Earlier today I had an idea for a meta-game a group of people could play. It’d be ideal if you lived in an intentional community, or were at university with a games society, or somewhere with regular Less Wrong Meetups.

Each time you would find a new game. Each of you would then study the rules for half an hour and strategise, and then you’d play it, once. Afterwards, compare thoughts on strategies and meta-strategies. If you haven’t played Imperialism, try that. If you’ve never tried out Martin Gardner’s games, try them. If you’ve never played Phutball, give it a go.

It should help teach us to understand new situations quickly, look for workable exploits, accurately model other people, and compute Nash equilibrium. Obviously, be careful not to end up just spending your life playing games; the aim isn't to become good at playing games, it's to become good at learning to play games - hopefully including the great game of life.

However, it’s important that no-one in the group know the rules before hand, which makes finding the new games a little harder. On the plus side, it doesn’t matter that the games are well-balanced: if the world is mad, we should be looking for exploits in real life.

It could be really helpful if people who knew of good games to play gave suggestions. A name, possibly some formal specifications (number of players, average time of a game), and some way of accessing the rules. If you only have the rules in a text-file, rot13 them please, and likewise for any discussion of strategy.

74 comments, sorted by Click to highlight new comments since: Today at 10:07 AM
New Comment
Some comments are truncated due to high volume. (⌘F to expand all)Change truncation settings

I'm not sure your argument of the benefits of this actually works.

Taking myself as an example: I play a lot of card and board games; I am very good at learning new games; I have been that annoying person who picks up a game after a quick five minute explanation and beats everyone else who's played it multiple times. But I don't feel like it's given me any extra insight into situations beyond board games. It could be that the problem is too much compartmentalisation, but my instincts tell me that the problem is that board games are just too tidy compared to real life. In real life there are just too many variables and the rules are often implicit or outright wrong, and without understanding how all the variables interact I have trouble working out how to improve my position.

I will say, however, that your plan sounds like a lot of fun. Just not necessarily that useful

I concur. As a member of my school's board game club, I have learned over two dozen games within the past year, and am frequently competitive during the first game. The core of most games is fundamentally the same: You have a position; your actions bring the game to a different position. Once you understand what moves you can make and begin to understand the benefits of various positions, you can begin to strategize using a human version of the simple Minimax algorithm. There are other reasons why having an experienced player teach the game can be better for meta-strategizing skills: They will give examples of correct play and are more likely to enforce the rules correctly, enabling faster learning. Additionally, a quite common scenario is to have a game with someone who's played three times, someone who's played once, and two newcomers (or thereabouts). In that case, it isn't hard for the first-timers to be competitive.
Are you trying to argue against generalizability here? This sounds like it could describe any situation in life (except that usually you don't want to use minimax).
I think the difference between board games and life situations is how much more tractable games are to this attack. You have excellent information on how the state can evolve (perfect information, once you know the "event card" deck or whatnot if there is one), and can quickly get a rough estimate of the probabilities of each. While even a DnD beginner can figure out what actions to take to maximize his expected damage output, his actual character, who has seconds to do what the player has an eternity to think about, could not. And many games have limited interaction between players, in which case strategizing is quite simple. Come to board game club sometime, and I'll show you games where you find yourself able to think significantly ahead (in a quite similar style to how you calculate ahead in chess) within a few turns of playing. Except many of them have low-enough long term consequences for most moves that you often only need to calculate ahead by one.
This is generally true of chess as well. Thinking ahead is definitely useful, but just caching a whole bunch of tactics, strategies, mating themes, and opening moves is enough to make you an advanced amateur without ever having to do much actual calculation.

I used to annoy my little sister by reading up on simple games with mathematically perfect strategies, then beating her every single time. Now she refuses to play with me. (E.g. Nim, and variations thereof. If you don't know how to play Nim, it's actually a pretty good example for this thread. The perfect strategy requires a certain amount of mental arithmetic and is non-obvious, so if you don't know it the game is pretty playable.)

Edit: In retrospect, I'm a jerk.

I've always wanted to be a Jernau Morat Gurgeh.

By the way, isn't Go a good example of a game that is hard to improvise and do well in? Just a reminder that domain knowledge usually trumps all else.

