Recommend
11 
 Thumb up
 Hide
119 Posts
1 , 2 , 3 , 4 , 5  Next »   | 

Abstract Games» Forums » General

Subject: Discussion about two lists of games in BGG's "Abstract Games" Category rss

Your Tags: Add tags
Popular Tags: combinatorial [+] [View All]
Herb
United States
Georgia
flag msg tools
mbmbmbmbmb
I am trying to spin through the 6,000+ games in the BGG database that are in "Abstract Strategy" category and put them on one or the other of two geeklists depending on if or not the game is a 2 person combinatorial abstract game (definition to follow later today in this tread):

Personally I didn't like the fact that games that have dice rolls were list as "abstract strategy." So I'm trying to get a relatively "clean" list of such games.

The following lists are simply too large to be useful other than for a yes/no classification. But that their real purpose to document the yes/no decision. Since the game will be on one list or the other, we have a specific place to discuss if this is an "acceptable" abstract strategy game.

Yes List: [geeklist=173526][/geeklist]

The name of the list is perhaps wrong. It really is to collect games that can be analyzed logically, or in which intellect is the most significant aspect of game play.

For the moment please concentrate on this list.

* Are that any games on the list that shouldn't be?

* Please make comments abut yes/no on the game's entry in the geeklist.

* Please add any games that you feel is a 2 person abstract strategy combinatorial game to the "yes" list.

* You can search the "yes" list which is part of user's 2pCombAbsStrat game collection. The search ability isn't great, but it is far better than trying to wade through 1500 games on a list.

* PLEASE if BGG has a game misclassified, or incompletely classified, according to their own definitions, then submit a correction for the game. I have done many.

No List: BGG "Abstract Strategy" Games That Are Not 2P Combinatorial

The "not" geeklist has gotten exposure faster than I had hoped. I had the "not" list hidden but I couldn't see it within the game's list of geeklists. That made trying to find a game on a large geeklist difficult.

There are 6,000+ games to spin through and I am trying to do it quickly, but not totally recklessly.

Please let me explain what I am trying to do in the "not" list

It would impossible to read through the remaining 4500 games in any reasonably short period of time. So I am looking for existing BGG classifications such as the mechanic "Dice Rolling" which would be inconsistent with the game be a "good" abstract strategy game. So I'm adding such games with to the "no" list and putting the reason as:
/* Dice Rolling*/

The intention is two fold.

First: It leaves fingerprints as to why the name is on the "no" list. So if the exclusion was wrong, please comment on the game's entry on the geeklist and it can be moved.

Second: It is a lot faster to mark the geeklist and move on than it is to read the game's description, and then have to read the rules for 20% of the games.

I wholly acknowledge that putting the game on the no list for "Dice Rolling" is arbitrary. It could be that dice are rolled at the start of a game to randomize them to use as men. The point to me is that I documented what my arbitrary decision was. So you can tell if the "no" classification is wrong.

Then I need to see if any game is on both lists and resolved those problems.

Then I need to look at the games which are not yet on either list.

Remember this is a work in progress...

I certainly need your help and participation.
-- I don't own 99.9% of these games.
-- I have no idea how many of them play.

Please make comments about a particular game's yes/no status in whichever list has it.

Games can certainly be moved from one list to the other. I'll move the comments too to save the discussion.


3 
 Thumb up
1.00
 tip
 Hide
  • [+] Dice rolls
Herb
United States
Georgia
flag msg tools
mbmbmbmbmb
Current definition: post 16170430
-------------
Definition of "2-player Combinatorial abstract strategy games"
... and how I am skewing it a bit to include more games.

Text book definition for a combinatorial game is:
(1) Two players who alternate turns
(2) Both players have perfect information
-- No chance, such as dice or shuffled cards
-- No hidden information, as in Stratego & Magic
(3) The game is finite – it must eventually end
(4) There are no draws or ties
(5) Normal Play: Last to move wins!


(1) Mostly ok. It leaves out Boggle for instance.

(2) The essence of a "strategy" game. (I beat you because I'm smarter, not because I was lucky.)

(3) A lot of games like checkers and chess have rules to prevent infinite repeats of position.

There are positional games where the play loops until one of the players makes a mistake. With perfect play, the game would go on forever.

