dumped at opportune times to net the appreciated gains. In Civ, cities may dissolve and be transformed back into units, and units may be liquidated to construct ships or cities or resolve events. Also of note is the "holding company" status of intermediary assets: train companies own and spend cash paid for shares in 18xx, and armies of units eventually "upconvert" embedded reserve assets into cities in Civ.
Games that share traits with Tresham's designs include Age of Renaissance, Die Macher, and Indonesia.
A man who loves your misery, Martin Wallace designs games about:
Keeping your investments from being derailed at every unexpected turn.
Martin Wallace is a big meanie, but not the overtly bullying sort. No no no; instead he butters you up first with the promise of great prosperity like some ARM-hawking mortgage broker who forgot his morning coffee, then ratchets up the required monthly payments until you're underwater, underleveraged, and under a lot of pressure from brass of all sorts to get your act turned around [nevermind that your demise wasn't even your fault].
The mechanism he is best known for is, appropriately, the taking of loans [which very often literally cannot be repaid]. To remind you of your inevitable financial distress, the tokens representing loans are often a friendly pitch black color or else penciled in bright cheery red. They are also given neighborly euphemisms like "poverty", "trouble", or the joyfully abstract "loss"; but no, they're all still debt... dirty, stinking, irreconcilable black holes of sucking debt. That slurping sound was all your liquid assets being drained dry. Martin Wallace drinks your milkshake.
It isn't just that the economies in Wallace's games have thin margins and are prone to predictive errors that necessitate short-term solutions in the form of a high interest payday loan. No no no; once again Wallace is not so kind. Instead, his games place you in the role of an upstart "idea person" in need of venture to realize [you foolishly hope] your dreams. It is not uncommon to have to begin a Wallace with literally no operating capital or assets, and the need to immediately go into debt in order to get off the ground. Of course, you will take on more of it later, often every other turn [if not every turn outright]. The "debt" systems he deploys are often quite nuanced and implemented in unexpected ways.
In Age of Steam, the players manage train corporations from initial venture funding to sprawling rail empire. The would-be Railroad Tycoons have no company assets and no cash... just the promise of profitable return. Players must begin the game by immediately extending a certain number of "shares" [to nameless investors not appearing in this film] for which they receive $5 each... and must pay $1 each in interest at the end of every round for the remainder of the game. The principal amount is never repaid by your company [these are shares, mind you, not bonds or loans], but the indebtedness is inescapable [you can never reduce your share issuance] and if you cannot pay interest out of a current round's revenue stream or your company coffers [retained earnings], your future income will be garnished. You can easily go bankrupt in the first few rounds through poor planning. The system requires a tight balance between cash inflows and outflows, that involves planning out revenues and cash needs in advance while servicing the minimum amount of debt necessary to fuel that economic engine.
The subtlety of the game, however, lies in the interaction of this system with the victory conditions; victory in the game is assessed [as it is in large part for most real-world business valuations] as discounted future earnings, where the estimate of future earnings potential is equal to the income of your economic engine in the final round of the game and the discounting factor is a $1 reduction in that income estimate for every share issued. There is a wealth of opportunity available in the game for near unbridled growth, in absolute terms, of your empire; the problem is that if you take out enough debt to fund all of that opportunity right away, the net effect will be a loss in the short-term and severe devaluation at the game's end. You need some absolute growth, but what you really need is a better rate of marginal growth in revenue vis-a-vis your growth in interest expense than that which your opponents can manage. It's a struggle to figure out how to put together long routes [which are more profitable] without simply bonding out all of the cash to pay for them up front; you also need to finance the incredibly important turn order auction.
In London, the players manage the rebuilding of London after the Great Fire, and continue to develop their part of the city on into the 20th century. It is a tableau-building game like Race for the Galaxy, Through the Ages, or 51st State. The key feature distinguishing it from these games is the need to intentionally limit the rate of growth of your tableau beyond the systematic / game-imposed limits set by your rate of resource collection and production capacity. As you lay down cards in your tableau and then activate them, you will receive a "poverty point" [i.e. foreboding black cube] for every card you have played; however, since most cards "flip over" after their initial activation, you are allowed to limit the growth of your poverty level as you lay down new cards by "stacking" them on top of previously-played cards [flipped or not]... poverty is technically collected, then, per stack rather than per card played [it's helpful for this post to think it through backwards like this; it's presented in the rules directly as one poverty point per stack, of course]. You would, of course, like to have a huge array of cards to activate all at once, but that will net you a ton of poverty, so you stack.
Poverty itself isn't an absolute evil; you can theoretically have as much of it as you like and still win the game. At the end of the game, what will happen is that each player will discard an equal number of poverty points [equal to the total that the least poor player has] and then be penalized for any amount remaining above this level; that is, for their relative poverty level as compared to their opponents' burroughs of the city. Oh yeah, we should talk about those burroughs; the chief way to manage poverty is by buying up portions of the city [to house the poor, I guess]. This is essentially a "VP grab" / "resource grab" option; you get to draw some cards and collect some VPs by paying some money. The long-term benefit is that every burrough you own in the city reduces your poverty allocation by one point every time you activate your tableau of cards. [A secondary way to manage poverty is through a number of cards that simply remove some of your amasses poverty points.] Oh, there are of course "normal" loans in the game, too, where you simply get a loan counter, an amount of money, and then pay interest every turn until you repay the principal.
In Automobile, the players build, market, and [hope to] sell automobiles. The game is one of staying ahead of a technological arms race to have the newest, shiniest cars and of accurately guessing at the demand for cars, old and new. You are penalized, in the form of "loss cubes", for not keeping pace with progress and for being over-anxious with both production and marketing. "Loss", here, is much less like debt than "poverty" is in London, since there isn't any immediate gain associated with it [in fact, you've just wasted a lot of effort at producing all the stuff you had to sell at a loss]. Rather, "loss" is the price of standing still, a luxury you can only afford if you are saving up resources for a critical moment. It costs money [and actions], after all, to innovate or to get preferential space on the sales floor. If you don't have money or actions to press ahead, you can hope to scrape by on what assets you already have, and the potential for accruing "loss" is the "debt" you take on for buying that extra time. If you choose to innovate, you'll probably be taking the more typical monetary loans that, as in London, are also available.
And so on.
These historically-themed, debt-driven, conflict- and confrontation-heavy economic Euro games are probably what Wallace is best known for [the keen reader will notice I left out Brass: Lancashire, which has its own cute little loan system reminiscent of Age of Steam]. He also has a penchant for historical conflicts of a non-economic nature, including Struggle of Empires, Byzantium, Liberté, and many others [his catalog consists of about a 50-50 split between economic and military history / empire-building games, if not slightly military-leaning]. His militaristic games don't have "debt" as such [certainly nothing much like any of the systems above]; they do, however, have concepts like "resource committal", "guns or butter", and "action economies".
In Bzyantium, for example, players simultaneously play a worker-placement cube-pusher [cubes are workers and resources] and an area-control world-domination game [where cubes serve as armies and movement / hit points]. The commitment of scarce resources [cubes, from a single supply per player] to either the action draft or the army development side of the game precludes and limits commitment to the other. The actions available for drafting are related to board control and economic efficiency; taking armies directly cannibalizes economic efficiency, as heavy military upkeep costs are imposed on the player that rise with the size of their army.
His military games also toy with the idea of playing multiple sides at once. In Byzantium, you play for both the Arabs and the Byzantine Emperor, and your score is essentially the sum of the two. Liberte has a system apparently much like König von Siam, where you progressively put political "stock" into the faction you hope will win, and then try to maneuver so as to achieve that victory for your favorite. Perikles has a funky area majority game driving control of either side in a number of different conflicts that players will resolve in a separate phase; and you can switch sides next round! And so on.
Games that share traits with Wallace's debt-driven / economic designs include Blockade Runner, Container, and Helden in der Unterwelt.
Games more like his historical titles include König von Siam [as named above], Tammany Hall [and not just the art!], and Wind River.
& Joris Wiersinga
Men who look too friendly to rip your heart, soul, and brain out, but do anyway,
Jeroen Doumen and Joris Wiersinga [aka Splotter] design games about:
Managing long-term supply-chains, or at least pretending to.
Around 1999, Splotter launched with an initial catalogue of about 6 or 7 games, championed by Bus and Roads & Boats. Together, these two games heavily influenced, or nearly created, the genre we now recognize as the resource management Euro game. Bus, with Keydom, is among the earliest games to make explicit use of the exclusive action-drafting mechanism we now know as "worker placement". Roads & Boats is perhaps the earliest example of a "resource conversion" Euro [i.e. cube-churner]. These are the two halves that together created Caylus [by conscious design choice on William Attia's part or not].
Later Splotter designs like Antiquity directly influenced the creation of the other huge worker placement resource game, Agricola, as described by Uwe Rosenberg in his "advent calendar" design diary. And the dynamic design duo of Jeroen and Joris hasn't stopped innovating and creating exciting games, with the brilliantly nasty Indonesia following Antiquity by a year, and the Merchant of Venus reboot Duck Dealer and satirical [and timely] Greed Incorporated following shortly after. The Splotter team have left behind the smaller titles that spotted their catalogue early in their history, and this has been a really good thing for the heavy game market.
Splotter games are about very long-term planning all mixed up with short-term efficiency. They are almost always entirely or very very nearly luckless, and are chock full of logistical planning nightmare puzzles to burn even the brainiest of brains. Splotter games are the Lexus of heavy Euro gaming: exceptional attention to quality in the design, and focused on "the relentless pursuit of perfection" in play. "Relentless" is, in fact, exactly the word to describe the demands are on your ability to plan multiple turns into the future. Making this task more difficult is the fact that even getting through one turn unscathed requires a lot of mental energy. Yet, if you're just living turn by turn, you're losing the game.
In Roads & Boats, the players are tasked with transforming 5 sticks, 3 donkeys, 2 geese, and 1 pile of rocks into as much valuable corporate stock as possible. Obviously, the path by which such a magical transformation may be accomplished must be a long and winding one, and that's where the roads and boats come in. By picking up the sticks and stones with your donkeys and moving them somewhere where you can build a quarry or a woodcutter, you can get a few more sticks and stones, then eventually build a mine to get some gold, then a mint to make some coins, then a stock exchange to write promissory documents against the gold coins... oh yeah, but you'll need some fuel along the way. Suffice to say, there are a lot of conversions that must take place. Trouble is, once you get past the first few "setting up" turns, you will have resources piling up and needing to be fed into every individual process in that long, winding production chain. At any given time, you might have raw lumber waiting to be made into building materials and gold coins waiting to be written into stock, plus everything in between. Ideally, every single turn would be an assembly line of simply moving progressively more inputs through a just-in-time production system so that money just comes pouring out the end; of course, this doesn't happen. What happens instead is that you marvel at your own inefficiencies and are lucky if you manage to scrape up two coins to rub together by the end of the game.
In Indonesia, the players acquire shipping companies and goods-producing companies that serve to distribute the staples of life [including microwave TV dinners] to the islands of the titular country. Challenge #1 is that only so many goods can be delivered per turn [each city only needs one of each good at first, unlike say Age of Steam whose otherwise similar goods-movement mechanism can service each city multiple times per round with the same type of good], so turn order and preferential access to routes is a key issue to manage in your long-term strategy. Challenge #2 is that companies may be merged; oh, by the way, "may be merged" means "by anyone", so that great shipping company you control could get merged with some lousy upstart, and a completely uninvolved third player could bid up and buy the ownership rights to the new larger, better, more profitable shipping conglomerate! So much for preferential access! The game builds step-wise as more and more cities have their demands fully satisfied, until eventually all of the cities on the map are accepting multiple shipments per turn and the players who have managed to position themselves and their asset holdings to take advantage of this economic development will find themselves rolling in revenue while players who planned less effectively will be left eating ramen noodles instead of the deliciously exotic sounding "siap faji" [those microwave meals]. From the outset, you have to make a plan to manage your liquidity and financial vulnerability.
In Antiquity, the players experience what it was really like to be a subsistence farmer in the Middle Ages; this game makes Agricola look like a cakewalk. There are steps each round for the assignment of famine, pollution, and graves. If you were lucky enough to have playtested the game, you are forever immortalized with your name on these gravestones! These and the pollution counters are the pivotal components / "resources" in the game, yet managing them is only a secondary concern. The primary concern is racing as fast as possible to one of 5 different victory conditions. It doesn't matter how much plague and pestilence you have going, if you manage to put one of these conditions together, you win. Of course, it is incredibly difficult to marshall the resources needed to do so without keeping death, disease, and decay at bay. The game itself is a combination of a territory-claiming, resource-gathering hex-map game like Settlers or any decent civ game and a special-power, city-building grid-filling game like Puerto Rico or any decent economic engine game. It is, essentially, an economic civilization 4x game writ [very] large. As in Roads & Boats, players start with almost nothing and must manage to grow that nothing as quickly as possible into increasingly more valuable assets; unlike R&B, there are not only player-centered inefficiencies to deal with, but game-driven obstacles to your growth. It is an intensely difficult and compelling game.
And so on.
Splotter games are not for the faint of heart. They are usually quite long, involved, and difficult to play well. However, the player is rewarded for their effort with games that stretch the limits of the Euro genre and push it in directions not yet taken, games that present new and interesting challenges worthy of analysis, and games that are [usually] brutally interactive.
Among modern designers, very few are as ambitious as Jeroen and Joris. They are designers as stouthearted as are the players they design for. For this reason, they don't have nearly the breadth of output as most of the other designers in this series [as I recall, only Tresham has fewer designs], but their depth is nearly without peer.
So, despite the price tag and wargame-like components, I think gamers would do well to pluck up their courage and give a good old-fashioned college try to these truly interesting creations. There's really very little out there that is at all comparable in epic scope and need for end-to-end analysis.