Diplomacy. Best played with exactly seven players. Playing time 4-12 hours face-to-face; can be played by email over the internet.

Obligatory warning: Playing this with friends is not advised unless you and your friends are very thick-skinned. I have never encountered another game that is as likely to destroy friendships.

1Paul Crowley12y
When much younger, I made lots of friends playing Diplomacy, so YMMV!
I've found Dokapon Kingdom [http://www.atlus.com/dokapon/] ruins more friendships. Course, the box actually has a warning about it, so I guess you can't claim ignorance. Also, this isn't a game suggestion. The game is too much based on luck, and balances often for the person in last place.
It's all in the game, yo.

You don't need a group of people to be in on this with you -- just find a local group of board gamers, and hang out with them every once in a while. I guarantee they will have so many games that you will be playing new ones all the time.

http://www.boardgamegeek.com/ is where it's at, I gather.

This is a bit different because typically, most of the people playing any given game have played it before, with only a few newcomers who learn the game as they play it. It wouldn't be a good way to play the meta-game, because you shouldn't expect to be competitive against someone who has played the game for a while. It shouldn't matter as far as "learning to learn games goes", ideally. However, I expect it will, because you're not competing against other people learning to learn games. If you want to win -- and most people like to win -- the incentive here is to keep playing this one game until you're good at it.
Try it before you knock it -- I bet that if you're the kind of person posting on Less Wrong, if you have someone who explains the rules well, and if you're really spending all your mental energy focusing on figuring the game out, you'll stand a good chance of winning a lot of games on the first go. That's my experience so far. That said, it might be even more fun playing against other people who are all new.
I strongly agree with this. In general, most people don't work to optimise their strategy, and a newbie who's used to thinking about long-term or meta-strategies is highly likely to win or come very close on their first attempt.
Depends on the game. I actually have a success story at this sort of thing: I went to a board game convention this summer where I learned a game called Galaxy: the Dark Ages at a demo. That same week, I actually made it to the final round of the Galaxy tournament (and may have done even better if I hadn't made a rather terrible careless mistake). I think the reason this happened is because a lot of the complexity of Galaxy is just due to interactions between players: the rules themselves are very simple. Diplomacy may also be like that (I know the rules, but I've never played, so I don't know what it's like in practice). You can use your existing knowledge of how Other games -- Dominion comes to mind -- would be harder to pick up in half an hour because
Probably not. People don't enjoy being thrashed and the situation can become socially awkward. Leveling the playing field somewhat by having players with existing competence makes things more challenging and is less likely to provoke people into just having sour grapes and refusing to play that game again.
I did something like this back in the 90s, with a guy in a local gaming club who had a big collection of boardgames, played a different one each time, and won about 50% of the time. It's important to bear in mind the typical boardgame does not have the depth of chess or Go -- those are classics for a reason -- and most of them are actually pretty much ideal for this sort of rapid pickup.
You have that right. There is nothing like the combination of having a lot of fun and winning. Oh, wait, different 'pickup'. Nevermind. ;)

No one mentioned Dominion yet? It seems to crowd out Magic Playing groups, which I take as a quality sign. Each game is very different from the other, but so far the same guy of my playing group consistently wins.

Dominion is a good choice for the meta-game thread because it turns out that there are definite meta-strategies that work. Once those are mastered, the game is still kind of fun but loses most of the addictiveness (well, for me anyway). I would argue that Magic has more actual strategic depth, largely as a result of its complexity and giant card base.

Chinese chess, surprisingly rare in many Western countries, but has very in depth strategy.

Settlers of Catan (the original. Most of the expansions just make things more complicated)- warning is a bit addictive.

Sprouts- an amusing game due to Conway.

Illuminati- A very amusing game by Steve Jackson. Can take a bit long.

Ninja Burger- another fun Steve Jackson game.

Note that some other Steve Jackson games aren't necessarily worth it (for example Munchkin is a lot of fun but is deliberately unbalanced and so isn't going to work well for the purposes discussed... (read more)

As far as Chinese chess goes, a good amount of skill in Western chess seems to carry over. In particular, the ability to pay attention to the whole board at the same time and remember what's going on where. There's probably some heavy opening theory that I never learned, but that shouldn't matter unless you're playing against expert players.
[-][anonymous]12y 5

Is it time? Will it work?