(4) This is a problem for mathematicians who are analyzing game play and expect a winner. We give each player 1/2 a point and move on...

(5) Again this is a problem for mathematicians. There are also misère games where last player to win loses.

I think we can more or less just ignore this point. For most games it would be true.

I'd add:
The game has no dexterity or "reaction time" aspects.

Yes, I know that chess has a clock and you can wait a split second too long to try to move and lose. Probably don't have quite the right phrase.

This leaves out Jenga style games where you're pulling out a stick and re-balancing. it. Obviously with "perfect information" it would just be a matter of physics.

So what I thinks works for us is:

A 2 player combinatorial abstract strategy game has:
(A) Two players who alternate turns
(B) Both players have perfect information
-- No chance, such as dice or shuffled cards
-- No hidden information, as in Stratego & Magic
(C) Typically the game is finite – it must eventually end. Exceptions are some abstract strategy children's games where infinite repeats are possible.
(D) The game has no dexterity or "reaction time" aspects.


I'm thinking that for (A):

* Some sort of exception should be written in for games
like Boggle. So long as the simultaneous play doesn't involve
players making hidden decisions which impact the play of the
other players.

* Perhaps relax this to just ignore king-maker problem. It
obviously exists. So a "Three man checkers" game would be ok.
There are only a couple of hundred games which 2 players
cannot play. So it is a relatively short list to process.


Your thoughts?
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Herb
United States
Georgia
flag msg tools
mbmbmbmbmb
There is a bit of a "hidden agenda" here. With over a 1000 games on the "good" list, I was hoping that I could talk BGG into have a "Perfect Information" mechanic. The mechanic could then be set as true for the items in that list.

As I pointed out in a suggestion thread, there is a huge problem with adding a mechanic. Initially all games are "false."

You set a handful "true", but 99% of the games aren't classified.
(1) It is misleading since a lot of games will be missed
(2) It creates an impossible workload for admins trying to process the corrections.

There are only a couple of hundred games which are either 1 player games or games with a minimum of 3 or more players. They have the king-maker problem. We could define our way around that by explicitly defining that aspect of the game play would be ignored.
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Ryan James
United States
Davis
California
flag msg tools
mbmbmbmbmb
This is somewhat beside the point, but I think the game closest to an abstract strategy that is disqualified due to dexterity is Polarity. I personally find it a wonderful game and much more based on strategic play than on dexterity.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Herb
United States
Georgia
flag msg tools
mbmbmbmbmb
autoplectic wrote:
This is somewhat beside the point, but I think the game closest to an abstract strategy that is disqualified due to dexterity is Polarity. I personally find it a wonderful game and much more based on strategic play than on dexterity.


Thank for the extra information.

Please put the comment on the Polarity game on the geeklist. So whoever comments there gets to help decide if it should get moved. If different threads are discussing the same game, or similar games, it is unmanageable.

I marked games, somewhat arbitrarily, based simply on the BGG classification. Hopefully 95% are right. I am really interested in views on how the game plays rather than some theoretical abstraction that I surmised.

Jenga type games should theoretically just be pure physics. Pragmatically the game doesn't play that way. A lot of the blocks look alike. They have different bumps and pieces of sand stuck in them, so it isn't just a physics problem. The game plays chaotically.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Jeremy Peet
United States
Milford
Michigan
flag msg tools
designer
badge
Way Out There...
Avatar
mbmbmbmbmb
This is what is currently listed on BGG as what defines an "abstract strategy":

Abstract Strategy games are often (but not always):

theme-less (without storyline)
built on simple and/or straightforward design and mechanics
perfect information games
games that promote one player overtaking their opponent(s)
little to no elements of luck, chance, or random occurrence

A little more open than your own definitions above (mostly because of the first sentence).

An arbitrary categorization without thoroughly knowing how the games actually play may work against you here.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Herb
United States
Georgia
flag msg tools
mbmbmbmbmb
Germflinger wrote:
This is what is currently listed on BGG as what defines an "abstract strategy":

Abstract Strategy games are often (but not always):

theme-less (without storyline)
built on simple and/or straightforward design and mechanics
perfect information games
games that promote one player overtaking their opponent(s)
little to no elements of luck, chance, or random occurrence

A little more open than your own definitions above (mostly because of the first sentence).

An arbitrary categorization without thoroughly knowing how the games actually play may work against you here.