Games that share traits with the Splotter duo's designs include the Series: 18xx, Neuland, Kaivai, and the games of Phil Eklund [though quite obliquely].
- [+] Dice rolls
Hobby Game Designer Compendium, Part 1--The Great German Invasion: A Tale Of Two Stefans; or, A "Herr K."-ulean Face-off
30 Jan 2013
Folks on BGG often ask questions like "What is the best Reiner Knizia game for me to get if I like auction games but want something thematic?" or "Is Through The Ages representative of Vlaada Chvátil's style? Could I like Galaxy Trucker even if I hated TTA?" or "What is it that makes Stefan Feld's games so wildly popular among Euro gamers?"
There is an implication, probably accurate, behind all of these questions that these and other designers have something uniquely theirs that distinguishes their work in much the same way that content creators are distinguished in other media: architecture, art, comics, dance, music, writing, whathaveyou. It is usually something obvious yet nameless.
We as gamers lack the type of robust classification and criticism system that other artistic media have [though we're making progress toward it], so it is difficult to explain what makes, for example, Knizia and Sackson and Colovini and Schacht's games similar, let alone what differentiates them from the Kramer / Breese / Friese or the Georges / Feld / Dorn crowds.
I hope in this [two-part] post to catalog some of the more popular and stylistically consistent designers that I am familiar with and to try to give form to nebulous concepts such as "a Wallace game".
There will be two parts to each entry: 1) A 10-word or less description of the designer's style that highlights their most defining traits; 2) A longer narrative of the mechanisms they are known for.
I'll also try to reach outside of each individual designer's own catalog to identify some games that might be seen as influences on or extensions of their design style as I've identified and described it.
A man who takes gaming seriously, Vlaada Chvátil designs games about:
Procedure, priorities, and preparation with phased inflows of pertinent information.
Beyond the silly rulebooks and fanciful themes, Vlaada Chvátil is a mechanical mad scientist. Within his ouevre are deck-building games, real-time games, drawing games, acting games, cooperative games, educational games, civilization games, fantasy games, space games, dungeon games, and everything in between.
The common thread through nearly every one of his designs is an emphasis on progressively creating an infrastructure that can respond flexibly to leverage continually unfolding information and opportunities, and can eventually withstand a procedural "check" phase that awards or takes away points for the quality of preparation.
Criteria on which the player's infrastructure will be judged or opportunities which the player will be asked to take advantage of are nearly always occluded behind hidden or uncertain information. Usually, the player is given a way to "peek" at bits and pieces of this information, and possibly to control or modify it, usually over the course of many turns.
In Dungeon Lords, the players construct nefarious weblike dungeons and fill them with traps and monsters in hopes of knocking out the would-be do-gooders who will come galumphing through each player's dungeon in search of some manxome foe to slay. Different classes of adventurers require different preparation, and players will be assigned 3 adventures each over the span of 4 preparatory rounds before the "combat" check phase occurs. The pool of potential adventurer assignments is revealed iteratively turn-by-turn, as are the monsters and cavernous rooms that can be used to build up a Rube Goldberg Machine of Death to stop the good guys. Part of the preparation involves maneuvering so as to be assigned the adventurers your dungeon is best suited for, and largely the player who succeeds most at this maneuvering will win. After the unwelcome mats are laid out and the adventuring parties are knocking at the front door, a largely deterministic and highly procedural phase of "combat" ensues to see who can handle the outgrabeous onslaught and who will be snicker-snacked.
In Galaxy Trucker, the players assemble amassments of surplus space scrap in overoptimistic pursuit of something bearing vague semblance to a spaceship. Different bits of scrap allow players to perform different actions and use different resources, from lasers for shooting down baddies to cargo holds for transporting tribbles and other biohazards to shields for avoiding the inevitable meteor swarm [you'd think they'd just drive around, right?]. Oh yeah, they do this in real-time, too. After a mad-cap tile-laying ship-building sub-game, a series of cards are flipped over in sequence to determine the parameters of and obstacles encountered on the "flight". Those players who were not well-prepared will fare poorly; those who were better prepared will fare... well, less poorly at least. After the first few learning games in which players figure out how to perform the complex task of lining up matched sets of 0-, 1-, 2-, or 3-pipe connectors between their ship parts, they will have the opportunity to take a [real-time] peek at most of the "flight" cards before finishing their ship preparation.
In Space Alert, the players will
commiserate over the failure tocooperate in the attempt to protect a hopelessly doomed space craft with a hopelessly incompetent crew and hopelessly poor engineering from being blown to smithereens by dangerous and very dangerous alien hostiles. Unlike most cooperative games where planning and action happen simultaneously, this game is played out as a real-time 10-minute planning phase followed by a procedural "acting out" of what was planned. During the real-time blindfolded elephant description phaseplanning phase, alien threats and system breakdowns will gradually be revealed to the players by the most obnoxious voice-over CD ever recorded. As this new information is made known, players will use increasingly many expletivesrefine their plans so as to make it look like they know what the hell is going onminimize the amount of damage taken. They do this by pre-programming three entire rounds of RoboRally at the same time. A procedural action round then tells them how many years the Federation will banish them to Rura Penthe for.
And so on.
Through The Ages and Mage Knight have a similar structure, only the "check" phases are both more frequent and less severe. They take the form of civilization upkeep and production in TTA, and of smaller monster combat or native interaction rounds in Mage Knight. The focus in these bigger games is on the information and opportunity inflows, which also come more frequently.
Vlaada has a few other stock mechanisms, too, with the most obvious being the stepped progression of two or three game phases, each involving more complicated actions or resources and more involved planning or "check" phases. Every game mentioned above features this. Dungeon Lords has two "years", with tougher adventurers in the second; Galaxy Trucker has three phases, with bigger ships to build in each; Through The Ages puts the players through three, well, ages.
He also reuses a type of "player ranking" mechanism throughout nearly all of the above, where players move back and forth along one or more tracks representing the current status of their infrastructure along some axis of interest, and are occasionally judged or rewarded [outside of the main "check" phase] for their position. Dungeon Lords has the "Evil-o-Meter"; Galaxy Trucker has the racetrack; Through The Ages has the Culture and Military tracks; Mage Knight has Fame and Reputation. This is commonly the primary source of interaction in his games.
Decision-making in a Vlaada design typically involves prioritizing between the actions and resources needed to address each part of the "check" phase, as informed by the progressive revelation of what it will consist of, or simply prioritizing between flexibility / adaptability and amassing enough infrastructure in a single area to reap a large payoff if you get the opportunity. There are always at least half a dozen concerns to balance in decision-making, with different payoffs for each.
In spite of the multifaceted complexity of his designs, they do not fall prey to the "18 kinds of resource chits to transform back and forth" or "there are 5 resources which are mostly identical except for color" trap that plagues big Euro designs like Le Havre, Puerto Rico, Goa, or Macao. Vlaada's games typically have only 2 or 3 primary resources that do wildly different things.
Games that share traits with Vlaada's designs include 20th Century, Rise of Empires, 7 Wonders, and Polis: Fight for the Hegemony.
A man with a game about donkey poop, Rüdiger Dorn also designs games about:
Ducks, rows, trails, breadcrumbs, influence peddling, interplayer meddling... and poop.
Rüdiger Dorn is a "classic" German game designer. His themes involve trading in the Mediterranean, building castles or palaces, wielding influence in political courts, and doing things with dragons other than slaying them. His games involve trading, auctions, tile-laying, set-collection, technological development, and action / resource management. To quote a recent review: "If that doesn't get your blood pumping, you're. . . well, every gamer I've ever sat at a table with."
It's true, Dorn's game are not glitzy. They are, however, part of the pre-Rosenberg style of Euro game, built on positive player interaction, mutual gains and goals, shared spaces, and dynamic incentives. One of the defining features of a Dorn game is that players gradually unveil and declare their intentions or interpretations of game state to each other by relatively small actions which join together to form a larger picture. Sequencing of small actions and relative positioning in shared spaces are the order of the day. Often, this dynamic involves laying down claim markers in that shared space in physical patterns that define certain actions or effects to be activated or offered, in what is sometimes referred to as Dorn's "leave a trail" mechanism.
At times, there is only one trail which belongs simultaneously to all players [Goa, Genoa]. At other times, there are multiple trails which are available to all players [Titania, Arkadia]. At yet other times, each player has their own "trail" [Louis XIV, Robber Knights, Il Vecchio]. These "trails" often grow one space at a time into a series of adjacent cells, which is reminiscent of Kalah; Dorn has, of course, a few games with variously direct nods to that classic [Emerald, Space Walk, Der Schatz der Erdgeister]. Whatever the incarnation, the mechanism typically involves ensuring that your strategic direction is aligned with the likely path the game will take, and has all the components in place so as to take advantage of any unexpected twists or detours from that path. Additionally, you must take care to maneuver yourself around the intended directions that your opponents gradually reveal to you.
In Genoa, one player per round controls the "tower", a stack of 5 action discs which leave a trail across the board, potentially activating any of a wide number of different action and resource collection spaces in the form of buildings in the port city of Genoa. The catch is that while there are 5 discs, each player is only allowed to take a single action per turn [and usually 3-4 actions, or even 5, will be available based on the direction the tower takes]. Every time the tower steps into a new space [leaving a disc behind as part of the growing breadcrumb trail], the tower player may negotiate with all the other players for the right to take the action; this usually ends with another player paying the tower player some amount of money and goods and getting the right to use that space as their one action for the turn. The tower player can also negotiate for the tower's next move [this is, in fact, the more common approach; there are rules that harm the tower player if she just moves the tower willy-nilly without taking offers first]. Of course, the tower player can just take her single action for the turn and then stop moving, but then she misses out on all the potential bribe money.
The goal of the game is essentially set-collection, either of resource cubes [in 8 varieties] to fulfill "orders" cards or of adjacent ownership of "privilege" cards associated with each building on the board. These cards and resources are picked up from various spaces on the board [along with a few other layers of possible things to collect and do], and the most valuable ones require you to collect resources from 3 corners of the board and deposit them in the 4th. This obviously cannot be done in a single turn, so the key to doing well in the game is to manage multiple set collection goals simultaneously and flexibly negotiate for something of value on every turn, wherever the tower ends up trailing off to. While you are in control of the tower, it is often to your advantage to choose a path that is less directly beneficial to yourself so as to move toward areas of the board that you think your opponents will pay you a premium for access to. Of course, if you can manage to align your preferred direction of movement with your opponents' preferences, you'll do even better. It's also possible for players to lay claim to buildings along the trail and earn commission fees for their use later.
In Goa, as in Genoa, a grid of available actions or resources is displayed in the center of the table, and players attempt to gain access to their preferred items from that offering. The mechanism here, however, is auction-based rather than negotiation-driven; additionally, the resources / actions on the board are removed from play [into a player's personal area] each round instead of remaining to be used on a later turn. The selection mechanism consists of each player taking it in turn to place a auction marker [of 5 total--surprise, surprise--in the full 4p game] adjacent to the marker the player before them placed, leaving a trail of to-be-auctioned items across the board. These are then auctioned off in sequence, with the player who placed the auction marker on each item serving as auctioneer [and receiving the bid money if they do not personally win the auction]. There is a great deal of gamesmanship in [as starting player] choosing where to begin the auction and [as any other player] where to continue the auction trail so as to ensure that the items you are most interested in come up for bid in the order you want them to. As items get removed from the grid, this element becomes even more important, since choices for adjacent items to claim for auction become more and more limited.
In Diamonds Club, players buy remarkably silly and ostentatious things to put in their equally silly and ostentatious palace gardens. Additionally, there's a fun little twist on the action-availability-altering nature of the "leave a trail" mechanism as it's found in Genoa or Goa. There is a public grid of available resources, as in either of the two other games, but players do not "trail" across them in making their selections; nevertheless, the selections depend on adjacency, except the mechanism disfavors selections that are adjacent to each other [maybe "playing hopscotch" instead of "leaving a trail" is more appropos]. To claim an item from the grid on your turn, you put a coin on it; the catch is that you also must put an additional coin on the item for every coin in a square adjacent to it [i.e. on every previously claimed item]. The progression of claims is kind of like watching a losing Bingo card fill up: Spaces get blotted out that are completely separated from each other [often in checkerboard fashion], and then only near the end of the round do players decide they need to pay a bit more to make some claims adjacent to previous selections. The remainder of the game is a bit of a mix of Genoa's set-collection and Goa's tech-track development [a mechanism I didn't discuss; it's similar in nature to Hansa Teutonica's system]. I find this part a bit blase, but the action selection is delicious.
In Titania, players direct three different lines of ships, which are owned by and available to all players, to various locations on the board which grant resources [or abilities to use up resources for points ] to the player who places in them or reaches them. These do variously uninteresting things, like giving you points or additional card draws [cards are used to limit the color of ships you may build on your turn] or resources which can build the "big point" items once the trail of ships reaches special destination sites on the board. The greater interest in the game seems to lie in the maneuvering of these shared slime trails of ships such that they go in directions that you will be able to leverage with your current asset / resource holdings on your next turn, but that your opponents will not be able to so leverage. If things go completely awry, midway through the game the entire board resets [all the trails disappear] and players begin again, with a slightly different set of incentives due to prior placements / construction on the board. It's a classic-style Hans im Glück title that seems to have been released 10 years too late, as it fits in much better with their offerings from 1996-2000 than from 2006-2010.
And so on.
Even in his less "trail"-driven titles, there is a great emphasis on the timing and sequences of actions and resource collection / utilization. In Jambo, for instance, a player gets [surprise! 5!] actions on their turn, which must be split up across the actions of drawing cards, playing tech cards, using tech cards, buying resources, selling resources [i.e. getting points], and deploying one-time-use special power cards. Managing a "chain" of actions within and between your turns that efficiently leverages whatever the deck throws your way is the key to success in the game. Getting the chain out of sync or in the wrong order will see you wasting turns on drawing cards [of which you may only keep 1, regardless of how many you draw] looking for something to kickstart your progress.
