Games, Rationality, and Fairness

It isn’t an “unfair” strategy, especially if you warn people beforehand, but it is a bad strategy as mentioned above. Retaliate, by all means, but dont throw away victory over some petty feud. If I were playing against you, I would conspire with other players to take you down. After all, if you attack me at the expense of everything else, my coconspirator would wipe the floor with you.

That would work largely to my detriment and mostly to the advantage of my ally, but having a rapid dog on your doorstep is too risky to tolerate.

I’m a fairly serious competitive board gamer, and I employ a similar strategy on very rare occasions, when playing with other serious gamers. But for me it has nothing to do with alliances.

It comes up in a situation like this: it’s a 3 player game, and I’m losing in the early going, but am not necessarily out of it yet. But there’s one scarce resource that I really really need to acquire to get back into it. (Ie, one place I really need to successfully build a settlement in a game of Settlers of Catan). Someone else has the opportunity to take that resource, and it’s fairly inconsequential to them, compared to how essential it is to me. I believe that if I do not get that resource, my chances of winning the game are very very slim at best.

In that situation, I will tell the player who is potentially going to take “my” resource that if they take that resource, I will consider myself to have no hope of winning the game, and will thus redefine my objective as insuring that that player does not win.
The trick here is to not just start threatening vendettas every time there’s any competition for anything. In fact, I only ever threatened a vendetta 4 or so times, and only ever carried them out twice or so. But of course it’s absolutely 100% crucial that you do not threaten the vendetta if you are not totally prepared to carry it out.

I’d say it can be a rational strategy. If you play with the same group of people your reputation will benefit you because people will hesitate to betray you. That’s a genuine advantage in the game.

It could go too far and become counter-productive. Other players might decide not to ally with you in the first place or, more extremely, not play games with you. So you’ve got to achieve a happy medium.

Play Illuminati with me twice and I will make it so you can’t ever do that again. :wink:

I had a friend who I was teaching to play the game, he was being all loyal to me, so I smacked him in the mouth just to teach him the final lesson. :wink:

I often use a tit-for-tat strategy in war games (i.e., games in which you eliminate other players: Monopoly, Risk, Diplomacy, Supremacy, Axis & Allies, etc). Not so much in games that are races (i.e., games where you cannot eliminate other players: Settlers of Catan, Civilization, Sorry, Hearts, etc). The key to successful use of tit-for-tat is proportionality. If someone snaps at my borders, I snap back. If someone presents an existential threat to me, I respond in kind. If someone betrays my trust, I try to take back what the cost me, plus enough extra that they ask for mercy.

As to whether this behavior is rational within the context of a single game, I think it is. If a player betrays another, shouldn’t they expect reciprocation? Why should a player think the can get away with taking advantage of another? There is rarely only one path to victory, and a betrayer of trust is unlikely to find to other allies, thus making them an excellent target. I’ll only go on a suicide attack if that player has rendered me truly incapable of winning. In that case who wouldn’t go after the one who crippled them? The alternatives don’t seem any more rational.

In most games you’re looking at cost/benefit analyses. You might decide to betray an ally because you figure you’ll gain more from the betrayal than you’ll lose from breaking up the alliance.

Let’s say you’re playing Risk. You’re the most powerful player on the board and you control ten territories. You’re allied with the second most powerful player and he controls eight territories. Now let’s say you see an opportunity to steal two of his territories.

Now if everyone were completely rational, your stealing the territories would make no difference in the game. After the theft you would be more powerful so the rational thing for your partner to do would be to accept the theft and continue to be your ally. It makes rational sense for him to be allied with the strongest player and forget the past. So if you know your ally will act rationally, the rational thing for you to do would be to steal his territories.

But suppose your partner is irrational and you know he’ll dedicate himself to fighting you if you steal his territories. It doesn’t make sense for him to do this - he’s going to have only six territories and you’ll have twelve so he’ll probably lose. But you know he’ll do it anyway even if it’s not the rational thing to do.

So now you have to weigh the benefits of gaining two territories against the cost of having a eight territory ally turn into a six territory enemy. And you decide two territories isn’t worth it so you make the rational decison not to take them.

So now you’ve got the situation where two hypothetical players were in the same situation. And the one who plays rationally would lose two territories in that situation and the one who plays irrationally would keep those two territories. So if you’re a rational player in that situation wouldn’t playing irrationally be the rational thing to do?

I don’t understand the people in this thread who are importing rules of ‘fairness’ and ‘implied goals’ to these games. So long as what you do in the game is within the written rules of the game, you are golden, in my opinion. So make that illogical move, lose now to win later, do what you want. That’s the real game. If people can’t handle that then they’re not interesting enough to game with.

To the OP:

  1. Is it “fair”? Yes, it’s fair, because it is within the rules, and not everyone plays to “win.”

  2. Will it cost you in the long run? Yes. It pisses off some people, and it will make people reluctant to ally with you in future games. You have to decide if these short and long term downsides are worth the potential upside of precluding an unscheduled end to an alliance.

  3. If you are slavishly loyal to alliances, you are not going to maximize your result in most board games. :wink:

I know a fellow who says he always plays the “Unstoppable Rageful Vengeance” trick. He’ll sit around and turtle till doomsday. In many games, this is enough to win, so people have to attack in order to keep him from victory. Then he goes into Unstoppable Raging Vengeance and annihilates them.

He doesn’t actually use it in every game, but in enough of them that I usually don’t want to play with him. :slight_smile:

Your strategy is neither unfair nor irrational. You treat each game as a costly signalling exercise. You signal that you are willing to pay a disproportionate cost if another player defects and that you will externalize that cost onto the defector. This means that the gains for defection need to be really, really high to compensate for the costs the defector will have to bear. You need to behave predictably in order to maintain your credibility.