Is the world really ready for The Nomic Conspiracy?

Do you mean the Ludic Conspiracy? Though it seems to me that that would at the very least heavily overlap with the Competitive Conspiracy...
Is there a point in distinguishing? If so, it is yours - but I don't know which game assigns it! Admittedly I really like the sound of The Ludic Conspiracy. I just don't know if I can know if I can play.
Yes, I was uncertain of that.
Incidentally, the game is still going on.
Hah! I don't remember, can it only be played when at Burning Man? If so, no amendments can be made until next year, which is a sure bet for some hilarity when the time comes around. If not, let me know if a vote is passed extending it to online play - I'd enjoy influencing the odds that I will win!
In the Beginning, Omega created Nomic...

Mao: A card game with similar principles to Nomic.


Discussing the rules of Mao! *distributes cards*
Saying the M-word when inappropriate, 2-card penalty.
I think we learned different variants.
I'd have been shocked if we didn't.
Yes, that particular game seems to have a very high mutation rate. It almost seems to be designed to be an example of a highly mutable meme. (Incidentally, if one wants to play a game explicitly about figuring out the rules that isn't silly like Mao I'd recommend Zendo.)
In-game chatter! Penalties all around. Not that we ever actually played that variant, too obnoxious
This is precisely what I was going to suggest; I had a very nice game of it just last night. Indeed (says AstroCJ, going on to discuss strategy, but not rules of the game), I think... hmm. This might actually be worth a top-level post. Since I'm going to dispense with all pretence of obeying "the rules", I'll rot13 the rest of this post. We never played the no talking variant either. Fb, yrg hf fcbvy gur zlfgrel nf dhvpxyl nf cbffvoyr: Znb vf n tnzr va juvpu gurer vf na rnfl (vfu) zrpunavfz sbe rnpu crefba gb zbqvsl gur ehyrf bs gur tnzr. V'yy qrfpevor n fnzcyr tnzr dhvpxyl. Cernzoyr/Ehyrf: Gurer ner n ahzore bs cynlref naq gurl rnpu gnxr n ahzore bs pneqf. Gurl gura cynl gurfr pneqf nppbeqvat gb n onfr frg bs ehyrf; zl snibherq inevnag vf bar jurer gur bayl onfr ehyrf ner "sbyybj ahzore be fhvg", ohg zl tebhc bs sevraqf abeznyyl cersref n fyvtugyl zber pbzcyvpngrq frg. Oernxvat n ehyr vaphef n zvabe cranygl, fhpu nf qenjvat n pneq. Jura lbh trg evq bs lbhe pneqf, lbh ner tvira bar-fubg cbjre gb zbqrengryl nygre gur tnzr, naq znl nqq, zbqvsl be erzbir n ehyr. Lbh gura erwbva gur tnzr naq ortva rasbepvat vg. Ab-bar vf gbyq jung guvf ehyr vf; rkcynvavat nal bs gur ehyrf vf ntnvafg gur fcvevg bs gur tnzr naq vaphef n zvabe cranygl. Qvfphffvba: Znb, nf qrfpevorq, qbrf abg fbhaq yvxr vg zhfg or sha. Guvf vf ragveryl gehr. V'yy nggrzcg gb tvir n yvfg bs gur ernfbaf oruvaq cynlvat gur tnzr: Bar rawblf fbyivat chmmyrf. Bar rawblf frggvat chmmyrf. Bar rawblf tvivat uvagf gb chmmyrf. Bar rawblf frrvat bar'f sevraqf ynl gur Dhrra bs Urnegf naq fdhrny, "va gur znaare bs n qehax cvt", na vpr pernz synibhe bs gurve pubvpr[1]. Fb, gurfr frrz yvxr npprcgnoyr guvatf gb rawbl juvpu ner pregnvayl cerfrag va Znb. Fb, jul nz V jevgvat guvf sbe yrffjebat va fhpu qrgnvy? Jryy, V guvax gung vg sbeprf bar gb: Haqrefgnaq jura lbhe sevraqf ner abg univat sha (v.r. lbh'ir abg qbar bar bs gur sbyybjvat, naq guvatf unir tbar jebat.) Abg sbez ulcbgurfvf gbb dhvpxyl naq grfg ntnvafg gur
Woah, you're kidding, right? I've never not played with a "no talking outside of Point of Order" rule. How would it even work if you can say things all the time, and thereby avoid learning when you're actually required to say them? Oh, and I get a card for saying P of O during P of O, and another for talking about the rules.
It's usually clear when one says something intended to interact with the mechanics of the game (e.g. saying "That's the badger" on the Two of Clubs). End P of O. grin
Teehee .... "End P of O" is just an abbreviation for the code to actually end a p. of o. (card to self for talking about the rules). Is it now? There are plenty of times when you might be tempted to say "Ouch!" or "Epic fail," but when the rule of Belittlement is in effect...
The thing that bugs me about Mao... I don't trust the other people I'm playing with to enforce all the rules properly, particularly when their are multiple rules to be processed for one card. A sane execution of Mao would involve a non-player taking the role of arbiter. Not to enforce rules but be available for appeal and reference. But I prefer Bartok regardless. After you play that for 30 rounds or so things get seriously complicated.
In Communist China, rules break you!
All your obsolete memes are belong to us!

