Thumb up
4 Posts

Outpost 18» Forums » News

Subject: Outpost 18 Dev Stuff rss

Your Tags: Add tags
Popular Tags: [View All]
Adam Wilk
flag msg tools

Hey guys. I’m Adam. I’m designing this game, Outpost 18. Welcome!

Rules —


Outpost 18 is a design with a lot of intent. It’s the game I’ve often “dreamed of” — small enough to it in a coat pocket, and complex enough to offer an engaging game on the go.

Micro games are often synonymous with light weight, but I don’t think that has to be the case. There’s a way to make something compact and still satisfying. Small and crunchy.

The challenge was to make a tight and snappy feeling strategy game with only 18 cards. Why 18 cards? It’s just enough to add complexity and weight, but still be extremely portable.


I wanted something that had hints of strategy and classic TCG themes — like aggression, card advantage, board manipulation, control, hand management, and so on. I wanted to make something that doesn’t require mental upkeep or board fiddliness — so there’s no tokens, no resource spending/tracking, and so on. Instead, cards operate on a threshold system that keeps information on the board all the time.

The game is broken down in to three resources. Ore is mined from the asteroid belt and refined in to advanced weapons. Ion is captured energy stored in cells that can unleash powerful technologies. Labour is the scarcest resource in the belt, and strong workers create better utility for your station.


Ore-based builds will generate a lot of offensive punch.

Labour-based builds will generate card advantage.

Ion-based builds will generate board control.


This thread is going to highlight the design history of the game, specific design tweaks, and broader stroke changes. Each post will highlight a design issue and how I approached it.
 Thumb up
  • [+] Dice rolls
Adam Wilk
flag msg tools

Outpost 18 is a game meant to be played quickly. Ten minutes is a match. Fifteen is a long match. So when play-testing began, I was surprised at the tester response. Games were taking longer than expected.

Sometimes, way, way longer.

It’s not uncommon for the first game to go slowly in a duelling card game. Any game with sufficient depth is going to have a learning curve. Outpost 18 is, by design, a heavier micro game. Cards interact and trade in unique ways. There needs to be room for this in a strategy game — something more than surface deep.

In Outpost 18, you can play (more or less) right away. But a seasoned player will have developed an understanding of what is left in the deck, the scarcity of resources on the board, and the value of hand management. Do you need to know these to enjoy a game? No, I hope not, anyway. But the room for mastery is there. And the better, more seasoned player should win almost every time.

That’s where things loop back to game length.

Like Game of Thrones, the problem is getting to the ending.

My first play testers reported back that their first game was thirty minutes long, or longer. Okay, no problem — you need time to reference rules, process all the new information, and figure out the song and dance, so to speak. But their second and third games were still long — twenty-five, thirty minutes, or more. Keep in mind: this game is meant to be played while waiting for a meal at the bar, or waiting for your Uber.

So I took to watching a few players in action. I kept quiet in the corner and let them explore their first game. They grasped the rules pretty quickly, and within five minutes, they started their first turn. And their cogs began to turn and try to unravel the puzzle in front of them.

A player mulled over which card to deploy, and opted for an early upgrade. Their opponent, armed with an extra card, eventually decided to do the same. Back and forth they went, turn three, turn four, turn five, upgrade after upgrade, stockpiling resources and card draw, until finally, a player opted to play an aggressive card.

That game lasted thirty-four minutes, and could have gone longer. The last ten minutes were slow, overly-calculated, and dry, as the card pool dwindled and nobody took initiative to end the game.

Yikes. So what gives?

Engine builders, duelling games, and resource management games in general have taught us that greed is good. In deckbuilders, you can hone the timing of your engine to kick in at the perfect time. In magic, card draw engines and hyper efficient mana cards create an accelerated sense of momentum. So when someone approaches a game with resources, the natural instinct is to take as many of those resources as one can, because greed is generally rewarded.

At it’s very core, Outpost 18’s conflict hinges on players using their ships to create favourable trades. The game progresses dynamically as a player uses their hand size, ships in play, and resources generated to create two-for-one or three-for-one situations where they can create a lead. Like in Hearthstone, a card’s strength is relative to its’ ability to trade in to your opponent’s tableau. When both players opt for defensive and greed-oriented strategies, the tension drops, and games get extended. And, since there’s only 16 cards in circulation, the bigger tableaus leave fewer cards moving between players.