I see Dorn, in general, as a sort of mezzanine designer voice in the "resource / action management" subgenre of Euro gaming, standing between directly interactive titles like Puerto Rico or Princes of Florence and today's more indirect titles like Agricola or Trajan. All of these games [like many of Dorn's] have player boards / player tableaus, and so all look like "multiplayer solitaire", but the systems that are built up around them have quite a diversity of depths of interaction. He is a few steps closer to Puerto Rico / Princes of Florence / etc, but in his funky action-selection mechanisms you can see inklings of Feld and the rest.
Rumor has it that he no longer considers his bigger designs necessary [halfway through the interview] in the spate of gamer-game explosions we see with every new Essen, but I think that his unique "classic German" style of design is still needed and welcome in the hobby. The fact that he had a SdJ nominee just this past year, and an as-yet well-received gamery-game shows that, whatever he says, he's not quite done yet.
Games that share traits with Dorn's designs include Hansa Teutonica, DruidenWalzer, Myrmes [seemingly], and Royal Palace.
A man of many faces, Stefan Dorra designs games about:
The Auction Grand Unification Theory in all its unified grandeur.
Though he has had at least 1 game published in all but 1 year  out of the past 20 years, many gamers probably don't even recognize his name, or if they do would be hard-pressed to identify any of his titles. His focus shifts regularly [from Euro games, to card games, to kids games, back to Euros] and he designs in genres [trick-taking, memory] that are not popular and for publishers [Hans im Gluck, Ravensburger] that don't export regularly. This combination has not helped his popularity.
Despite having two relatively successful recent titles with co-designer Ralf zur Linde [of Finca fame], Milestones and Pergamon, his most popular and highest-ranked game remains the 15 year old auction super-filler, For Sale. This is appropriate, since Dorra's designs are typically characterized by auction-like [though not often explicitly auction-based] mechanisms. In his designs, Applecline's Auction Grand Unification Theory finds full expression, even moreso than in those of Applecline's favored auction-heavy designer, Reiner Knizia.
Nearly all of Dorra's games involve players competing for the ownership of a pool of resources, which tend to be distributed either winner-takes-all or one-per-player [with order of claim an important consideration]. This is the 1st of Applecline's 5 elements of auctions [and also involves the 5th to a large extent]. Dorra also is fond of mechanisms wherein the value or the cost of some resource steadily rises until it is claimed or bought [the 4th, and possibly 3rd, element]. Winning a Dorra game consists of laying claim to the most valuable pool resources by expending the least of your own.
In Medina, the players jointly build up the Arabian city of Medina by filling it with palaces in 4 colors. Each player may claim 1 palace of each color during the course of the game, but before this claim is made, the single palace [allowed at a time] on the board of that color potentially belongs to anyone. On their turn, a player can extend any currently unclaimed palace that they like. This leads to a game of chicken that is a sort of reverse Dutch auction, where the value of the item [the palace] being offered steadily rises [as players extend the palace] until someone bites and lays claim to it. There are other bonuses, and secondary ways to extend even completed palaces, that muddy up the waters and make evaluation of the growing palaces quite difficult. By restricting each player to claiming only 1 of each color palace in the game, Dorra creates a very difficult timing and valuation exercise; the restriction also makes the game rather spiteful and dynamic, as players who have already claimed a palace of a given color will try in any way they can to limit the value of others of that color.
In Buccaneer, the players take the role of pirates intent on plundering a series of merchant ships. The only problem is that there isn't any clear leader to captain the crew! Players take it in turn to add a pirate to an existing crew controlled by an opponent, and so become the captain of the entire lot of loutish mutineers. Once a crew has become large enough, its captain can choose to send it out to plunder one of the available merchant ships. When this happens, the gold earned from the raid is distributed to every player who had a pirate in that crew, according to the value of the pirates [which have a range of values around 1 to 5], with the captain keeping any remaining amount. There are two auction-like mechanisms here, the "bidding" for control of the larger crews by adding to them and the intertwined "bidding" for part of the bounty that the crew eventually receives from the raid [whoever captained the venture]. The captain also gets an additional bonus treasure token that goes toward set collection. The effect would seem to be [I haven't played this one] a similar "chicken" style reverse auction as in Medina.
In Intrigue, the players engage in what might be the most direct and barebones negotiation game ever. The items to be negotiated over are salaries paid out to the owning player every turn. The means of negotiation is outright bribery to the employer whose estate pays the salaries. When a player is alone in seeking a job, the negotiation is simply for the amount of the salary [there are 5 possible values]. When 2 or more players are seeking the same employment, the negotiation is for both the job itself and the salary. The trouble is that all of the bribe money is given before the job is assigned and that the employer is under no obligation to give the job to the person who bribed him the most, nor to give the salary promised when the bribe was accepted. This breaks one of the cardinal rules of Applecline's understanding of auction [#5, that the highest-bid always wins], but it remains an auction game in its essentials, only one based on trust rather than evaluation. There is also a variant from the author wherein the loser gets their bribe back and promises are binding; with or without it, the resource deployment here seems to me to exemplify Dorra's style perfectly.
And so on.
Dorra is very fond of the Simultaneous Action Selection mechanism, as well as game structures that reveal [fully or partially / limited by memory] the assets of the other players that they will use in the "bidding" parts of the game. This combination lends his games a distinct "game theory" like flavor, where you mostly know the other players' incentives and how they interact with your own / how you can influence them, and have to guess what they will do in order to succeed.
In Streetcar, for instance, the players' tile holdings are made public, but their connection goals are hidden and must be deduced. In Sluff Off! and Nyet!, the players' bids reveal considerably more about the strength of their hand [particularly in what suits] than a typical trick-taking bidding phase does. In Turn the Tide, the players' hands of cards are not redealt after each round of play, but simply passed to the next player in order.
He seems to go wherever the wind blows, focusing most recently on mid-weight Euros like Pergamon and Milestones. I'd like to see him get back to his trick-taking and funky-auction roots; his designs in these areas are among the best offerings in those genres. I seem to have been a small part of a spark of interest that seems to have recently been lit under the long-languishing MarraCash, and hope that continues.
Games that share traits with Dorra's designs include Basari, Doge, GOPS / What the Heck?, Metropolys, and Nicht die Bohne!.
A man neither wizard nor warrior princess, Stefan Feld designs games about:
Efficient management of actions and infrastructure in pursuing competing demands.
Feld's games are nearly always multifaceted, full of many interlocking mechanisms and subsystems, but the thing that binds them is an emphasis on action diversity and action efficiency within the framework of a system with either very clear "limits to growth" behavior or a system of "demand" thresholds placed on the player by the game or their opponents [in the latter case, the mechanism chosen is typically majority scoring].
While his games are often economic snowballs of leveraging increasingly large infrastructure, they are not economic avalanches that simply build and build without limit. It is in the novel limiting factors he deploys that players typically find the most of interest in Feld's games. Where a typical snowball game like Through the Ages limits players through increased overhead / variable costs, Feld's games do so through difficult to control action activation systems.
There are usually blatant tradeoffs presented between specialization and diversity in a Feld game. The value of specialization is Feld's fondness for triangular scoring / income systems, majority bonuses, and infrastructure development. The value of diversity is the flexibility to take advantage of nearly any outcome in a system of action-taking that is often not kind to players. Striking a balance of specialization and diversity, and catching a stroke of luck, are keys to winning a Feld game.
In Macao, the players take the role of Portuguese merchants arranging trade routes with the far east. The game is a relatively straightforward cube churner in large part: Players will collect cubes in various colors, use them to build up a personal infrastructure that can turn the rest of them into money or points, or use them to pursue various other secondary set collection goals. At the beginning of the game, the 6 different cube colors are largely identical and interchangeable. A player may choose any direction to develop their infrastructure and have it be as successful as any other. The problem is that the availability of cubes is not identical across the 6 colors, but determined randomly by the memorable windrose mechanism. The infrastructure / buildings are designed such that specializing in a single color is generally the best way to get points if the timing and availability of those cubes works to your favor; of course, the wind[rose] of fate is fickle and this won't happen, so you have to find a way to create an infrastructure that can use a wide and unpredictable variety of resources effectively.
In Notre Dame, the players take the role of Parisian patrons contributing to the wealth of the city and the raising of the eponymous cathedral. There are 3 essential resources: cubes, coins, and victory points. [Sounding familiarish? Yes, this is another relatively straightforward cube churner.] Aside from earning VPs directly, the game gives players various actions that can turn money or cubes into VPs [yes, kind of exactly like Macao does] and such. The game has built-in incentives toward specialization in any of the 3 resource areas, in that the more resources you devote to particular actions in the game the more productive they become in their ability to grant you their rewards [an action that lets you get more cubes, for instance, gives you 3 instead of 1 if you built up its capacity; same for the VP-granting action, and the money-granting action]. The catch is that the activation of your action infrastructure is determined by a card draft, and your opponents are very unlikely to pass you what you need to activate your specialty. There is also a limiting variable cost [rats] that must be managed, as well as a bit of a pseudo-majority system [the cathedral].
In Trajan, the players navigate a complicated system of action and bonus action activation fueled by a Kalah-based system. Every action [of six possible] taken really demands specialization to function to its fullest potential, but the Mancala rondel is not setup to allow players to easily specialize. A large part of the game is in increasing the effectiveness of the actions you choose to take less frequently [by assigning bonus scoring opportunities to them, or by using them to accrue bonus action opportunities]. Beyond the sheer planning difficulties of working the Mancala system, there are set collection demands to be met [on penalty of VP loss] each turn that require players to divert attention from their most direct path through the system. There are majority bonus mechanisms as well. The game has a sort of "cobbled together" feel, in that every action drives its own little "mini-game" of sorts; balancing the demands and scoring opportunities of all of these microcosms is the chief challenge of the game. More than in most of his other games, Feld's love of competing, mutually limiting demands on your efficiency is displayed quite prominently here in what seems to be his magnum opus.
And on. "So" doesn't show up 'til two rounds later.
Feld likes penalties almost as much as he likes triangular scoring efficiencies. In many of his games, there are losses to be avoided as well as gains to be had. The loss is usually triggered by a lapse in efficiency beneath a certain "bare minimum" threshold. Pseudo-paradoxically, it's not always easy to meet the "bare minimum" if you're looking to be really really efficient in the use of any one particular action or scoring opportunity.
Nowhere is this really felt more than in In the Year of the Dragon, one of the more brutal of his games. The eponymous "year" in the game consists of a string of event tiles laid out before it begins, each of which specifying some condition that must be met to gain points or avoid penalties [usually the latter]. Players have to carefully choose which "checks" to intentionally fail as it is nearly impossible to complete them all and stand a chance of winning.
The challenge, then, is to manage waste [keeping it low] while maximizing output [efficiency]. "Waste management" might be a relatively good two-word description of Feld's style; a single wasted action or wasted resource usually carries with it a broad array of consequences [both foreseeable and unpredictable]. Keeping these wasteful outcomes to a minimum while continually building up a larger infrastructure is one of the more interesting features of his designs.
Games that [allegedly; I don't know this genre well] share traits with Feld's designs include Troyes, Glen More, and Village.
A man whose hips don't lie, Reiner Knizia designs games about:
Asset management, diversification, market vs liquidity risk, and incentive structures.
One of the more polarizing designers in the hobby, Herr Knizia is also among the most prolific, with over 400 unique titles to his credit in the BGG catalog. He is a mathematician and his games are known [deservedly or not] for being dry, calculating affairs with awkward to explain scoring mechanisms, as attested to in a microbadge created by his fans:
The overarching design feature of Knizia's games is captured by a famous quote attributed to him: "When playing a game, the goal is to win; but it is the goal that is important, not the winning." [This quote, coincidentally, also has its own microbadge: ] More than those of any other designer, Knizia's games proceed from their goals, victory conditions, and [yes] scoring mechanisms.
Despite his reputation, very few of Knizia's games actually have complicated scoring mechanisms, however. What they usually do have are multiple competing scoring mechanisms or scoring opportunities. The types of goals Knizia likes to design toward are those that reward maximizing the gain of a multidimensional portfolio of assets by navigating and managing a structure of ever-changing risks.
Knizia's designs favor scoring mechanisms wherein VPs are very rarely secured immediately and the bulk of a player's final score is in constant flux and remains uncertain until very near the end of the game or round. This provides the risk management element he is known for; a large part of succeeding in a Knizia game is comprised of capitalizing on assets with uncertain or low liquidity.
A large part of gameplay in Knizia's designs is the competitors' bumping, grinding shift of the shared game space until favorable conditions are reached for a player to capitalize on their holdings. There is also a strong element of asset portfolio management as players seek to obtain new assets that best fit their perception of the prevailing "market" trends.
In Lost Cities, the players have a relatively simple goal: Play cards [1-10 in 5 suits] to exceed a sum total of 20 points by as much as possible in each suit played, while laying down as many scoring multipliers as possible for the suits with the highest totals. The problem is that the cards have to be played in increasing numeric order, and that all multiplier cards have to be played before any number cards are played... and of course players aren't commonly dealt cards in a helpful x2-x3-x4-1-2-3-4-5-6-7-8-9-10 sequence as they'd like. Playing the game well consists of two aspects of risk / asset management: 1) Correctly gauging your likelihood of success in a given suit so that you can lay down the bonus multipliers before you start playing in that suit; 2) Biding your time through discards and "throwaway" suits until you have sufficient information to make the determination needed in #1. Playing the game very well consists of doing this simultaneously across all 5 suits in the game based on the information your opponent is implicitly giving you by their plays.