This means that when you are betrayed, the defector had better cut your throat instead of just kneecapping you. You invite spectacular, game-ending betrayals instead of run-of-the-mill ones that can more easily be recovered from or turned around. I used to play a lot of Diplomacy, and this is the distinction between a “lunge” and a “stab”. If this works for you or if your fellow gamers can’t pull off a serious stab every game against you, then the strategy is fine.

Pretending that you treat each game as independent is a non-starter. Everyone has a style, has tendencies, or has other idiosyncracies that can be exploited in repeated play. You just make yours extrinsic, so it gives people more leverage to object. But this is not really a fair objection, especially when the best response your strategy is pretty obvious.

It sounds like he can win whether people attack him or not.
If he’s the best player by far and the other players don’t know how to stop him, then any strategy will do.

Little Nemo I don’t understand how you view attacking you back as irrational. You by attacking your weaker partner have shown a lack of good faith, and as such have become the most powerful person in the game. By being reduced as a threat, he now is in the perfect position to lead an alliance against you who is destined to win if he cannot ally against you. At that point no one has any incentive to ally with you because you’re a double-crossing snake, and also because you are certain to win if anyone allies with you.

So it is completely rational to ally against you under the scenario you outlined.

Suppose you were all just sitting down at the table and the territories were distributed at random. You have six territories and somebody else has twelve. Everyone else has less. It’s pretty obvious that if you ally with the twelve-territory player, the two of you together will probably mop up the other players. So at that point, allying with the strongest player is rational.

So saying that you won’t ally with him because he stole your two territories isn’t rational. Breaking up the alliance won’t get your lost territories back. And declaring war against the most powerful player in the game probably isn’t a rational strtegy either. Rationally speaking, the past is the past. You should no more hold it against him for stealing your territories than you should because he stole your comic book in the third grade. Rationally speaking, you have to do what’s best for you based on what the situation is now. You should just suck up the loss of the two territories and remain allies (although you should also keep some forces behind in your remaining territories to avoid future losses).

Are you playing some variant with shared victory? Because if there’s only going to be one winner, helping the guy in the lead just makes sure you come in second.

Right but what incentive do they have to help you win after you just screwed them? How is it best for you to help him win?

I assume that most players are hoping that they will gain more from the alliance than their partner will. There’s forty-two territories on the board. Right now, the lead player outnumbers me by a 12:6 advantage. If we stay allied and conquer the other twenty-four territories and divide them up between us, then his advantage falls to 24:18 and I have a better shot at beating him in a one-to-one battle. Not perfect (and maybe not achievable) but it’s still more rational for me to try to seek a battle at 4:3 odds rather than at 2:1.

In a broader sense, the premise of this thread is that you’re making choices between the options that appear available. If the player in the lead has an insurmountable advantage, then there’s no point in playing the game until it’s end - just play until somebody take a lead and then call it a win for him.

And if the betrayed player is just forming a new alliance with the hope of using that alliance to win the game then he’s still playing with the goal of winning. His plan is rational both for revenge and for victory. The debate here over whether the grim strategy is a good idea is based on the premise that the betrayed player is sacrificing his chance at victory in order to destroy another player’s chance at victory. He’s asking whether it’s rational to choose revenge over victory in some cases.

Years ago, I was in a multiplayer hotseat game of CivII with 2 coworkers. We had a standalone PC at work and during lunch hours we’d get some grub and play and play for awhile. (one of the players was my boss). My boss was into strategy games and he played to win. We formed an alliance during the game against the AI players and mutual defense against the other coworker. But I ended up getting my butt kicked for awhile but the AI (boss did try to help, but…) I ended up bottlenecked and my tech was falling behind. During the end game when the other coworker attacked me, boss revoked the Mutual defense treaty. I don’t blame him. It was what I would have done. It put him in a superior position to win, which he did. (by letting us beat up on each other, he was able to become the only viable superpower and thus win the game by spaceship launch before we could do anything).

My wife is the same way. We used to play risk a lot with a groiup of guys in the barracks. (before we were married and when my wife was in the army). If you had an alliance with her she’d back you up. If you betrayed the alliance she was a monster. (she often won). And if she could win by betraying you…she showed no mercy! I think I won one game back then and she was pissed at me because I bragged all night about how I pwned her. As she says I’m a good loser, but a poor winner.

I am used to playing against the AI in games like the civ series or Alpha Centauri, and I am kind of the same way. The AI is predictable and will betray you sooner or later. I usually position myself to be able to crush them for any betrayal as soon as it happens. But I also position troops to sneak attack them preemptively if its to my advantage.

No one wants to play a strategy game to lose, after all.

I disagree.

Your goal is to win. If you ally with the strongest player, he will remain stronger than you, and you are highly likely to lose. If you ally with one or more weaker players, your alliance may be able to take down the leader, leaving you a better shot at winning.

Just watch your back!

Where the game calls for it, I will employ this attitude. I don’t say ‘strategy’ because it’s certainly not done with any idea of winning in mind. Winning is the most boring part of most games. All I want to do is build my little empire, my stack of zorkmids or whatever. People who interfere with that take me on as an albatross around their necks for the rest of the game.

I’m not seeing the case for this being unfair. First of all, you are allowed to have goals extrinsic to the game. Or, more to the point, try assuming you aren’t and follow the absurd consequences. You are allowed to act on those motives.

By going after only people who fuck with me first, the decision of whom to use my unfriendly resources on is taken out of my hands, which is of tremendous utility to me. The fact that I’m willing to sacrifice the win that I wasn’t all that enthusiastic for to begin with is a powerful disincentive for people to take my fun away.

Of course, if other players drop out, then there’s no need to fret about who did what to whom. But as long as people have a choice whom to dick with and they dick with me, then I’ll have my pound of flesh.