So how does a designer try to solve this problem?

I considered a bunch of things. For example, I could set a hard limit to the number of upgrades a player could play consecutively. Or a hard limit for how many could be played in your tableau.

Ultimately, I opted for three changes.

I added a single line to the rule book. To paraphrase: when a player destroys an upgrade, he takes that card in to his hand. Essentially, by being aggressive, you are taking their stuff, and everyone likes taking someone else’s stuff.

I created a set of ships that provide an alternate solution. The Freighter, Vessel, Magnet and Lotus. All offer passive resources, while still allowing you to field an offensive threat. In addition, they offer another layer of complexity as cards played as ships are harder to remove or target than upgrades.

And I buffed the late game. When the tableaus get long, the big guns need to be big enough to end the game, or do enough damage to end it in a turn or two later. By adding a bit of power to the capital ships and game-ending abilities, we can subtly push the game to a faster resolution if two players want to passively build.

Is it enough? Time will tell. When you first get your hands on it, I don’t doubt your first match will take a while. But before long, you’ll be playing in the coffee shop, waiting for the rain to die down, fighting for galactic dominance and control over the asteroid belt with someone who is really eager to knock your station out of the sky.
 Thumb up
  • [+] Dice rolls
Adam Wilk
flag msg tools
Information Overload

Games are a series of interesting decisions. Right. Heard that before, maybe a billion times? But really — how many decisions is too many, or too often? And how do you fit that in to a tiny game of only 18 cards?

It might be hard to parse from just a glance at the rule book and the cards themselves, but Outpost 18 is a game of hand management. Each player starts with an engine drawing one card a turn, and player tableaus can grow to support multiple draws, multiple plays, and a whole whack of other things.

And depending on the gamer, it can be a lot to take in.

How much is to much?

Take a game of Magic: The Gathering. You are dealt seven cards — thats a lot of information, right? Maybe. But on your first turn, how many decisions do you have? Almost the entirety of your hand is unplayable due to mana costs. You choose a land to play, and maybe a one-drop. In the mid game, with three or four lands in play, your decisions open up a bit, but it’s still rare you have more than two playable cards at any time. By the late game, you’ve got a lot of decisions — but hand size keeps you from having that many to track.

In Outpost 18, every card is a decision between ship or upgrade. And there is no cost to playing a card to your board. Player’s start with a minimum of four of these decisions, or six for the second player. Understandably, not all of these decisions are weighted equally — depending on when a card comes up in the course of a match, the choice might be very apparent — but, the raw amount of possibilities is very high for a small card game.

Mental processor requirements

Watching two new players try to play their first game is an enlightening experience. The second player in particular has a lot of processing to do on their first turn — three cards, six possible choices, each with branching impact for the way the game will play out. That decision process for a new player takes time. Maybe too much time.

The +DRAW mechanic has a similar underlying issue. Outpost 18 is about hand management, card advantage, and engine building. Your hand represents your possibilities. Player’s who opt for +DRAW in their early tableau suddenly have four new decisions every turn instead of two. Combined with inexperience, and a new player’s tendency to build slow resource-focused engines (see previous post), and bam, you’ve got a recipe for very long think-times between moves.

Information overload. Analysis paralysis. Brain cramps.

So what’s the balance? It depends the kind of game you are shooting for. With Outpost 18, I wanted a game that plays fast between experienced players. I want a game that plays well after a dozen matches, after a hundred matches, or more. By cutting out decisions early on, or limiting a player’s ability to build a draw engine, the game would undoubtedly be easier for new players. But what about the guy on his fiftieth match?

Analysis-Paralysis informed design decisions

Limit hand size to three.

This has a few levels to it, actually. Three cards represents six potential choices, which is probably beyond a comfortable scope for most, or at least clawing at that upper limit. If the hand size limit was four, for example, we’d be juggling eight choices in a turn.

Additionally, Outpost 18 only has 16 cards in play at any time. Without a small hand size, players could generate enough card advantage to empty out options.

Drawing cards at the end of your turn.