In Ra, the players "bid" for various lots of tiles depicting various aspects of a stereotypical ancient Egyptian civilization: impressive monuments, Pharaonic dynasties, cultural developments, farms along the Nile, and so on. Every category of tile has its own unique way of scoring based on set collection goals: monuments score for matched sets and for diverse collections, Pharoahs score in majority-bonus style, cultural developments are primarily used to avoid penalties but can provide points for sufficiently diverse sets, farms don't score points unless you also obtain a "flood" tile, and so on. You keep a certain portion of your collected portfolio of tiles from round to round [3 total], and you are challenged to find ways to both capitalize on what you've already collected and build a larger portfolio at the same time. The chief difficulty is that most scores are a threshold to be met [gaining a certain number of monuments; obtaining majority; etc] or a binary on/off switch [getting a flood, cultural development, etc]. Success is not a continuum; you either liquidate your assets or you don't.
In Tigris & Euphrates, the players lay tiles in 4 colors to develop "kingdoms" in ancient Mesopotamia, usually gaining corresponding color cubes for so doing. The goal is the scoring mechanism Knizia is most known for: "Your score is the number of cubes collected in your least numerous color." The most basic move is to lay a tile into a kingdom and collect a matching cube if you have the matching "leader" in that kingdom; the problem is that you will never win the game this way, since you simply won't get enough tiles / plays over the course of the game to build up enough cubes in each color if you collect them one at a time. There are better ways to get cubes: by winning a "conflict" with another kingdom and gaining a cube for every tile it had in the relevant color, and by building "monuments" that provide free cubes every turn to various leaders in that kingdom. Every play changes your ability to take advantage of these high-income moves, and winning the game consists of continually aligning your hand of tiles to the shifting opportunities on the board. And, to do this in the context of diversifying across all 4 colors.
In Amun-Re, the players bid for control of various regions depicting various aspects of a stereotypical ancient Egyptian civilization: impressive monuments, Pharaonic dynasties, cultural developments, farms along the Nile, and--wait a minute... didn't we just do this? Yep; like Ra, this is a game of portfolio development along multiple axes with the eventual goal of the most effective liquidation possible. A player's assets consist of 3 regions per scoring round, with attached farms, pyramids, and temples. A sealed bid at the end of each round determines the profitability of the farms and temples; the pyramids score 3 ways: on their own, in completed sets, and for majorities on either side of the Nile. Every bid for a region must be weighed against its potential contribution toward each of these 3 areas of competition, and the subsequent building phase is a further refinement of that potential. Guiding these decisions is a 3rd axis of refinement in the form of bonus scoring cards that require your portfolio to meet various criteria. These are scored all or nothing, and converting on them is key to victory.
In Stephenson's Rocket, the players invest in the development of the British railway system in 3 distinct ways: by claiming goods that will be delivered from each city the new railways connect to, by claiming stations along the route that serve as switchovers between towns, and by claiming stocks in the railroads they hope will be most successful. Each of these scores in a separate majority-bonus subgame, but the 3 types of assets [goods, stations, shares] are interdependent, since the scoring of all 3 depends on how and where the tracks are connected. The game has a palpable sense of emergent cooperation between players, and the most important part of decision-making in the game is determining whose levers to pull and buttons to push and in what manner to do so. All 3 types of assets are particularly illiquid, and converting them into VP once claimed is a difficult process for which you need the help of other players. Of course, this help comes unwittingly; you can't simply tell your opponents what to do. There is, however, a "veto" phase in which you can tell them what not to do.
And so on, ad infinitum.
Of course, the good doctor has a plethora of other design proclivities. A GeekBuddy of mine, Laszlo Molnar, has cataloged a number of the more common elements of Knizia's designs in a number of Geeklists. Topics include: Double Tile-Laying, "Area Surrounding", Riffs on Acquire, "Linear Adventures", Polyominoes, Maze-/Grid-Building, and of course the "Highest Lowest Score" concept.
My personal favorite Knizian quirk [though it is not uniquely his, of course] is the two-action turn. Action-Point Allowance games are great, of course, in that they create very dynamic turns, but they require balancing of action costs and can often lead to analysis paralysis. Knizia gets around that while maintaining the rapid pace of game state change by allowing only two actions.
The two-action trick shows up in various incarnations in Through the Desert, Tigris & Euphrates, Stephenson's Rocket, Samurai, The Lord of the Rings, Jäger und Sammler, Genesis, and probably others that I am not familiar with. Relatedly, perhaps, Knizia is also fond of the two-choice turn [in various forms], from Ra's "Draw or Ra!" to Loco's "Play a card and Draw a chip".
Games that share traits with Knizia's designs include MarraCash, Yspahan, Medina, Key Harvest, and [I think] Mahjong.
Ein Mann über alles, Wolfgang Kramer designs games about:
Multiscalar tactical decision-making with elements of persistence and repercussion.
Ironically, Herr Kramer himself is also repercussive. He has been an active hobby designer for nearly 40 years [his first credited design in BGG's database is Legemax from 1974], longer than even such industry stalwarts as Alan R. Moon or Richard H. Berg, and nearly rivaling the unsung Reinhold Wittig. Kramer is often credited [accurately or not] with creating the victory point track and the area majority and action point genres.
His magnum opus, El Grande, is the oldest non-traditional game [i.e. excepting Go, Tichu--a minor Zheng Fen variant--,and Crokinole] in BGG's Top 50 ranked games, a testament to the longevity and continued relevance of his work. This is particularly appropriate, as longevity and continued relevance are mechanical, as well as ludological, hallmarks of his game designs. Decisions in a Kramer game have impacts on multiple time scales simultaneously.
Kramer employs a lot of multiple scoring round structures and special scoring opportunity mechanisms, where the infrastructure that players build up persists throughout the game and is leveraged multiple times [and often in multiple ways] to earn points. This gives his games a rich tactical backdrop, since every move must be weighed against its immediate use, or at least its use in the most imminent scoring opportunity, as well at its infrastructure value for bigger opportunities to come.
In Hacienda, the players expand land claims and animal herds and are judged in two scoring rounds for the size and connectivity [markets reached, water surrounded] of their claims and herds. The multiscalar element arises in that the early game is about money and resource management, and players often take the most direct and efficient routes to market [which generates cash] so as to conserve. The easiest next step is simply to expand "close to home", continuing to deliver to the same market and develop the same land; this is not a very valuable action, points-wise, however. The game rewards expansion to multiple markets, which are costly to reach, as well as careful consideration of the shape [as well as size] of land claims and herds so that wells and reservoirs [which grant points] may be dug around them; often the most potentially valuable shapes are not the most efficient. There is also another special scoring opportunity, the eponymous haciendas, which may be bought for very large chains [which require costly specialization of resources]. The tactical balance is very rich.
In Colosseum, The Princes of Florence, and Master Builder [all closely related], the players navigate an auction round aimed at set collection followed by an action / scoring round in which the sets are judged, and then repeat the process until the end of the game. In all of these games, the large majority of a player's collected resources persist after having been scored, whereas the set collection goal to which they were applied is removed from play; this means that assets continue to turnover and be scored toward multiple set collection goals. In Colosseum, even the previously completed goals persist as scoring bonuses toward future scoring. In the early stages of these games, players will be forced either to pursue smaller goals or to settle for fulfilling fewer of the requirements of the goals that are available to them. Tactical decision-making in the game is a balance between getting the best resources for completing your immediate goal of interest and getting the resources that will most flexibly allow you to complete future goals; the player that wins the game will be able to make these two time-scales overlap and reinforce each other.
In the games of Series: The Mask Trilogy (Kramer / Kiesling), as well as the apocryphal member Torres, the players work [pseudo-]together to develop and/or explore a shared landscape and are rewarded multiple times during the game for control [through majority or dominance or other claim] of portions of that landscape. In Mexica and Java in particular, a sharp distinction is also drawn between game elements that provide points immediately, elements that provide points during mid-game scoring, and elements that provide points only in the final scoring. This "now, then, later" method of scoring is a hallmark of Kramer's designs, and is particularly strong here. In all of the four games in the "trilogy", the shared landscape persists throughout all time phases of the game and its development shapes the actions of players in each portion of the game, with a general trend of catch-as-catch-can in the early game where very little has been developed that shifts toward a positional struggle for the highest scoring claims for the end game. These games also prominently feature Kramer's proclivity toward action-point allowance games and each have a large menu of possible player actions to spend points on.
In Heimlich & Co. and City, two old roll-and-move family-style games, the same features are evident despite the simplicity of the games. In Heimlich, players are assigned secret identities from among a pool of 7 colored pawns. On their turn, they roll a die and make moves for any combination of pawns they like up to the number of movement points rolled. Spaces on the board consist of scoring values from -3 to +10. The object is to get the colored pawn matching your secret identity onto a higher scoring space when a scoring round is activated, but to do so in such a way as to not tip off your identity [because your opponents will simply see you favoring that color and move it to a lower scoring space; additionally, there are variants where you gain / lose points for guessing at secret identities]. The real object of the game is to figure out a way to keep your pawn on the higher scoring spaces for multiple scoring rounds at a time. There are numerous cute ways to do this: you can hold back on a 6 or 7 and be content that something else will be a better target for moving off to a low scoring space; you can choose to move the scoring activator [which sits on a space on the board and triggers scoring when landed on] to a space you think the next player will favor [while your pawn remains on the place it just scored for]; and so on.
In City, reimplemented as The Market of Alturien [my personal choice for most underrated game on the Geek, by the way], players buy spaces on a [mostly] circular track, build trading houses on those spaces, roll dice to move pawns around the board so as to land on spaces, and receive money when the pawns land on their spaces based on the number of houses they have [yes, this is Kramer's take on Monopoly]. Players control the direction in which pawns may move [they don't simply circle the track], and have the ability to purchase or develop properties anywhere [unlike Monopoly, players have no on-board avatar; the pawns are all neutral buyers that anyone may move]. This leads to players trading off between prime real estate [intersections, mostly, but also competition for area majority bonuses] and areas that are likely to be hit soon by the moving pawns; these long-term vs short-term tensions are typical of Kramer's style. Additionally, there is an added incentive to prioritize the long-term development, since every intersection has a special rule for buyers; in general, only the buyer pawn that was actually moved on any given turn will pay out to the business it stops on, but if any pawns remain on any of your intersection spaces from having landed there in a prior turn, they also pay out. This drives some fun "take one for the team" play in moving pawns off of these repercussive payouts.
And so on, for decades.
Kramer is also well known for being a collaborative designer. Only between 20% to 25% of his top 50 or so games are sole-author creations. His primary co-design partner is Michael Kiesling, but he has worked with at least a dozen other authors [the next most frequent being Richard Ulrich]. I suspect the built-in feedback loop / sounding board of the co-design relationship is largely the reason why Kramer's games always feel well-conceived, well-playtested, and well-balanced.
Kramer is kind of an unsung hero [at least in the English-speaking circles I game and/or chat with; maybe it's different in mainland Europe / Germany?] "hiding in plain sight", as far as I can tell. He's at least as good and as influential as Knizia, yet judging by "fandom" on his designer page and ownership of his microbadges, he's only 1/3rd as popular. He's also dwarfed by relatively "young" upstarts like Feld and Chvatil, being about 1/2 as popular as either. It's hard to fathom.
More than most designers, Kramer is consistent in the quality and style of the games he has produced over the years. He no longer dominates the BGG Top-100, and his "big" newer games [The Palaces of Carrara, Artus, and Seeland] are certainly not flying up the charts, but unlike, say, Reiner Knizia, I think this is an artifact of the average BGGer's tastes changing and not the result of Kramer's design quality changing. He was and remains a designer worth watching.
Games that share traits with Kramer's designs include Alhambra, Vikings, Factory Fun, and Vegas Showdown.
- [+] Dice rolls
I have an apology to what is left of my subscriber base:
I have been incredibly slow of late in putting up posts.
I have two or three posts actually in progress, but they languish for want of impulse, creativity, or motivation. Take your pick, as they probably all apply.
I think I can make it through what I intend to write about here in one setting, and I've got nothing better to do [up at night, can't sleep, contemplative, etc].
A Force to be Reckoned With
If I were to identify one element or trait that describes what I enjoy in games, which would be enough to provide sufficient information for a reasonable summary to a fellow gamer, it would be dynamism.
By this, I do not mean simply that the games I enjoy are exciting or interactive in play, but rather that they possess a quality that is related to the [a?] rigorous definition of "dynamics" as a field of study.
Dynamics is the study of forces [from Gr. "dyna-" = "power", "force", etc]. More particularly, it is the study of the effects and interactions of forces. That is, to motion and change as a system or a process.
The processes of motion and change in general are typically present in any game, however, so what do I mean in particular by the application of this term to game design as a differentiating factor?
1) Dynamic games have changing structures of costs and incentives. Such changing structures are state-dependent rather than sequential or stochastic. They both depend and impinge primarily on player choice.
2) Dynamic games have interdependencies between players and the systems they inhabit in the game. The forces which act on players in the game come from multiple sources and have multiple interlocking effects.
This is not a direct representation of these concepts, but gets across the same gestalt in short order. Dynamic games depend on themselves [that is, on their individual instantiations in a play session] and not on their rules.
The essence of this self-referential graph [the world's first "Quined" graph?] is reproducible in any number of configurations, and each minor shift toward a new one would necessitate sweeping changes to the whole.
The "rules" of the graph affect how it must be "played", but the shape and result of the graph depends only on some initial seeds [how thick the lines are, what the dimensions are of the shapes, etc] and their dependencies.
The flow of a dynamic game is the same as this "dynamic" graph. The rules govern how the game is to be played, but the values and incentives that are the game as such grow out of the rules rather than are set by them.