The "theme" aspect is puzzling. I've chosen to ignore it. If BGG has the game listed in the "Abstract Strategy" category, then I'm accepting that it is an "Abstract Strategy" game.

Personally, I think that many games in the "Abstract Strategy" category are heavily themed. I think many of the games in the "Abstract Strategy" category don't fit their own definition. But to argue that point would be beating a dead horse. That is how BGG has them classified.

So readers -- Please submit a correction if you think a game is "too" themed to be abstract strategy. You'll have to argue the point on a case by case basis.

Mainly I'm trying to find the 2-player games with "perfect information."
-- There are 6,000+ games in Abstract strategy
-- 1,500 on list now with perfect information
-- 1,500 that i can cull fairly quickly, a month or so...

That leaves 3000 odd games to still look at!

It seems next to impossible to just scan the themed games and make a decision (with 95% confidence) that the game has perfect information or not. I start reading about how "the bright sun has warmed the princess's day," and my mind goes numb. yuk

So the themed games have to be studied to make a decision and that takes forever. I'm hoping that this pile will not get too large. It will also probably be games that a lot of people know, and will hopefully help out with the characterization.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Jeremy Peet
United States
Milford
Michigan
flag msg tools
designer
badge
Way Out There...
Avatar
mbmbmbmbmb
The search for all the games that fit the "perfect information" category will be a long arduous (valiant) effort.









1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Maurizio De Leo
Singapore
Singapore
flag msg tools
badge
Avatar
mbmbmbmbmb
I appreciate the effort.
I have indeed tried similar things, first by voting "non-perfect information" games out of the Abstract Games Subdomain, then by perusing the tag system (mostly piggybacking on Russ effort) and finally by using the "want" and wishlist-5 (do not buy) options.

However the scale of what you are trying is surely bigger and certainly better planned. I'll try to help you out: I own many of these games (trying to collect them all) and have played quite a few.
4 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Herb
United States
Georgia
flag msg tools
mbmbmbmbmb
megamau wrote:
I appreciate the effort.
I have indeed tried similar things, first by voting "non-perfect information" games out of the Abstract Games Subdomain, then by perusing the tag system (mostly piggybacking on Russ effort) and finally by using the "want" and wishlist-5 (do not buy) options.

However the scale of what you are trying is surely bigger and certainly better planned. I'll try to help you out: I own many of these games (trying to collect them all) and have played quite a few.


I believe that it is necessary to be very public about it to get the additional support needed to make this work.

I don't own, nor have I played, 99.97% of the games being listed.

Without the help of others this project will not succeed.

megamau wrote:
... certainly better planned.


Like all good computer programmers I started whacking it out before I planned carefully enough. Now I need to do some rework.
The problem always is that you don't know what the issues
are until you've really start doing the project. But a couple
of days of more careful planning would have helped. To make
matters worse? I am a PMP certified project manager. I had a
great team of a dozen programmers at IBM. gulp


I think I can use a computer program to parse some download that I have to generate a html document with a link to the game and then a list of "all" the BGG tags that indicate trouble for the "not list". Then I could click on link and swipe the list of BGG tags. So:
-- if there were 3-4 BGG tags that indicate trouble we could all be pretty sure that the game has hidden information.
-- if only 1 tag then the "not" characterization is weaker.

I am not trying to tag games capriciously. Part of the problem is that many of these games have 2-3 "problem" BGG characteristics. I was trying to multiple select parameters in the BGG Advanced Search so as not to get a lot of second hits on a game.

I am trying to add a game and tag it as fast as I can. Trying to check to see if game is already on the list (by looking for geeklist in game) slows thing down considerably. If the game is already on the geeklist and I add a second copy with a different tag, then cleaning that up takes extra time.

With all the clicks and waiting for BGG and my browser (I have slow DSL connection) it takes at least 1 minute per game.

I wonder how much bandwidth having BGG "adblock" stops? I could also disable picture gallery and cut down on some of the graphics there too.

Not sure if Firefox add in "Adblock +" stops graphics from coming down line, or just dumps them. The visual annoyance is stopped either way. Not sure what this does for bandwidth though.