This one is getting more commonplace in card games nowadays, but it’s worth mentioning. By shifting card draw and decision space to the end of your turn, you now have the entirety of your opponent’s turn to finalize your plan. This also has the side effect of speeding the game up, which is great.

Only four draw engine cards.

Of all the design theory tossed around on this blog, this is one thing that may change before publishing. Currently, the +DRAW cards are limited to four upgrades. And the more +DRAW a player generates, the more decision space there is. Since layers draw cards either through their tableau or by destroying bits of their opponent, the game is playable and very winnable without relying on +DRAW tableau pieces.
 Thumb up
  • [+] Dice rolls
Adam Wilk
flag msg tools
State of the Game - September 2019

In this blog, I'm going to update you on game design and balance issues as the game has evolved over the course of the Kickstarter.

A game to play /or/ a game you play

A backer recently told me he'd put Outpost 18 in his top 5 lists of tabletop games, alongside some pretty wonderful designs. That's a pretty amazing accolade for a micro game, but more than that, it's proof of concept. My friend and fellow game-designer, Keith Burgun, uses a phrase I'll steal here: "games you can live in." It's a quick little phrase that summarizes my design philosophy.

The basic difference is this: a game to play is a fun outing, a something to do and move on. A game you play is something more, it's something that you invest time in to and it keeps rewarding you. Something you can come back to and keep getting more out. A game you can live in becomes part of a defining characteristic in someone, and in a wonderful way.

I'm confident that Outpost 18 has the tools to be that kind of game.

How 2.2 changes have aged

Base-breaking and comebacks: 2.2 introduced a few changes. Most notably, Junkrig and Dreadnought were re-developed to provide options to players who did not have cards invested in their tableau.

Over time, Dreadnought has proved to be a really valuable card, and especially for player 2. Dreadnought allows the second player to threaten for more damage than otherwise possible, which changes the game dynamic in a really interesting way. Moreover, with the way destroyed upgrades return to a player's hand, Dreadnought is an excellent finisher.

Junkrig also works on a few levels. It's a wonderful card for finishing off a heavily defensive player, and otherwise fits in nicely.

6 Def upgrades, up from 5:

Not much to say here, other than it seems to fix the issue. Def cards are very reliably drawn when needed, with just enough variance that things can get interesting.

Ship power nerfs:

All of the small +resource ships have been set down a bit to 1 power or less. This was simply to slow down the early game a touch and add some more variety to openings.


This version of Outpost 18 is producing really good games. Between two skilled players, there is enough interesting action and decision making that I am confident I could send the files to print tomorrow and deliver an excellent game.

There is one final issue I am looking at over the next week: player winrates.

Now that we have a fully functional digital game (, we can monitor player winrates over a modest sample size. Player 1 has had a historic advantage, which is beginning to slow down, and it's getting closer to something reasonable.

Currently (65 game sample size on version 2.4), player one wins around 65% of the time.

That said, there are a few things at play here. The game isn't 'figured out', meaning there is a lot of room for experimenting and learning. The sample size is relatively small, so fluctuations happen. A good number of games has some digital bugs that definitely influenced the win.

But, there is an undeniable feeling at high levels of play that player one is in the driver's seat. And, maybe that's OK. But it feels like there needs to be one more tool to get player two in the lead.

Applying Dreadnought learns to Lotus

That's where I'm at now with what I take to be the final step in this design. The Dreadnought redesign taught me that a card using hand size as power helped player 2 and made win rates more evenly distributed. Nothing feels better for player 2 than a turn 1 dreadnought.

Lotus is at a weird spot with ability text right now. There's only a few scenarios where it's ability is useful: paired with dreadnought, or paired with a big labour combo. Lotus is particularly useful for player two because it is played as a ship, therefore it is protected from direct attack most of the time.

Playing lotus also represents a drop in tempo -- it is a 0 power ship, and only provides resources.

The current ideas for an ability rework involve discarding cards in hand for a return in +Play for the turn. That way, a player with more cards (inherently player 2) can turn card advantage in to a tempo advantage, but at a high cost. Player one can still get use out of this card, but only if they are trading cards efficiently.

Still have to brew on it for a while. Thought I'd share the process.

Thanks for reading!
 Thumb up
  • [+] Dice rolls