Seeking Change for a Change
What in the world is a "[state-dependent] changing structure of costs and incentives"? Perhaps there are more compelling instantiations than the following, but essentially all that I have in mind is the situation in which numbers or values in the game are not hard-coded into the rules but are dependent on some portion of the game state.
- In Acquire, the price to be paid [buying or selling] for a stock is neither preset at a certain level, nor increased in predictable sequential fashion over time, nor adjusted randomly* by some "random walk" process but changes solely based on player actions and is determined by some dynamic metric: the size of the associated hotel chain.
[* the probability distribution governing this, and thus the expected and possible values at any point, would still be part of "the rules"]
- In Web of Power, the value of a[n area majority] scoring region on the board is not preset to a certain amount, nor changed by fiat of a player, but depends on the number of houses each player manages to get into the region. Similarly, the incentives and costs of the ambassador scoring opportunities are not preset values, but depend on house placements.
- In Medieval Merchant, the income gained by players from their cities on each turn is not a fixed amount, nor even an amount based on a mathematical progression or payout schedule [a la Power Grid's bureaucracy phase], but dependent on both a player's choice to build or tax and on the level of competition in each city in question when the choice is made.
- In Modern Art, the value of the paintings being auctioned is not inherent in the items themselves, nor a probability function around an average payout when the cashing-out-your-chips phase comes, but is determined instead by how many other paintings by the same artist were sold during that round of auctions as well as in previous rounds.
- In Hacienda [an underestimated light-/mid-weight rail game], the money earned "bringing a herd to market" [connecting to a city] is not dependent on preset values for the city or for connections in general, but on the size of the herd and its farmland at the time a connection is made; money earned at "harvest" depends on the size of the farm.
This is not a review-oriented post, so I hope these brief snippets will suffice as examples. There are many more, of course, that could be given.
Essentially, any game with a preponderance of the words "per" or "for each" or "based on", or with emphasis on quantities such as "the most" or "the highest" or "the number of", probably fits at least partially into this broad categorization. This doesn't seem a terribly thrilling or uncommon concept at all. Certainly not worth fussing over. Surely every game has something like this?
Or does it?
Counter-examples abound in mass-market titles. The values and payments made in Life or Monopoly, for instance, are nearly all hard-coded into the board and the rules. Relatively low-dynamism games are found in the hobby, too. Settlers of Catan is impressively devoid of it [aside from the "longest road" battle], yet the preset values and board are necessary to support trading and expansion.
Is it a bad thing for a game not to be "dynamic" in this way? Are the types of games I list inherently better or deeper just because of this feature? Well, no... and that's not the point of this post. It's just a preference of mine, and something that I find interesting for its own sake. It also strikes me as one of the more modern design features in our entire hobby; it is like a cooperation between player and designer.
To a large extent, this kind of design allows and sometimes forces players to set their own goals in the game and find their own "paths to victory" within the game space. Because values drive player choice in a game, allowing the values to vary based on player choice means that players can have a hand in shaping their own destiny. Their choices now will affect the substance and content of the choices they face later.
That's compelling to me, and it's more than just a typical engine-building game of piecing together the requisite parts of a combo-driven victory-point generator. Getting a building that generates 5 stone a turn and next turn one that turns 3 stone into 10 VP doesn't change the relative cost structure you face, at least not in the dynamic self-referential way I have been discussing. It's just a choice within the over-arching incentives.
It's a fine line, perhaps, between that example and a "dynamic" game as above. It might be as simple as there being many different stone -> VP buildings available, so that each player has to choose their own rate of exchange. It might be that players could get X VP per stone where X is defined as the number of Y buildings the player has access to. These seem insignificant modifications, but I think they increase dynamism.
And I like dynamism.
I like it so much that I instinctively look for it in games and seek to add it via houserule or variant if it doesn't seem to be prevalent enough. A not-too-unrecent example was during playtesting / communal development and criticism of an Ace variant for the Decktet resource-management / tableau-building game of Magnate. The initial suggestion was that an Ace would count as a value equal to the number of other cards you had in the column that shared its suit. Yay! Dynamic value!
What I took umbrage with was the comparatively more boring cost of playing the Ace: 3 resource chips. I suggested that it, too, be dynamic, based [like the eventual value] on the number of matching cards in the column where you want to play it. The designer shot back that he didn't see either the need for the added complexity or the value to the game of having the cost dynamic. I eventually conceded both points, begrudgingly, but I've come to appreciate, I think, the wisdom of his perspective.
There can be a problem with too much dynamism and too many floating values, and not just in terms of rules complexity. The problem is one of calculability and choice. To calculate anything accurately in the context of mobile objects [or values], you need a fixed vantage point for reference [or at least a point fixed relative to the moving body]. In game terms, if all of your costs and all of your values are dynamic, you either have to set a maximization function on the distance [if that's the appropriate relationship] or fix one end for reference.
It can be difficult to accurately determine, when either the low or the high point of a range is variable and both are independently difficult / costly to move, whether it is more effective to decrease the low end or increase the high end [assuming "maximum separation" is the goal]. In practice, you will probably consider one end fixed in the short-term and work on moving the other, then consider it fixed at a certain point and shift toward the original fixed end. This is clunky and unnecessary, unless the exact timing of that shift is something of interest for the design.
Much better is a mixed-model.
If you preset the value of one end of the range, you can maintain the difficulty and interest of controlling a dynamically-determined value on the other end while simultaneously maintaining the ability of players to make sane judgments among alternative courses of action. I have only chosen the "profit margin" between value and cost as an example of convenience, but I think the principal of a fixed reference point applies to nearly any "dynamic" game.
That reference point can be something as trivial as a hard end-game value for leftover resources, for example. "Each $5 is worth 1 VP at game end" or "Each 3 cubes of the same color leftover are worth 1 VP at game end" or some such. Even if the game is such that players never take advantage of this scoring, it is incredibly helpful as a standard for measuring other sources of scoring potential against. It allows for true and accurate opportunity cost analysis.
For instance, if you have the opportunity to spend $2 to convert 3 Stone to 3 VP or $4 to convert 2 Wood to 3 VP, and both of the above end-game conditions apply, you can use the relative distance moved from those two fixed reference points at end-game to compute the respective values [at least short-term] of these two moves. For Stone, you lose 0.4 VP in $ and 1 VP in cubes, for a gain of 1.6 VP; for Wood, you lose 0.8 VP in $ and 0.67 VP in cubes, for a gain of 1.53 VP.
Even if the $ cost of getting Stone or Wood is subject to change and if different scoring opportunities might present themselves next round, this is a cold-hard mathematical fact about the relative value of these options now. If you did not have the base-line for comparison, however, you'd be forced to guess at future game states to determine whether it was better to have $2 more or 3 stone more. You may still do that, of course, but now you have a better context for it.
We're All in This Together
Much more interesting still is the "second-order" dynamism wherein players directly, or mediated through some game system, impact the changing incentive structures faced by their opponents. Most [all?] of the games I cited as explicit examples above fall into this category, but it's not a necessity that a "first-order" dynamic game have this feature. It is, however, for folks like me who enjoy this sort of thing a bit of a holy grail feature. We are like moths to so many candles.
There are two things I can do to your incentive structure in a good second-order dynamic game [and the strength of such a game seems to me to subsist in how interesting and difficult they are to do]:
1) I can make choices that overtly and directly change your cost and/or value calculi. For instance, I could urbanize a town right out from under your most profitable route in Age of Steam [making it a lousy 2-stop route] which might lead you to build track elsewhere, deliver different cubes in the scoring round, and so on. Or, I might plop down a green/blue monument in Tigris & Euphrates using green tiles in a kingdom where I have a blue leader, hoping you'll then want to knock the green dude out.
2) I can make choices that shape the environmental context in which the players operate during the game, leading to intermediate or long-term changes to your incentives. For instance, I may begin building up a military powerhouse in Through The Ages, causing you to reconsider the peaceful strategy you had planned and maybe add a bit more colonization to the mix as a backup. Or, I could purchase an out-of-the-way shop in the streets of MarraCash, which might make you consider what would be worth owning along the way.
If you've ever wondered what the memorable "torque the incentive grid" comment is all about, I don't feel massively out of place interpreting clearclaw's commentary for him as related to exactly these types of activities.
These types of games require not only the type of opportunity and incentive analysis discussed above, and the ability to manipulate your own dynamic value axes to your benefit, but in addition the bewilderingly difficult skill of being able to interpret and predict the mental analyses being performed by every other player at the table along the same axes so that you can interject your malevolent will into their finely honed machine such that their value structure disintegrates. Games like this may be light-hearted, but are not for the faint-hearted.
Note that these games are not always explicitly mean-spirited. In fact, there are many that are quite "friendly" and constructive in their own way. Marracash that I mentioned above is a wonderful example. The game is setup to provide you with incentives to help your opponents do well, and for them to do the same for you. There is not a whit of destruction or loss of accrued value [loss of potential value, sure] in the game. Rather, the game consists of trying to get other players to do as much for you as possible while offering to do as little as possible for them.
Note also that even though the words "offer", "help", "threaten", etc are bandied about in discussion of such games that relatively few of them allow explicit negotiation. What is meant, instead, is that the interlocking of players' tactical decisions is so great and the interdependency of players' strategic positions so high that a well-executed move by a given player implicitly "invites" a particular response on the part of their opponents. The invitation might be "Let me tag along! I'm only getting half the points you are!" or other seemingly innocuous interactions implied by positions taken.
The skill in such games is largely in setting up these "invitations" convincingly and interpreting the actual innocuousness of those presented to you. It is not quite as simple a mechanism as the meta-gaming "bluff" in your typical "let's take over the world together" American-style game, but rather something a little more subtle yet [importantly] laden with more information. A bluff, of necessity, entails that some important element of the offered invitation is hidden or concealed. In a game with this sort of dynamism, the offer is typically wide open for perusal in its entirety; it is only the implications of it that are uncertain.
Whether you are destructive or constructive, a quiet strategist or an active haranguer, I think you can find something in this general genre of games to excite you. I don't think you should write them off.
Whether you will fall in love with this type of game as I have depends on other factors, particularly your tolerance for incessant [if only implicitly conducted] judging of alternative schemes of action against each other with regard to their likely impact on the incentives that will guide your opponents' future actions. Additionally, because these games are played "above the table" [between the players, eyes up; not within the players, nose down], there is an oft-encountered side-effect that the games present a drab or "abstract" appearance. This is because the players themselves, rather than the story or the artwork or even the theme, provide the interest in the game.
While this isn't a review post, if you are interested in this sort of thing, I think most of the games I've highlighted above [Through The Ages excepted; it just provided a single useful example] would be wonderful introductions to this sort of thing. They are, of course, my own personal biases. If you're interested in other personal biases concerning this basic style of play, they really aren't that hard to find once you know what to look for. I don't name-drop other users often except to interact directly with content they've put up elsewhere [see above], but if you're clever you might find a few folks hidden in my profile, most of whom are there because they share my taste for this sort of thing.
Whatever you do with this post, more than anything I hope that next time you're staring at a little kingdom of black boxes on the player board in front of you that you think about how much more fun and dynamic that game could be.
- [+] Dice rolls
Very rarely do I make a blog post with the direct intention of one-upping another blogger, writer, or podcaster, but in this particular case the entire impetus behind this post is that I think I can do better.
The inspiration for this post was chiefly last month's episode of Ludology wherein Geoff and Ryan talk about randomness in gaming. [Another recent influence was Sam Mercer's "non-random die-roll" challenge.]
Perhaps because of the podcast format or perhaps because of time limits or breadth of scope*, the Ludology guys really swept over a lot of ground rather quickly and made a bit of a muddle of things in the process.
[* I believe this to be the case rather than misinformation. Geoff, at least, has a number of excellent "Game Tech" segments that display quite clearly that he is capable of concise and accurate analysis of randomness.]
Geoff and Ryan touched on an awful lot of topics that are central to a proper understanding of randomness, but didn't really cover any of them in much detail or introduce them in a way that I found worked very effectively.
So, in this post I intend to discuss what randomness is, why it matters and occurs in gaming, how we can interpret and analyze randomness, and when it is or is not an appropriate game mechanism for a game design.
Alea iacta est! Or, what is randomness?
"Random" is the term we use to characterize data-generating processes that produce indeterminate results whose relative frequency of occurrence is defined by a probability function [which may not be known or knowable].
Admittedly, that's a bit of a mouthful, so let's break down the individual elements.
Randomness occurs in and is primarily concerned with active processes. While passive events like receiving a "random" phone call from a friend, experiencing a "random" natural disaster, or encountering a "random" person on the street are typically characterized exactly as such, this is a different [more colloquial and imprecise] use of the word than that which we usually intend when we talk about randomness in a formal sense. Unless you are considering "waiting for a friend to call", "tracking natural disasters over time", or "observing people on the street" as an active experiment [in which you are mostly collecting non-data: nobody called, no disaster happened, no people were encountered], the untimely and unexpected event that happens "out of the blue" isn't random, only rarefied.
Furthermore, randomness is concerned with data, or information, about the outcome of said process. We might say "I got a phone call from Amy at 4:00 today"; "The hurricane had wind gusts of 100 mph and came aground on August 29th"; or "3 men and 5 women were walking down Main Street", and the thing that randomness would likely be concerned with [were these actual active processes] would be the numbers or things [gender, dates, etc] that could easily be represented as numbers. Randomness isn't [usually] concerned with the fact that you got a phone call or didn't ["It was random! It came out of nowhere!"], but with when the phone call happened, how strong the hurricane was, when it hit the epicenter of damage, or how many people of each gender were encountered.
The most common and easily understood example of a "data-generating process" is [fortunately for us!] a die roll.
The die is cast [alea iacta est]--the process--and then we read off the obtained result--some data is generated.
There are processes that generate data that are nevertheless not random. An easy example would be me asking you for your name.
Presumably, if you're not a class clown, you will respond in the same way regardless of how or when or by whom the question is asked.
