Antipatterns: Elimination
Posted by Rob Herman at May 4th, 2006
Inspired by the popularity of design patterns in computer science, a couple of authors recently wrote a book entitled “AntiPatterns.” The antithesis of a design pattern, an AntiPattern is a commonly made mistake, an easy trap to fall into that might seem like a good idea at the time but turns out to have negative consequences.
It’s a good idea that didn’t turn out to have the massive impact of design patterns. But I’ve certainly seen games that make the same mistakes that have been made many times before, so here you go.
Name: Elimination, or Last One Alive Wins
Problem: You have three or more players. Every player has the opportunity to attack any other player in some way.
AntiPattern Resolution: After undergoing a certain amount of abuse, players are eliminated. The last person remaining at the end wins.
Issues: There are several. First and foremost, players who are eliminated have nothing left to do. In an online computer game, this isn’t necessarily a bad thing—these players can just start the next game. Civilization 4 and Warcraft 3, for instance, have online free-for-all matches that don’t suffer from this problem. In a board game, though, the players have to look around for something else to do. Not so bad in a big party, but if you just want to gather the players to play a particular game, it makes it tough.
Second, usually, attacking taxes your resources. This creates an incentive to stall and hope someone else attacks and does your work for you—which causes the game to stall.
Examples:
- In Lunch Money, you have a hand full of attacks that you can unload on whoever you want. The game has theme but little in the way of strategy. You lose when you can no longer convince people to stop hitting you, and they decide to just finish you off. If you’re lucky, everyone else will be finished soon as well and you can get on to something else.
- In the board game Titan, you form legions of creatures that roam around an battle other players’ legions. The goal is to eliminate all other Titans, which are unique and fairly powerful creatures. The designers had some good ideas for of ways to mitigate the second problem. As you win battles, your Titan games experience, which makes it increasingly difficult to assassinate and eventually a powerful fighter in its own right. Also, winning fights earns you Angels, powerful creatures that can be summoned into other fights; bolstering your ranks with Angels helps make up for losing your other creatures in hard-fought battles. However, the first problem is pretty much insurmountable, in my opinion. The game is so massively complicated that it’s completely unsuitable for a party game; you’d have to gather a group of people specifically to play it. And if one player gets an unlucky break and gets killed off early, does that player just sit and watch TV for the next 3 hours?
- As I mentioned before, WarCraft 3 has a free-for-all mode. The first problem is mitigated because if you get eliminated, there’s no big deal; you can start another game right away. The second problem is mitigated in a fascinating way. You have “hero” units that can gain experience as you kill enemy units. With little experience, hero units are expensive and not particularly powerful; but with experience, they are overwhelmingly powerful. It’s important to fight a series of small engagements as the game goes on to build experience; but not so much that you are vulnerable to an overwhelming attack by someone else. If you do see someone in a weakened state and can finish them off, you are in a good position to take their resources for your own.
Brother brother and best friend across the road spent a lot of time trying to make M:tG work for three players. In that game, it always came down 2 people attacking each other, and the thrid killing off the surviver. We had the exact same problem with Risk.
IIRC, the Vampire card game by white wolf required you to attack the person to your right only, and you could only activly defend against the person on your left. These two behaviors use a few common resources, but mostl they where seperate. I may not be remebering everything correctly though. Not sure.
It seems like one other way to mitigate this is to award some sort of short term reward for victories that cost recourse to compensate for the lost resources. This is actually the opposite of what Risk does, where you get long term benifits from early battles, but you are usually pretty weak right after the fight. This encourages the third man out to move against you right as you are removing the first player.
As an avid (probably obsessive) M:TG player I have considered multiple variants for the three player game. It can suffer horribly from both problems mentioned in this article. First by creating the “No one can win” scenario, then (in many games played by more experienced players) leaves the remaining two players in a stand off for a while because both have built up their defenses and saved the cards that deal with their opponents threats. Fortunately http://www.wizards.com/magic/welcome.asp?regionset=true has a lot of variants to try that mitigate these issues. One of the most inventive I think is “Zombie” where who ever dealt the killing blow to the first one out adds that person to their team. The player gets some amount of life back and continues to play from the board position they had. This encourages players to be offensive early so they can wipe someone out and gain the advantage. It also lets player do something after they have been eliminated.
I see this pattern as part of a larger issue. When designing a game, you have to ask yourself whether you want it to feel like anyone could win up until the last turn, or if you want someone to be able to become the likely/obvious winner before then. (For these purposes, I consider a game where people are eliminated to be equivalent to a game where they are forced to play for a while after they lose hope of winning. Either way, it is boring for them.) Traditionally, games have been about elimination or the rich getting richer, but the modern Euro trend has brought an emphasis on keeping it “everyone’s game” until the end.
I don’t think this issue can be solved as easily as saying that one way is always wrong. True, designers who don’t know what they’re doing often assume that their game has to involve player elimination, and players who don’t know what they’re doing often start a player-elmination game in the wrong situation, but that doesn’t mean that it isn’t the right mechanic to use sometimes. For one thing, Euros often feel contrived, and you can come away from some games feeling like the first 2/3s of it hadn’t really mattered. If anyone can win at the end, why bother trying hard at the beginning? Also, games where everyone stays in the running until the end usually feel more abstract than games where you can knock people out or take a commanding lead. Theme is important to me in games, and the visceral thrill of knocking someone out often makes up for the frustrations that go along with it.
I wouldn’t call elimination an antipattern so much as saying that it’s an element with significant strengths and weaknesses, and a game designer needs to carefully consider them when deciding how the game will work.