I don't know about board games, but after 15 years of playing video games with varying levels of intensity/time spent, I've gotten a lot better at quickly picking up new ones. I assume others have the same experience.

Me too. I assume it has something to do with reaction times and hand-eye coordination, or learning certain puzzle-solving skills. Video games in the same genre tend to be very similar anyway, more so than board games. Racing games in particular seem to be almost identical in the gameplay, additional weapons or abilities notwithstanding.

If you lack human opponents, you can try Toss, but it is currently going through important stages of AI development so you might want to wait till a future release (or try out the svn version, visit our mailing list, etc.)

Middleman, by Eric Solomon

I have played this game (badly, when I was younger) - I recommend it. At least one copy of the rules has been posted online at the time of this comment.

I've wondered whether a computer program which could deduce competent strategy and tactics from game rules would be a good step toward AI.

I don't have a link handy, but apparently there was a pretty successful version of this made for US military war games, that managed to beat the generals at least once.
Ships that are nothing but a gun that floats rule, apparently. :)
How would you get the program to deduce strategy and tactics from the rules in the first place? ;)
It could be done. Endgame positions (which will probably result in a win in three moves or less) can easily be analysed completely using basic game theory - I could probably write you a program that plays tic-tac-toe, or Nine Man's Morris. More complex positions are the real problem, but a lot of human-playable games have basic strategies like "try to keep as many of your pieces on the board as you can (Chess, drafts, Go, and so on)", "try to get your pieces to form rows (Connect-4, Nine Man's Morris)", or even "bet higher than you should to bluff your opponent (Poker and gambling games)". Perhaps it could learn from past mistakes too. Not saying it would be easy to make, that it could beat a human at an arbitrary game, or even that it could beat a specialised AI designed for that particular task. But making AIs adaptable makes them much more human-like, if that is what you want. PS: Toss [http://toss.sourceforge.net] might actually be able to do this. The description isn't very clear; I'll check for you.

I don't really play the type of games getting mentioned here. But I'd be interested in seeing what four LW regulars would do with Bridge. Much of the game consists of conventions of play and well known strategies (and nearly everyone learns these strategies and conventions while learning the rules). I'd be interested to see how much would be duplicated and what alternative conventions new but very intelligent players would come up with.