Your answer--the data generated by the process--is deterministic. You only have one name, and you provide that information to whoever wants it. Yes, obviously there are slight variations for formality, nicknames, etc. Even allowing for these variations, you do not have a set of possible names to provide to askers at random. If you are in a formal situation, you will give your formal name. If you are introducing yourself to a new friend, you might use your nickname. You would never write down "1 = Alex", "2 = Bobby", "3 = Chris", "4 = Dana", "5 = Erin", "6 = Francis" and then roll a die when asked what your name was. No parent is that mean to their child! You wouldn't even do this with variations on your name ["1 = Nathanael", "2 = Nate", "3 = Nate Straight", etc]. Instead, you [the data-generator] choose the result.
Using the die roll example, a deterministic process would be to roll the die and add the number of pips on the top and bottom faces. Because of the way dice are designed, this will always give you a result of 7. Regardless of how the die rolls, where it lands, what number is on top, or any other input or environmental factors, the outcome of the data-generating process will always be the same. This is also something that randomness is not concerned with. Instead, we are only concerned with processes that produce results that are not deterministically generated by the input factors. Importantly, even if the method of determinism is unknown, the process is still non-random. Imagine you have six levers, a black box, and a light. One lever [only] turns on the light. Even if you don't know which one it is, it is not a random process.
This is because it is not the case [even though it's the most likely interpretation by a careless / casual observer] that each lever has a "1 in 6 chance" of turning on the light.
There is no "chance", in fact. One lever has a 100% "chance" of turning on the light. The others have a 0% "chance" of turning it on. The only uncertainty is our own ignorance.
At best, we might be able to say "If we choose a lever at random"--say by rolling a die and counting off--"we have a 1 in 6 chance of choosing the lever that will turn on the light."
This may seem like pedantry, more relevant in a philosophical discussion of free will and determinism, but it is an important observation on what it means for an event to be "random".
Relative Frequency of Occurrence
The introduction of percents and fractions in the prior example is a good segue to the next element of randomness, that of relative frequency [as opposed to an absolute count].
When we discuss random events, we nearly always do so in terms of the proportion of occurrences of an event to the total number of outcomes or repetitions of the process.
Imagine a gambling game where you win the bet if you roll a 2 at least five times in ten die rolls. It matters quite a bit whether the game consists of rolling one or two dice.
Whether you roll one or two dice, there's [absolutely] only one way to get a 2. But what you care about is the fraction of the total "ways" the dice can fall that are 2's.
In just a moment, we will begin enumerating sets of possible outcomes. Some poor intuition [based on absolute rather than relative counting] about how randomness works might lead someone to say that because "2" is in both the set of all possible outcomes for the roll of one die and the set of all possible outcomes for the roll of two dice that the chances of getting a 2 are the same in either scenario. This sounds silly, but it is a type of error that is actually made by real people in more subtle real-world scenarios involving randomness or uncertainty and prediction. More precisely, confusion of absolute vs. relative frequency is appallingly commonplace [absolutely and relatively]. It is so pervasive that psychologists have discovered [pp. 4-7] it to be an innate inability in children to distinguish absolute and relative quantities in interpreting randomness.
Adults make similar mistakes, though. If presented a problem something like "Is a cure that saves the lives of 9,000 out of 100,000 infected people more effective than a cure that saves the lives of 90 out of 100 people?", people will incorrectly answer that the cure with the higher absolute effect must also be the cure most likely to help the relevant population. The confusion is also present in the use of phrases like "5% increase" to describe something already measured in percentage points and something measured in absolute terms. "Inflation increased by 5%" [YIKES!] might be written by someone who really means "prices increased by 5%" [who cares?]. Even more insidiously, people will misinterpret the fact that "there are a lot of black people in prison" as an indication that there is a high probability that a black person in society is or will become a criminal.
Probability Functions [Distributions]
What we need, then, is a way to talk about the relative frequency of the indeterminate results of a random data-generating process. This is what probability is for. Generally speaking, the probability of an event is the ratio of the number of times it happens to the total number of times we have looked to see whether it happened [the frequentist / empirical definition, if you care]. We might also conceive of it as the ratio of the total number of ways an event can happen to the total number of possible outcomes of the process that [sometimes] generates the event [the classical / theoretical definition, by the way]. We used both definitions previously in discussing the die roll game, and both are relatively [heh] intuitive. Under either definition, careful observation [for the frequentist] or counting [for the theoretician] will very often derive different probabilities for different events.
The mathematical equation or graphical representation that defines or describes the different probabilities associated with every individual possible outcome of a data-generating process is known as the probability function [or probability distribution] of that process. Sometimes the function is very simple [for a single die, it might be written as "Pr(x) = 1/6", read "the probability of event x is 1/6" where "x" can stand for any number in the set of possible outcomes], and sometimes it is more complicated. Often it can not be written in a single mathematical equation and is either presented graphically or as a set of equations. Another possible presentation of the relatively simple probabilities involved in gaming is just to make a two-column table, with the first column listing all the possible outcomes and the second column recording their associated probabilities of occurrence.
An important characteristic of a probability distribution is that the sum of all of the individual probabilities calculated for each possible outcome of the process will always be 100% [because something will happen].
And, by looking at the probability distribution, we can readily discuss the relative likelihood of an event as compared to another [by comparing their probabilities], as well in the context of all possible outcomes.
Lemme 'splain; no, dere ees too mush. Lemme sum up.
In summary, when we say a process is "random", we mean that there is no way to tell what the outcome of each run will be, but that we can predict the likely distribution of results over multiple repetitions.
Additionally, if we know the set of possible outcomes of the process and the probability distribution for it, we can at least say how likely a given outcome is even if we can never be certain it will happen.
Urning a living. Or, what are the types of random events?
Before we can discuss randomness in the context of gaming, we need to define at least one more element of the proper discussion and precise interpretation of random events.
If you've ever studied probability formally, you surely have encountered a problem like the balls-and-urn example. Probably you thought "What the balls is the deal with 'replacement'?"
In the classic urn-type problems, "with replacement" and "without replacement" are conditions that serve to mark off two different types of sequences of random events. If you pull a ball out of an urn, note its color, and then toss it back in ["replace" it], the next time you or anyone else pulls a ball out of the urn the probabilities of getting any particular color will remain the same as they were before your first pull. We say these different pulls [events] are independent of each other. If, however, when you pull the first ball out of the urn, you note its color and then stick it in your pocket ["without replacing" it], the next time you or anyone else pulls a ball out there will be one less of that color in the urn and the probability of getting that color [or, indeed, any other] will be different than they were before your first pull. We say these events are dependent.
Note that by "dependent" we do not mean that the actual color of the second ball you pull from the urn "depends" [in any way] on the color of the first ball you pulled out. What is meant is that the probability of getting any given color for the second ball depends ["is affected by"] the color of the first ball you pulled out [and didn't "replace"]. Again, this sounds like pedantry, but precision is very important when talking about probabilities. Even though we will commonly say that the "events are dependent", we primarily mean that their probability distributions are dependent, not that their eventual actual outcomes are. If the probability distribution for the second event in a series changes based on what happens in the first event, the two are dependent. If the probability distribution for the second event is the same no matter what happens first, they are independent.
In gaming, the most common examples of dependent and independent events are card draws and die rolls, respectively. If you draw a card from a deck, put it in your hand, and then draw another, the probability distribution for that second draw depends on your first.
Contrarily, if you roll a die, write down the result, and then roll another die, the probability distribution for that second die roll does not depend on any way on what happened in the first die roll. The common adage is that "dice have no memory".
Bonus Content: It's time to play.... Probability Mythbusters!
Most people with an modicum of probability training understand the "dice have no memory" thing, and even if they subconsciously feel that a string of six die rolls that results in 6-6-6-6-6-6 is more unlikely / odd than a string of six die rolls that results in [exactly] 6-5-4-3-2-1 or [even still] 6-2-3-1-5-4, they probably understand that each of these particular results [exactly as given] is equally unlikely. If, after the series of six 6's, they were asked if the odds were any different of the next die roll being a 6, they'd probably answer [out loud] "No, still 1/6 chance" even if they subconsciously felt otherwise. Where intelligent, educated people still do go wrong with predictable frequency is in the interpretation of the so-called "law of large numbers". The mistake is easy to make, and is related to the [rarely made] mistake of attributing dependence to a string of die rolls.
The mistake is this: The law of large numbers says [roughly] that with an adequately large sample size [number of die rolls], the observed relative frequency distribution will "look like" the predicted or theoretical probability distribution. That is, if we roll a die 600 times, we would expect 100 1's, 100 2's, 100 3's, and so forth, based on the probabilities of each possible outcome of a fair die. In any given series of 600 actual die rolls, we will observe slightly different frequencies [the odds of actually getting exactly 100 of each face are astronomically small], but the overall set of outcomes will probably be fairly evenly distributed across each result. Now, let's say you've already rolled 300 of those 600 attempts, but have only seen 25 5's [half of the expected 50 5's]; how many 5's do you expect to roll in the next 300 attempts? Will the overall frequency of each result "even out" in the end?
Many people's intuitive response is wrong. The correct answer is that the next 300 rolls will probably have 50 5's [same expectation as before] and that the overall number of 5's in this particular set of 600 rolls will likely be 75 [rather than the 100 expected], meaning that your final set of results will almost certainly be quite uneven.
The law of large numbers only says that as you increase the number of rolls in your sample, the extent to which you can expect the observed distribution in those rolls to match the predicted distribution increases. It does not say that a sample already observed to be unevenly distributed will be "corrected" by additional repetitions.
The additional repetitions themselves would have to be unevenly distributed [in the opposite direction] for that interpretation to be true, but the law says that they'll likely be even.
A further misunderstanding is that there is a magic number or a numerical threshold of rolls beyond which you can "start expecting" evenly distributed results from your sample.
Gamers will lament that "there are not enough rolls for the results to even out" in a game or ask "how many rolls should I have if I want to be sure that my game is balanced?"
The truth is that you always expect an even distribution, but that you can never be sure of it. The law only says that increasing rolls increases the expectation's accuracy.
Death and taxes. Or, is the only thing certain uncertainty?
The Ludology podcast I referenced previously began with the observation from Ryan that [paraphrased] "What most people seem to mean when they say a game is random is really that the game has a lot of uncertainty." This calls to mind two questions: 1) Is there any merit to this characterization?; 2) Is the distinction itself even significant? I'll attempt to tackle #1 a bit later. First we should ask, is there really a difference between randomness and uncertainty? Isn't is part of what randomness means that the results aren't certain ["indeterminate" was the word I used previously]? Is Ryan making a distinction without a difference for the purpose of filling in time in the podcast? I don't think so, but I don't think the Ludology team kept up the distinction carefully enough throughout the podcast, and probably assigned a bit too much weight to it.
Randomness is not distinguished from uncertainty, wholesale, but is a particular type or instance of that broad category. Something [the price of eggs in China, say] can be "uncertain" because it is unknown [by the person interested in knowing it], or because it is patently unknowable. Presumably eggs have some relatively deterministic price in China, and my uncertainty over what that price is does not arise from the fact that the price is randomly determined, but from the fact that I simply don't have the information I need [first-hand experience or a trustworthy second-hand account] to determine it. That this particular type of uncertainty does not arise from randomness does not mean that all uncertainty is simply a lack of information. When I drop an egg, the shape of the resulting splatter is uncertain, but it is also random and fundamentally unknowable.
I've used the distinction between "unknown" and "unknowable" previously [in describing probability distributions], so I should clarify how I intend to use these terms going forward. There are random events [the roll of a die] with known and knowable probabilities that nevertheless generate unknown and unknowable ["indeterminate"] results. There are also random events that generate these indeterminate results and also have unknown and unknowable probability distributions; you don't even know what you don't know! A stupid example might be to go to a library, roll a d100, count off to that shelf, roll a d100, count off to that book, roll a d100, count off to that page, roll one last d100, count off to that word, then read off the 3rd letter. In some very broad sense [frequency of letters in the English language] there might be some knowable probabilities, but practically speaking you're ignorant.
You might catalogue all the books in the library [to see the overall probability distribution], but even then the actual books on the shelf will change day-by-day in such an unpredictable fashion that the probability distribution that defines your random choice remains basically unknowable. But, whether you can know the probability distribution that you're drawing from when you conduct some random process, the process itself still generates results that you cannot possibly know beforehand because of the randomness of the die rolls or whathaveyou. All random processes, by definition, generate indeterminate, or unknowable, results. And it is the distinction between the knowability and unknowability of results that sets randomness apart within the broad "uncertainty" category. This is only one kind of uncertainty; the other occurs in processes which generate uncertain results that are nonetheless not fundamentally unknowable.
What I think Ryan was getting at when he characterized gamers who decry particular games as "too random" as being really more concerned with "uncertainty" is the type of tenuous, unstable, tip-of-your-tongue knowledge of pseudo-predictable processes that, say, economic agents are faced with. The one thing that characterizes and drives the economy more than any other is uncertainty, not only about what government regulators will impose on economic agents but about what other economic agents themselves will do. This is uncertainty of the second order [in my little system above]; it is uncertainty about events or outcomes that are unknown but are not [theoretically] unknowable. Human beings do not generally make decisions randomly [in fact, it's a "failing" of humanity that we cannot even imitate randomness well] and the economy is driven by human decisions.
Economic activity is fraught with the particular kind of uncertainty that Ryan had in mind when he described games that are often lamented as "too random". "Players" in the economic "game" are attempting to guess the plans and future actions of their "opponents" [or "temporary emergent alliance partners", really] in order to respond preemptively to those actions to realize better gains, be better prepared to respond to market demand, or whatever. This process is characterized by a high degree, as expected, of uncertainty; marketers and strategic planners often lament "I wish I only knew ____" or "If we could only predict ____ then we could ____". But, the uncertainty here is one of ignorance and not of blindness. Fundamentally, the behavior of customers or markets [i.e. "groups of customers"] is knowable. There is one problem, and I think it's what Ryan was after in his analysis.