I'm too cheap to have cable TV installed and get faster access.
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Randall Bart
United States
Winnetka
California
flag msg tools
designer
Baseball been bery bery good to me
badge
This is a picture of a published game designer
Avatar
mbmbmbmbmb
It would be easier to understand your lists if rather than using "combinatorial" to mean "perfect information" you just said "perfect information".
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Russ Williams
Poland
Wrocław
Dolny Śląsk
flag msg tools
designer
badge
Avatar
mbmbmbmbmb
Barticus88 wrote:
It would be easier to understand your lists if rather than using "combinatorial" to mean "perfect information" you just said "perfect information".

Hmm, FWIW I recall there being less consensus on what "perfect information" means than on what "combinatorial" means. (Specifically, I think there was some long past debate about whether backgammon and other games with randomness but no hidden info are "perfect information" or not, and different math papers about game theory used the term differently, or something.)
https://en.wikipedia.org/wiki/Perfect_information
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Randall Bart
United States
Winnetka
California
flag msg tools
designer
Baseball been bery bery good to me
badge
This is a picture of a published game designer
Avatar
mbmbmbmbmb
russ wrote:

Well using "perfect information" to describe a game where you don't know something is just asinine. But regardless, to me Bridge is a combinatorial game. You deal out all the cards and all of them get played, so it is a perfect combinatorial game. There is a lot of hidden information along the way, but that has nothing to do with being combinatorial. OTOH, anything with dice is not combinatorial, because you are drawing from an infinite supply of die rolls.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Herb
United States
Georgia
flag msg tools
mbmbmbmbmb
Barticus88 wrote:
It would be easier to understand your lists if rather than using "combinatorial" to mean "perfect information" you just said "perfect information".


Thanks for the heads up.

Sorry, I don't mean to confused the two. I really want to separate out the "2 player combinatorial abstract strategy" games.

I asked other geeks to put any games that didn't have "perfect information" on the BGG "Abstract Strategy" Games That Are Not 2P Combinatorial geeklist because perfect information is one of the requirements, and there are a lot such of games that yet are not on the list. Statistically, it will eliminate far many more games than anything else.

Obviously all the "2 player combinatorial abstract strategy" games will have perfect information - when played with 2 players. Thus most of the games in BGG's "Abstract Strategy" category that have perfect information will be "2 player combinatorial abstract strategy" games.




Part of the problem here is terminology. We're jumping into the middle of game theory without defining the whole classification tree.

It seems that the terms "Perfect Information" and "Imperfect Information" are best used to describe combinatorial games. There is a whole laundry list of game terms used for classification.

Quote:
Know the difference between a theorist and a terrorist?
Spoiler (click to reveal)
You can negotiate with a terrorist!





Boggle is one game that bothers me. A strict adherence eliminates Boogle because it surely isn't "combinatorial." But both players are essentially playing sequentially and independently against the oracle who knows the complete list of words. They don't interact, one player's list doesn't effect the other.

Also Boggle certainly seems to have hidden information. Each player certainly hides his list from the other. First, let's define one shaking of the cube and its play to create a list of words as a round. Yes players score points for a round. But one round's play doesn't effect any other. So although each player hides his list of words from the other(s), the hidden information revealed doesn't impact play in that round, or in any other subsequent round.

It seems like the "definitions" need to be tweaked somehow to allow Boggle in. So a game theorist who is studying games wouldn't strictly classify Boggle as "2 player combinatorial abstract strategy" game, but it plays that way.

In fact this is the only type of play that I can think of that allows multiple players yet does not suffer from the king maker rule!!! Yes multiple players could combine their lists, but this would be obvious cheating.

As with any game, a strong player could play against a team of other players as a handicap. For example, one parent against two children. But playing with such a handicap doesn't imply that the children would be cheating.



 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Russ Williams
Poland
Wrocław
Dolny Śląsk
flag msg tools
designer
badge
Avatar
mbmbmbmbmb
Barticus88 wrote:
russ wrote:

Well using "perfect information" to describe a game where you don't know something is just asinine.


The reason quoted in that article for backgammon being "perfect information" is that the proposed definition is that you know perfectly everything which has happened so far (including all past die rolls). You don't know future die rolls, but - so what? You don't know many things about the future, including player decisions.

(That said, I agree that many people use "perfect information" to exclude randomness - which is why I think the term is problematic. It means significantly different things to different people.)

Barticus88 wrote:
But regardless, to me Bridge is a combinatorial game.