Radical experimentation is simply illegal in tournament Bridge. It was an interesting game a few decades ago when artificial systems were nominally legal, but resulted in judicial harassment. Then you had to balance the in-game advantage against the human factor. (ETA: by "interesting" I did not mean "fun" but "relevant to the post, ie, to exploiting rules in real life") ETA: my favorite bridge story is that HPF Swinnerton-Dyer [https://en.wikipedia.org/wiki/Peter_Swinnerton-Dyer] was once allowed to bid 8 clubs, a loophole that was quickly closed. My only source is a blog comment [https://quomodocumque.wordpress.com/2010/07/11/two-now-irrelevant-ethical-questions-in-sports/#comment-4762]
You would not believe how nauseating the thought of playing a game like that is. It's taking a game then removing all the parts of it that are fun. I can see why you would use those sort of rules if you were a bridge club full of old women verging on senility. They could maintain a high level of bridge performance based on heavily crystalised knowledge even though their ability to handle (and remember) novel information is shot to pieces. But why have a tournament at all players are basically primitive bots? I think I'll pass on that and go play some basketball instead. And use a full court press! ;)
Do you think gender is relevant for wanting to depend on heavily crystalized knowledge?
That doesn't follow from my comment. I do believe that gender is relevant to the makeup of strereotypical bridge clubs. For reasons including but not limited to the selection effect of mere survival. An observation that is implied in my statement that you may object is that on average women, particularly old women, are more predisposed to control via social convention than males of a similar age. The nature of human social dynamics is such that it rewards different social strategies differently.
Seconding this: When I worked at the nursing home, men made up about 15% of our long-term resident population. Also, I expect older men to be less comfortable in situations that are dominated by women than younger men are, due to cultural differences 60+ years ago.
My understanding of the game is that the fun part is trying to determine whether you have the points and suits between you to go for a minor or major slam, and then finessing those extra two or three tricks you need to actually get it. The rest is just window dressing to build tension.
So "A bid of X means I think X is a good bid" is right out, then? :P
You are allowed to use nonstandard conventions, as long as you tell your opponents what conventions you are using.
conventions, yes, "artificial systems," no.
What do you mean by "artificial systems"?
lmgtfy [https://en.wikipedia.org/wiki/Bidding_system#Classification]
Thank you for providing a resource you endorse as explaining the technical term you introduced, in a public manner so that all readers can benifet from you efforts. For future reference, it will facilitate civil discussion to not combine this action with a suggestion to use a different approach which does not provide all these benifets.
I don't understand bridge, but I'm sure there's a limit to what would be tolerated. Example: take an existing convention, and use a format-preserving encryption on it; the FPE should be simple enough to execute with a second or two of thought, and needn't be cryptographically secure, but would almost certainly be beyond anyone's capacity to cryptanalyze in real time without computational assistance. To extend it further: just use a convention as a n-ary encoding of your hand, and then encrypt it. How specifically illegal is this? Even it all information about the convention is given to your opponents, it does them essentially no good---probably even if they have the key, just because it would confuse them so thoroughly.
You would be expected to explain your conventions to your oppenents in the same way you would explain them to a new partner. The conventions themselves may be complicated, but the explanation should not add complication. Conventions have the purpose of making in-game actions to reveal information about your hand (to your partner, though your oppenents are allowed to "listen in").
But a convention that requires multiplying points along a discrete elliptical curve, for example, could be done with sufficient practice and initial aptitude, but even the most thorough description wouldn't really help any but the most amazingly gifted if opponents don't get more than a few hours to practice it. I've actually tried part of this with RSA, and I think ECC would be even easier implement, if harder to learn the underlying group. In other words, public key cryptography is completely possible to do in one's head, given practice and assuming small key sizes (I used 16 bits, which is laughable unless you happen to be denied access to a computer). Do I think any pair of bridge partners would be able to perform a key exchange after just learning the convention, much less do an on-the-fly cryptanalysis of such a verbal transaction? No.
While that sounds really cool, if computing your convention is more complicated that consulting a giant lookup table, it would be reasonable to expect you to describe it that way. There are 35 bids, (plus passing, doubling, and redoubling), which are ordered so you cannot make a bid that comes before that last one made, and the last bid made is important for the next phase of the game. You have less than 6 bits with every bid. You want your first bid to transmit info about your hand, not part of a key. In the next phase, you play one of the cards in your hands in each trick, your hand starts with 13 cards and is not replenished, and you do not know when setting up your convention which cards you hold (the point is to communicate which cards you hold), there are restrictions on which cards you can play, and you also want to play the card that wins the trick (and preparing to win later tricks is why you would be communicating about your hand). You have less than 4 bits in this phase.
Ah, thank you. I've never tried to learn bridge, so I had no idea what the specifications were. Upon further reflection (and a quick reading of the rules), I realize that I am probably not sufficiently considering the intelligence and dedication of the most intelligent and dedicated bridge players. Given the allotted bandwidth, I suspect that existing conventions are not optimal, but might be surprisingly close.