The problem is that large numbers of actions or decisions that generate uncertain yet knowable results can appear truly random when aggregated. You could ask an individual investor what they intend to do if stocks fall to a particular level; you could predict that behavior based on simple economic laws; you could even model the behavior of a large number of individual investors with a complex system of equations and coefficients. But, because the entire stock market is so intricate and interwoven, it is beyond the reach of practical predictive or interrogative ability to arrive at much of a useful conclusion at all about what will happen. This is the essence of the so-called "random walk hypothesis", which says that stock market behavior is best modeled by a truly random process [the hypothesis does not really say that stock prices are actually randomly determined, only that no non-random model predicts them better].
The stock price chart above looks believable enough [for some definition of "believable"]. It is entirely fictitious, however, and is essentially [though not actually, for sake of example I guess] the result of a random process: Somebody took the decimal digits of pi and plotted ups and downs based on the odds or evens in that irrational and [for practical purposes] random number sequence. Yet it has been found that similar random processes are statistically indistinguishable from actual, non-random stock market "trends". With a tip of the hat to Clarke's third law, what we find is that the result of any sufficiently complex uncertain process is indistinguishable from randomness. This is a good explanation for why someone might characterize a game that is full of player-generated [i.e. human-generated, so non-random] uncertainty as being "random"; I think it remains to be seen whether people actually make that characterization.
Roll and Move! Or, what is the role of randomness in gaming?
Randomness does one thing: Determine the outcome of a data-generating process. An appropriate analysis of randomness in gaming focuses on that definition. We find ourselves with two questions, then, in our interpretive framework: 1) What are the "data" in gaming?; 2) What are the "processes" in gaming? Answering these two questions can give us a clear structure for analyzing what randomness does when inserted into various parts and functions of a game design. Data and processes are more fundamental concepts than those we typically talk about in articles like this: mechanisms, dynamics, rules, themes, categories, interactions, etc. Think of "data" and "processes" in gaming as being analogous to assembly or machine code in contrast to higher-level programming languages. Software [a game] is written in a higher-level language [mechanisms], but depends implicitly on the underlying assembly language structure [data and processes].
Some examples of data and processes in gaming:
- A number representing a monetary value or price
- A number representing a threshold or criteria to be met
- A number representing a movement or action allowance
- A number representing a limit or count of components or their use
- A category representing the ownership or availability of a given item
- A category representing the color or type or other characteristic of an item
- A category representing the identity or function of an item appearing in quantity
- A category representing the size or shape or other physical attribute of an item
- A vector representing the physical position of an item in relation to other items
- A vector representing the orientation or configuration of a given item
- A vector representing the movement or future course of an item
- A vector representing the influence or range of an item
- An action that involves obtaining or identifying a number
- An action that involves manipulating or transforming a number
- An action that involves comparing or relating numbers or categories
- An action that involves reference to or otherwise checking a category
- An action that involves implementing a vector to move or activate an item
- An action that involves modifying a vector to reposition or configure an item
The particularly bright students in class today might notice that the real analogy here is between variables and functions ["data and processes"] and programs or software [mechanisms or games], not between assembly and high-level code. You are [obviously] correct, but that metaphor doesn't have the virtue of capturing the "under the hood" / two-level feel that I'm after of dissecting games down to this detail of analysis. I'll concede. We don't typically talk about games in these terms, just like we [that is, non-programmers] don't typically talk about computer software or programming in terms of variables and subroutines, but of overall mechanisms. It is at this level, however, that randomness makes itself felt most directly on a game, and so it seems appropriate to begin our discussion of randomness at the exact point at which it contacts the data and the processes that generate or use it. To understand randomness, we need to think like game programmers.
Does it make a difference if the data in a game [the numbers, categories, vectors, configurations, etc] are determined by random algorithmic processes or if they are the [thinly] veiled or otherwise obscured result of a set of deterministic choices? Does it matter if the "program" to call up a required number in a game is sufficiently complex as to be practically indistinguishable from randomness ["check the price of eggs in China; subtract the weight of coffee left in the bag in your cupboard; then multiply by the number of players in the game"; or, more applicably, generating a number associated with the amount of cubes left on a particular tile in the last turn of a game of Dominant Species after the final action selection and activation round is through... recall that nothing is random after the initial setup of each turn] or is truly random? And if it does matter, how so? What is the effect of randomness, included or excluded, on how a game plays?
I think one of the larger effects of randomness in general is to shift the focus of the game from player-vs-player to player-vs-game. There are, essentially, only about 2 or 3 different possible ways to obtain a piece of data necessary to play a game; it is either 1) Randomly determined by some game process [roll a die]; 2) Deterministically generated by the actions of the players [a choice]; 3) Predetermined by the structure of the game [a preset value or a value that changes by phase or other passive function]. #3 can be ignored for our purposes, because the heart of our discussion is uncertainty and a preset table of values based on criteria which exist in the game at the time represents total certainty. We are left with uncertainty that is resolved by observing player action or by resolving a random process. It is not that randomness removes or prevents player interaction; it is that it reduces its dimensionality.
Consider a roll-and-move "toy game" with a few pawns for each player that race around a track and can knock each other off by exact movement to convergence, or even an actual roll-and-move game like Pachisi or Backgammon. But, let's limit ourselves to such games that are actually games [and not like Chutes and Ladders or Game of Goose which have no player choice in response to the die roll and are literally "roll and move" pastimes]. Even when the available movement is determined by a die roll, players still have a choice of which pawn to move and may choose [as available per the die roll] to interact with each other by direct conflict [knocking a pawn off the board], by blocking / limiting [tying up a point or space], or by racing / applying game tempo pressure [moving ahead more quickly toward the end of the game, probably at higher risk, by passing up the potential to set up knocking or blocking moves].
What might happen if we choose to substitute the random process [a die roll] generating the "movement allowance" "data" for a non-random, yet still uncertainty-laden, process dependent on player choice? Let's say that our game was based around a single die roll previously; now let's give each player a set of tokens numbered 1 through 6 from which they may freely choose their movement points each turn. Let's add the wrinkle that they must use all of the six tokens one time each before they may reuse / refresh their supply of possible movement values. The overall distribution of movement speeds by game's end will be roughly the same [namely, uniform] in either the original "roll and move" game or in our new "choose your own adventure" game. Additionally, all of the old strategies [knocking, blocking, racing, etc] will essentially still be available [racing / risk is the most vulnerable, but it was also the least interactive].
But what will have changed? There will now be a new element / dimension of your opposition's strategy and tactical responses for you to consider when making a move. Previously, if you left yourself potentially open for attack, the question was "If they are able to, will they knock me off?" Now, the question becomes "Seeing that they are able to, will they knock me off?" The reason the latter has greater dimensionality has to do with the opportunity cost function involved in player choice under either scenario. If your opponent needs a 5 to knock your piece off, and they roll a 5, their decision whether to knock you away depends on what else they can do with that roll of a 5. If, however, they are free to choose among any of their remaining movement numbers [let's say 1, 4, and 5] and 5 is among them, their decision depends not only on what else they could do with a move of 5, but a move of 4, or of 1, or of whatever else they are able to choose.
Suppose if you leave yourself open for attack that there is nothing even remotely better for your opponent to do with a movement rate of 5; if that is their only choice, they will with 100% certainty choose to knock your piece. Under the die-rolling scenario, your decision of whether to move yourself into the position of being potentially knocked off the board does not depend on what you expect the other player to do at all, but on whether you are willing to take the risk that the game [effectively] knocks you away [with your opponent serving merely as the vicar of the cruel winds of fate]. You will weigh the possible gains of that move against the probabilistically weighted cost of having the piece knocked, and make your choice. If, however, we are working under the scenario where the other player may decide either to use a 5 to knock you or to use another available number to do something else, things will be different.
In the non-random case, your decision depends on your expectation of your opponent's evaluation of the 5 move against the potentially "better" uses of other movement values available to them. Simply because the data point "number of spaces your opponent can move" will be determined by a non-random choice rather than a random process, there now exists a broader range of possible moves over which you must anticipate and prepare for potential responses by your opponent. If you are "surprised" by what happens on your opponent's turn, it will not be surprise that a 5 came up on the die and you got knocked [because it was the only reasonable action] but because your opponent evaluated the game position differently from you and chose to either knock you away or not based on that evaluation. Your task is not merely to compute expected values, but to judge and assign subjective probabilities themselves to the particular responses possible by your opponent.
In short, the uncertainty and the richness of opportunity cost decisions and positional evaluations is often greater in games where more data is determined by player choice than by random outcomes. This is neither inherently bad nor inherently good, and is itself a trade-off. Games that have less randomness must generally be more mechanically complex than games that have more randomness, or else risk dominant strategies; if you could choose your movement value completely freely in our toy game above, you'd probably always choose the highest possible value. Only the complex number-cycling rule added to the luckless game keeps the game playable, whereas the die-rolling game needs no such overhead. More complex games tend to be more chaotic, and this is only amplified when the dimensionality of uncertainty is higher [as it generally is in games with more player choices]. Is the added richness of decision evaluation worth this cost? Who knows!
The boys here have made a few changes to the classic luckless game of Chess. But, the substantive nature of the change really isn't that they've added die-rolling, but that they've added data [as an aside, if they merely wanted Chess with die-rolling, they could have played its ancestor Chaturanga]. The essence of what has been added is data about the board and the pieces and their interaction: What pieces can move first? How to pieces attack each other? What if anything does the terrain of the board do to pieces? These are concerns that require very very little data in Chess: Players take turns [Data: "1" or "2"]; combat is quick and decisive [Data: "1" or "0"]; terrain consists of black-vs-white for the bishops and home-rows for pawn promotion [Data: "B" or "W"; "Home" or "Not"; or just "1" or "0" again]. As true thematic gamers, however, Jason and Marcus want to know more about the state of the game. This requires a finer level of detail and more things to collect data about.
Why did they choose to generate this new data randomly? It could be just that, as role-players, they had a bunch of nifty polyhedral dice that they couldn't stand seeing neglected on the shelf. That's a simplification [even abstracted out of our fantasy world of the Foxtrot comic strip] and the real reason probably isn't as silly. The reason I think a lot of game designers resort to randomness to generate data is that they are attempting to strike a particular balance between the three competing factors of uncertainty, chaos, and complexity that is difficult to achieve in a game entirely driven by player decisions. Because these three factors tend to be interdependent, there is a bit of a [self-referential] butterfly-effect to increasing any one of the three. The minute you add additional complexity to a game, you increase the amount of uncertainty [due to player choice or random chance--it doesn't matter], and that in turn increases the chaotic dispersion of outcomes.
That broader set of possible outcomes itself increases the decision complexity [apart from the rules / mechanical complexity] and difficulty of "ply analysis". In short, the resultant difficulty of learning to play the game [much less learning to play it well] and to make reasonable predictions concerning your opponent's moves increases exponentially [I believe] because of the tri-fold sensitivity of uncertainty, chaos, and complexity to each other. There is a tendency, I find, for games without any chance elements to fall victim to one of two patterns of design that don't suit most gamers: either the game 1) is too simplistic to be consistently interesting [it risks solution]; or 2) is too complex to be enjoyable without intense study [it resists learning]. The problem is that, from a simple germ of an idea for a luckless game, any significant increase in "bulk" will tend to result in a momentous preponderance of ever-growing uncertainty, complexity, and chaos. Getting a luckless game to be accessibly deep is incredibly difficult because the system has no checks or balances or limits.
Adding randomness to such a system can actually decrease the extent of chaos and uncertainty in a game because it places a hard limit on the amount of information that must be considered in making a decision, or the amount of game data that can conceivably be affected by player choice [and is thus not susceptible to typical probability analysis which, as we'll see, is actually a very reliable decision-making tool]. Imagine a role-playing game in which every initiative check, every damage value, every hit percentage was deterministically generated by some player-driven function; the game would be a slog to play. "How many CP will you bid for initiative? How many CP are you going to spend to attempt to interrupt?", "Compare the speed and reach of your weapon to the speed and reach of your opponent; now determine whether the modified strength value of your attack is stronger than the defense value of your opponent, modified by armor if your weapon is piercing", and other similarly arcane systems of data generation would arise. [These games actually do use randomness elsewhere.]
Randomness effectively cuts short the look-ahead tree of decision-making. There are fewer possible counter-moves to consider the ramifications of, because there are fewer things which your opponent can control in the game. At a certain point, you must resort to calculating the probabilities of particular outcomes, which [although often unintuitive] are relatively easy to work through and use in decision-making. Even if you're not the analytical type, there's a certain comfort level attached to a decision which will largely be met by a random response as opposed to one that will be met by the freely chosen response of your opponent. To run into "bad luck" is one thing; even if your decision was demonstrably "wrong" mathematically / probabilistically, you can claim to have been "taking a chance". To fall into a blunderous oversight like Marcus' neglect to ready his resurrection spell is different; you rightfully feel embarrassed and a bit silly, and you have no one but yourself to blame. That was the one piece of data he had to take responsibility to generate, and he screwed up!
Does this all mean that games with random elements are necessarily lighter games? Well, not if phrased in those exact terms, no. But if we appropriately apply ceteris paribus conditions, then yes. Absolutely. Games with random elements are, in fact, necessarily lighter [because the look-ahead is shorter] than they would have been had the random data been generated by a player-driven process instead. [The "ceteris paribus" condition, if you're not familiar with the term, is that we are comparing games to themselves / alternative default assumptions, not to other games... Jason and Marcus' D&D Chess is still a heavier game than Tic Tac Toe, even though the former is random and the latter is not.] This doesn't mean that games with randomness are all light games either. There is a particular type of depth that deals with understanding and intuiting probability structures, and it is often extremely difficult to play luck-driven games well, but the broad concept of randomness / probability remains more predictable than player-driven choice [because humans are irrational creatures].