How in the world is Bridge combinatorial? That flies in the face of any definition I've seen.
https://en.wikipedia.org/wiki/Combinatorial_game
wikipedia on combinatorial games wrote:
CGT does not study games with imperfect or incomplete information (sometimes called games of chance, like poker). It restricts itself to games whose position is public to both players, and in which the set of available moves is also public (perfect information).


Barticus88 wrote:
You deal out all the cards and all of them get played, so it is a perfect combinatorial game. There is a lot of hidden information along the way, but that has nothing to do with being combinatorial.

Yes it does, according to every definition I've seen...

Barticus88 wrote:
OTOH, anything with dice is not combinatorial, because you are drawing from an infinite supply of die rolls.

Not sure what "infinite supply" has to do with it. Nothing about the definition of "combinatorial game" requires a finite number of possible moves or positions. Winning Ways has examples of infinite combinatorial games.

If you're not bothered by the hidden info of Bridge, then Backgammon can be viewed as combinatorial in the same way you view Bridge as combinatorial by simply rolling all the die rolls in advance. (A million rolls should suffice, for example, if you want to quibble about it not being clear how many must be rolled.) Then it's "merely" hidden information, just like the cards in Bridge are hidden.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Geir Erik Ø
Norway
Sandnes
flag msg tools
Geeks!
badge
Hello!
Avatar
mbmbmbmbmb
From when do the all information need to be shown? In Navia Dratp and Shuuro you make "armies" hidden, but once placed on the board, all the information is given.

In Gobblet all the information is given, but is hidden during the game, so you can forget information. Is this a valid game?
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Herb
United States
Georgia
flag msg tools
mbmbmbmbmb
Backgammon & randomness vs perfect information

Ok, I think we can get into trouble on this one.

First I think we just have to agree that we're going to hammer out our definitions, without resorting to vague and unknown authority.

To me backgammon does not have "perfect information" because it has randomness. In reality the game isn't deterministic. But if a game doesn't have any randomness then it is surely deterministic.

I rationalize perfect/imperfect information this way. I assume that there is a malevolent oracle who controls the randomness in the play. Thus he can, at very few appropriate moments, thwart the better player by selecting a bad option for the better player. He can also help the poor player at a very few appropriate moments. Thus the better player almost never wins against the poorer player who he should defeat easily.

This cuts at the heart of the notion that knowledge is power.

The problem here is that I (knowingly) have tried to jump into the middle of a whole classification structure without having a whole hierarchy to use to sort the games.




Can we just agree that "BGG perfect information" means no randomness in play and move on, is this a show stopper for this "project" that needs to be thrashed out?

Would a poll be helpful with majority rules?

 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Russ Williams
Poland
Wrocław
Dolny Śląsk
flag msg tools
designer
badge
Avatar
mbmbmbmbmb
herace wrote:
To me backgammon does not have "perfect information" because it has randomness.

OK, that's what it means to you, but as far as I can tell, established definitions of "perfect information" in mathematical game theory don't exclude randomness, and indeed allow it, as in the case of Backgammon.


Quote:
In reality the game isn't deterministic. But if a game doesn't have any randomness then it is surely deterministic.


Possible tangent: This raises another controversial issue, namely games with simultaneous decisions. (E.g. Rock/paper/scissors, to take the classic simple example.) Some people say they have randomness, some say they don't, some say they are deterministic, some say they are not.



Quote:
Can we just agree that "BGG perfect information" means no randomness in play and move on, is this a show stopper for this "project" that needs to be thrashed out?

I'm not sure I see the value in making a BGG-specific definition which is knowingly intentionally different from an established mathematical definition.

It sounds a bit like saying "Can we just agree that "BGG positive" means "greater than or equal to 0" instead of just "greater than 0"?" or "Can we just agree that "BGG prime" means "odd"?"

I mean, sure, we could do that, but it's probably not very wise to...

It seems more valuable/useful to try to use existing game theoretical terminology correctly/consistently instead of fragmenting into a non-standard BGG-specific dialect.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Herb
United States
Georgia
flag msg tools
mbmbmbmbmb
Geirerik wrote:
From when do the all information need to be shown? In Navia Dratp and Shuuro you make "armies" hidden, but once placed on the board, all the information is given.