The essential value of randomness to game design is that randomness is manageable whereas players are not. Or, at least, players are a particularly intractable admixture of jackasses, smartasses, dumbassess, wiseasses, and other asses yet to be assessed. The winds of fate are cruel and capricious, but they are not crafty and do not conspire against you. We tend to personify "good luck" [Lady Luck, innumerable gods and goddesses, etc], but general bad luck doesn't seem to anthropomorphized as frequently. I think, perhaps, this is because we instinctively feel that misfortune that could be attributed to humanity is much more malicious and ill-intentioned in character than simply drawing the short straw. Human representations of bad events are given such traits as vengeance, fury, cruelty, and so on. This carries over, of course, into gaming. Even beyond the typical social meta-gaming aspects of good-natured in-fighting and non-productive revenge-seeking, human actions are subject to a type of caprice that dice just aren't. After all, they have no memory!
If some piece of data or some process is essential to the proper functioning of a game, it is dangerous as a designer to leave it up to the players to generate that data or process. This can be a particularly compelling design move, no doubt, but it is not one that should be taken lightly. There is nothing inherently wrong with leaving a critical bit of uncertainty in the game data up to the roll of a die or the turn of a card. In fact, I believe it is sometimes the appropriate design choice in order to maintain the desired level of depth appropriate to the game and its audience. There are games that break in "mixed company" [analytical vs casual players, experienced vs rookie players, etc] because there is not enough randomness to prevent a player with a significantly better understanding of the depths of the game from leveraging that knowledge to secure the win. Often this is fine, even desirable; sometimes it is clearly not. So, if these are some of the roles randomness can play in a game, when is it appropriate to design in random elements?
Designers tend to introduce randomness into a game as an easy, and easily manageable, method for creating numerical data that needs to have a degree of uncertainty [die rolls for hit success or attack damage; price or price movements of a commodity; etc]. Another common place for randomness to occur is in the generation of qualitative information such as the availability of assets [power plants, plantations, plasma missiles] or the content of a player's asset holdings [drawing or being dealt a hand of cards; receiving resources based on a die roll; etc]. One other important implementation of randomness often occurs in the setup or procession [phase changes, etc] of factors that determine the prevailing environment in which a game will be played [what buildings will be available; how will resources be distributed at the beginning of the game; what will the game terrain look like; etc]. In the podcast, Geoff used the terms "input randomness" and "output randomness" to distinguish between possible applications of randomness within a game design.
I think these are fairly good terms, but I'd like to add one more category and draw the distinction a bit differently:
Random outcomes occur when a game process has an uncertain effect on the state of the game, which will be resolved by a random mechanism rather than by player choice [Geoff's "output randomness"]. All of the most obvious examples of randomness in gaming, including roll-and-move as the exemplar, fall into this category.
Random opportunity occurs when game players are collectively faced with an available set of actions or effects which are uncertain until revealed and are determined randomly [Geoff's "input randomness"]. Players may or may not have equal ability to pursue the opportunity presented, but their access to it is unrestricted.
Random empowerment occurs when the available opportunities in the game are presented differently to different players, or when players do not have equal access to every possible action, and when this asymmetry is determined randomly. This is a form of "input randomness" and "random opportunity", but worth separating.
[This basic structure is something I've used for awhile; my very first post in this blog touched on them in #7 and 8.]
These three terms in particular are useful [or distracting, I suppose, depending on your proclivity] because they are familiar from a field of philosophy more relevant to most of us than Geoff's mechanical-mathematical metaphor of inputs and outputs: namely that of political or social justice, fairness, and equality. They are also useful because this association emphasizes the result of including randomness in a game rather than the mere nature of it. Randomness, of these varying types, tends to introduce unforeseeable inequities of varying severity into a game. When only outcomes differ [from identical process inputs, as they do in an "output randomness" situation], the process may be said to be unfortunate, unkind, unfair, and perhaps even unjust. When the opportunities themselves also differ ["input randomness"], but the outcomes follow deterministically [or even randomly, really], the process may still be unkind, but would probably not be characterized as unjust or unfair. When the empowerment, in particular, to take advantage of or gain access to available opportunities differs, the process is probably somewhere in between.
In Entdecker, as an example, a random process determines each player's income for their turn, from 2 to 5 gold on a 4-segment spinner. Even though there is no systematic bias [because both opportunity and empowerment remain equal for all players], the process still generates inequality of incomes and feels unfair.
In Löwenherz, as a counter-example, a random process determines the availability of gold income for all players on a turn. The value, from 4 to 8, changes each turn, and the more players that choose the gold action on a turn the less they receive individually. Although the low income card may come at a particularly inopportune moment for you, you are not being treated unfairly.
In Pfeffersäcke, a middle-of-the-road case exists for randomly-determined income generation. Players are dealt discs randomly with values ranging from 2 to 8, which determine the cities they have control over "opening" for trade. Being able to open a higher value city means you determine when it enters into the game and have a greater ability to get in on its high income before other players can. You still have to work at it, so this feels only partly unfair.
Using our prior analysis of player-based vs game-based interaction, it seems to me these different uses of randomness increase along the dimension of player interactivity in roughly the same order as they do in perceived fairness. At one extreme is the system in which the success of your actions or growth of your resources is not a matter of choice at all, but the result [output] of a random event; this feels very non-interactive. Along the way toward total player-driven interactivity, we progress through systems in which a considerable portion of your fate is determined randomly by the game and not by your decisions [you have low or random empowerment], and systems in which the only thing holding you back is the interaction between your and your opponents' responses to randomly arising opportunities. There is greater player agency, unsurprisingly [but I also think unobviously], in games with only random opportunity instead of random outcomes. I think, in part, this distinction can help to make sense of the ever-controversial Ameritrash vs Eurogame / "thematic" vs "strategic" categorizations and explain both why they are and what they mean.
To Dice or Not To Dice? Or, wherefore art thou, oh Rolleo?
One of the prevailing and most often cited differences between Ameritrash and Eurogames is not just the use of more randomness in AT titles, but the use of particular types of randomness designed into each sort of game. It is by no means a rule without exception, but in general "Ameritrash" games tend to contain more randomness in the generation of outcomes than of opportunities, more "output" than "input" randomness; contrarily, "Euro style" games [that include randomness at all] tend to contain randomness of a different sort, primarily concerned with the generation of random opportunity. Empowerment is, again, a middling case, used by both Euro and AT games [I think perhaps one distinguishing sub-categorization might be the number of ways in which randomly distributed empowerments can be deployed in a game, with Euro games tending to have more possible uses for your random power alottment]. You will not tend to find "rolling for damage" or "skill checks" or random movement in Euro games; similarly, you will not generally find complicated random resource distributions such as Yspahan's tower or Assyria's food card display or Age Of Steam's cube production in an Ameritrash title.
I believe a large part of the rationale for these observed preference in sub-cultures of gaming, as well as for the "thematic" / "narrative" alternative [and sterile] description for the "Ameritrash" genre, has to do with the concept of agency: how it interacts with story-telling and how various types of randomness create or limit it. If a game is intended to convey a "story", I believe there is a counter-intuitive need for player's in the game to have low agency, which as we've discussed already necessitates greater "output randomness". This has to do with story as a literary device. For all of the character development and birds-eye, action-driven writing that goes into telling a good story, it seems to me that the plots of most great stories involve the protagonists playing very passive roles and being put into situations where they have low agency. If something doesn't "happen to" the protagonist, if there is no "ordeal" they must go through, if nothing with a "will of its own" gets in their way, then the story simply doesn't have the same dramatic appeal. Stories all but require "outcomes" that are not the fortuitous result of being in the right place at the right time [opportunity] or of having just the right abilities or assets to tackle a situation, but the unpredictable result of something that "goes wrong" or otherwise behaves in an unforeseen way.
Overcoming unforeseen obstacles and dealing with unintended consequences are the making of all of our heroes. Heroes don't have control over everything; they have relatively low agency to change the "way things are", despite their ability to adapt and overcome and perform great feats of strength or fortitude. BGG's resident solo gaming blogger, Patrick Carroll, once drew the distinction between "god games" and "hero games", categorizing "hero games" as those with more "hidden information and chance" and "god games" as being more "deterministic". I think he got this exactly right, although I would go one step further and clarify that a "hero game" is one in which the outcomes [in particular] of your decisions are more uncertain or random, where there is more "output randomness". I think these are the feelings and stories that Ameritrashers are trying to experience when they play a "thematic" or "immersive" game, and what they mean when they say [or at least BGG says] that these games focus on creating a "dramatic narrative". It isn't just that Ameritrashers like throwing around buckets'o'dice, but that the particular randomness associated with using such mechanisms to generate random outcomes helps create stories.
In contrast, Eurogames tend to provide greater player agency, because the experience they are intended to create is one of world manipulation [at least on a microcosm-ic level] and self-actualization. Even the player avatars in your stereotypical Eurogame theme tend to have greater real-world agency than the typical AT character: Players become wealthy aristocrats, powerful politicians, kings and emperors, factory managers, corporate executives, and so on. Themes and characters associated with struggle and low agency, as in Agricola or Antics!, are by no means the norm. Very few AT games, in contrast, focus on the Euro tropes of "building", "managing", "control", or "development". These represent agency, and the addition of high levels of randomness associated with the outcomes of game processes [as is desirable in an AT title] would serve to reduce player agency and destroy the "god-like" role of player-as-overseer. When random outcomes are used successfully in Euro style games, it is always in a very specific context to model something that is clearly beyond the limits of the particular type of agency being assumed by players in the game; it is rarely done merely to introduce obstacles for players. Obstacles in Euro designs arise from non-random factors such as complexity or logistics or efficiency.
How and why does randomness ever get into a game design in the first place? A lot of designers seem to follow the pointy-haired-boss moel and simply insert random features as a way of creating the impression of value to a particular demographic. Designers, whether working in the Ameritrash or Euro or any other genre, need to really appreciate the different types of randomness and the impact they have on a game apart from their mere immediate mechanical implementation. I think the really good designers out there are better at doing this [choosing the right form of randomness, if any, for their games] than we generally give them credit for. Rather than looking at the impact of randomness on a game or the way in which it shapes a player's immersion into the system, we tend to just look at the sheer amount of uncertainty that a particular random mechanism creates. I think that the core misunderstanding of randomness by gamers, rather than being Ryan's "they really mean just that the game is full of uncertainty", is that many gamers tend to view randomness as an artifact or an ornamentation of a design rather than as a structural architectural feature upon which the game design is built.
Randomness, as we've already discussed, is a very low-level game design feature; it is a foundation rather than a decorative element. It tends to be glaringly obvious, however, when a game process is handled randomly [the most iconic symbols of gaming, after all, are cards and dice], so we think that it is the superficial, surface-level portion of the game and that all of the depths of the design lie beneath it. This simply isn't the case. Well-designed games, when they include randomness, are built around the underlying randomly-generated data. The randomness in such games exists not only to provide simple uncertainty, but to create particular types of decision-making situations and to control the numerical data of the game in a specific way. Game mechanisms exist merely to manipulate the various data [mostly numeric] of the game; it is in the interaction of that data that the game itself really subsists. Randomness serves to create particular genres of data interaction. Surprisingly, it can do this with a degree of uncertainty while maintaining a greater level of predictability and control than can many systems relying on players to manipulate the data directly.
Hallmarks of good implementation of randomness in a design include openness, intentionality, and sensitivity. Openness has to do with the susceptibility of the random processes to appropriate player analysis. Die rolls are familiar and easy to interpret or predict, so it is no surprise that they continue to be popular methods of generating data randomly. It is not satisfying to play a game in which a particular piece of relevant game data will be determined randomly according to a probability distribution that is obscured or inscrutable or simply too complex to be grokked. Particularly helpful are games like Settlers or Dungeon Petz that provide summaries to assist players in interpreting the possible and likely results of random processes. I think the combat mechanism in Eclipse serves as a recent standout in the "openness" of its random resolution. It is extraordinarily clear how the process works, what the probabilities are, and how every modifier or buff affects the relevant parameters. Related to openness, I think, is the concept of designer intention. If it is not clear from playing the game why the designer included a random element or what the random process is supposed to simulate [or even do] in the game, it will also often be the case that the random process is overwrought and too complicated for its own good... as if the designer just gave up and left everything up to dice, the easy fall-back solution.
One evidence of strong designer intentionality in including randomness is in the number of systems in the game that use random processes and in the number of different types of processes that are used. Well-designed games, built to include randomness for very specific purposes, use only as much randomness as necessary and each random process usually has a unique and appropriate resolution method. Finally, another evidence that the designer has thought about the impact randomness will have on the game is that the sensitivity of game outcomes to the varying data of random processes is appropriate to the length and depth of the game. It is not merely that short games can have more randomness and long games should have less, but that long games with significant player investment and decision-making should not generally have randomness that can overwhelmingly determine the outcome of the game. Inherent in randomness is the concept of variability of outcomes; if one of the possible outcomes, however rare, can have such a dramatic impact on the game that it all but entirely determines the eventual winner, it ultimately doesn't matter how well controlled or designed the rest of the random processes are. Contrarily, if the randomness in a game has possible effects appropriate to the game's design, there can be a very large amount of randomness even in a very long game and the game can still be a satisfying experience. This doesn't even mean that randomness must always be insignificant; it simply means that randomness should be appropriately significant for the amount of risk, preparation, and potential reward it offers.
Editor's Note: Apparently there is a character limit in blog posts? The blog continues and concludes below in the first comment.
- [+] Dice rolls