In Gobblet all the information is given, but is hidden during the game, so you can forget information. Is this a valid game?


My notion was to allow an variable setup or a random setup setup, but no randomness or hidden information once plays starts.

So if there are 10 cards in deck and you get 5 and I get 5, then I have "perfect information."

But let's think about the two of us playing poker. Using a bizarre definition I could view that I know what all the cards could you have and all of the cards that the deck could have. Thus I could calculate the odds exactly. In actually humans aren't that good, and the whole game is based on the notion that I know the odds better than you. Thus it is more appropriate for me to view your cards as "hidden information."

In reality this is a "fuzzy" measurement. When does a game require too many combinatorial possibilities for a game to be considered as having "hidden information"?

I glanced at Navia Dratp. I surmised that you have some number of pieces which you play on the board initially and then you can chose from the remaining pieces which you want to bring into play.

Since the game is deterministic (my perception..), I'd say that it belongs on the "good" list.

I go back to my notion of a malevolent oracle who can control anything in the game that the players don't. In poker he could stack the deck and give the poorer player much better odds of winning. AS i understood Navia Dratp, a malevolent oracle couldn't influence the game.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Russ Williams
Poland
Wrocław
Dolny Śląsk
flag msg tools
designer
badge
Avatar
mbmbmbmbmb
Geirerik wrote:
From when do the all information need to be shown? In Navia Dratp and Shuuro you make "armies" hidden, but once placed on the board, all the information is given.


Do you mean that both players simultaneously decide on the composition of their armies (and then presumably reveal them simultaneously)?

If so, this is (to me) clearly different from a random setup, since it's simultaneous player decisions instead of an impartial random setup.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Randall Bart
United States
Winnetka
California
flag msg tools
designer
Baseball been bery bery good to me
badge
This is a picture of a published game designer
Avatar
mbmbmbmbmb
russ wrote:
The reason quoted in that article for backgammon being "perfect information" is that the proposed definition is that you know perfectly everything which has happened so far (including all past die rolls). You don't know future die rolls, but - so what? You don't know many things about the future, including player decisions.

By this definition, rock-paper-scissors is perfect information, because you know nothing and there is nothing to know. Taken to its extreme every game is perfect information, because you know everything except what will be revealed in the future.

Quote:
How in the world is Bridge combinatorial? That flies in the face of any definition I've seen.
https://en.wikipedia.org/wiki/Combinatorial_game

Well now I know where this bizarre use of "combinatorial" comes from.

1) That is actually a link to an article on combinatorial game theory, not combinatorial game.

2) I never heard of combinatorial game theory, but it is obviously the fusion of game theory and combinatorial analysis.

3) So when you say "combinatorial game", you don't mean the game itself is combinatorial, you mean that the game can be subject to combinatorial analysis.

Is there somewhere you said "combinatorial game theory" or "combinatorial analysis" that I could have picked up the context for using "combinatorial" here? I would call these "combinatorial analysis games" or "combinatorial game theory games" or "CGT games" for short. I wouldn't call them combinatorial games. By common use of "combinatorial", Clue is very much a combinatorial game. (Of course some people may not think "combinatorial" has a common use.)
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Herb
United States
Georgia
flag msg tools
mbmbmbmbmb
russ wrote:
Quote:
Can we just agree that "BGG perfect information" means no randomness in play and move on, is this a show stopper for this "project" that needs to be thrashed out?

I'm not sure I see the value in making a BGG-specific definition which is knowingly intentionally different from an established mathematical definition.

It sounds a bit like saying "Can we just agree that "BGG positive" means "greater than or equal to 0" instead of just "greater than 0"?" or "Can we just agree that "BGG prime" means "odd"?"

I mean, sure, we could do that, but it's probably not very wise to...

It seems more valuable/useful to try to use existing game theoretical terminology correctly/consistently instead of fragmenting into a non-standard BGG-specific dialect.


I agree that I'm trying to use the existing terminology in a bit of an unorthodox manner.

I'm trying to draw a line at only having games that have no randomness in play because I thought we could all agree on those that do. It seems to me that the difference between randomness in poker and randomness in backgammon is a matter of degree. They both certainly have a randomness factor. So yes/no on randomness seems better than trying to decide where between 0 and 1 is the randomness clip level.

I'm trying to choose an interesting enough name and concept that folks want to participate. My perception is that if I went strictly by how Berlekamp/Conway/Guy defined a 2-player combinatorial strategy game that there would be a very, very limited number of games in the database. No ties for instance.

I am truly open to suggestions as to how to rename/restructure this. Using my notion to exclude games where a malevolent oracle could influence the game, what would you call the remaining "combinatorial" games?

-- I'm using combinatorial only to mean that players play sequentially. But it bothers me that this excludes Boggle from the list for instance.




What I'd like to have is the ability to set characteristics and define each one for each of the 6,000 games in the abstract strategy category.

So is setup:
-- fixed
-- variable (controlled by players, possibility better player gets advantage in setup, could always randomize by making variant of "real" game)
-- random eg roll dice to use as men

Is play:
-- deterministic
-- nondetermnistic

Is the play combinatorial?
-- yes
-- no

-- Subquestion Is the game Impartial?
--------Yes
--------No


But at this point to sort out to sort out those possibilities with geeklists is unmanagable.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Herb
United States
Georgia
flag msg tools
mbmbmbmbmb
Redefining a "Two Player Combinatorial Abstract Strategy Game"

A 2 player combinatorial abstract strategy game:

(A) Is in the "Abstract Game" category in BGG.
(B) Two players who alternate turns
(C) Both players have perfect information(Note 1)
-- No chance, such as dice or shuffled cards
-- No hidden information, as in Stratego & Magic
(D) Typically the game is finite – it must eventually end. Exceptions are some abstract strategy children's games where infinite repeats are possible.
(E) The game has no dexterity or "reaction time" aspects.


I'm thinking that for (C):

* Some sort of exception should be written in for games
like Boggle. So long as the simultaneous play doesn't involve
players making hidden decisions which impact the play of the
other players.

* Perhaps relax this to just ignore king-maker problem. It
obviously exists. So a "Three man checkers" game would be ok.
There are only a couple of hundred games which 2 players
cannot play. So it is a relatively short list to process.


Note 1. The assumption is that any nondeterministic part (something that the players do not control) of the game could be controlled by a malevolent oracle who could influence game towards player 1 or player 2. For example (a) dice rolls in backgammon, (b) order of cards in a deck, or (c) which tiles a player would draw from a bag in a (supposedly) blind draw.

Your thoughts?



edit...

I'm changing name from:
FROM: Two Player Combinatorial Abstract Strategy Games
TO: Two Player Combinatorial "Abstract Strategy" Games

I think quotes around Abstract Strategy is something that we all can understand and agree to.

 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Geir Erik Ø
Norway
Sandnes
flag msg tools
Geeks!
badge
Hello!
Avatar
mbmbmbmbmb
russ wrote:
Do you mean that both players simultaneously decide on the composition of their armies (and then presumably reveal them simultaneously)?

If so, this is (to me) clearly different from a random setup, since it's simultaneous player decisions instead of an impartial random setup.


In both games, if I remember correctly, you choose your pieces before the game starts. So you take some choices before you take the first turn. You don´t know your opponents choices until you start setting up the pieces. So it is clearly not random setup.

But on the other hand when the game start all the information is open to both players. This is why I want them on the yes list.

The game can be uneven because of the army selection, but I can´t see that both sides have to be at equal strenght. In most games if you only play one game, the starting player will have an edge. This will also be desiced by a choice or a random element like a coin toss.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Herb
United States
Georgia
flag msg tools
mbmbmbmbmb
? rename Geeklist to Two Player Combinatorial and Deterministic "Abstract Strategy" Games

Move "randomness" like dice to Deterministic?

-- I think current geeklist of games, as is, would be ok with this change.

"Abstract Strategy" simply meaning that game is in BGG "Abstract Strategy" category. This of course has little to do if the game is "abstract."

(1) Still have defined away ties that are not strictly allowed as Berlekamp/Guy/Conway define a "combinatorial" game.

(2) Defined away variability/randomness in setup. Not sure where this falls in grand scheme of game theory.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
1 , 2 , 3 , 4 , 5  Next »   | 
Front Page | Welcome | Contact | Privacy Policy | Terms of Service | Advertise | Support BGG | Feeds RSS
Geekdo, BoardGameGeek, the Geekdo logo, and the BoardGameGeek logo are trademarks of BoardGameGeek, LLC.