A lot of the hobbyist designers on
BGDF really, REALLY hate player elimination mechanics. There's quite a bit of absolutism regarding this; people who would NEVER think about playing or designing something that includes this "vile" feature. Which is bad.
First of all, in general, absolutism regarding any kind of game design is bad. It locks you into specific things that you KNOW work, without trying something new that MIGHT work. Obviously, some things are pretty tough to work around, say, like gravity, if you are trying to do a
Jenga or
Villa Paletti styled game. Bu,t if you don't bother including the laws of physics, the logical creation of game rules and mechanics should never be locked in to "always includes" and "never includes". Player Elimination is something that maybe needs to be looked at a little longer before being tossed away like a three day old Chic-Fil-A sandwich sitting in a car in the hot summer sun.
I can understand why most people don't like it. In most implementations, it's a pretty harmful, even brutal thing. Much like school bullying, games where the sole purpose is to get rid of everybody like Monopoy can lead to long downtimes between game for the unfortunate soul cast aside early. And I don't think I've ever actually finished a game of Risk without the board being thrown off the table, showering the tokens everywhere.
But under the right circumstances, Player Elimination can work.
Case 1:All decisions, in game designs, (and life I guess), come down to making the best choice given the understanding of known circumstances. Your best choice SHOULD be based on the most preferred outcome of these choices. Being eliminated from a game can be just another part of this equataion. Clue, otherwise known as Cluedo, is a prime example of this. On any turn of this game, a player can choose to eliminate himself by guessing the the murder case wrongly. The game begins with a player having, just throwing a number out at this point, a 95% chance of guessing incorrectly. As the game continues, that number slowly decreases as you gain more information until you get to the point where you are 100% correct and 0% of guessing wrong and being eliminated.
However, what if you feel that another player is very close to guessing also? Maybe it's worth taking that 10% of guessing wrong instead of the 95% chance that you won't have another turn? Since you only get one crack at this, and Player Elimination is factored into your decision as a large weight of the percentage (let's face, without P.E., this becomes a non-decision), it's an important feature to have.
Another game to look at, would be Tournament-styles Poker. While going "all in" in a regular poker game would just mean visiting the closest ATM to get back in the game, tournament play is another animal. Typically, in any given poker hand you are comparing expected value with your cards versus what you think other might have versus pot value. This kind of thing let's you determine how you should play a particular hand. Epected value being what it is, even playing CORRECTLY to your expected value, does not mean winning every time; just that over the long haul over many hands in the same situation, you will come out a winner. However, in tournament play, since you just can't buy yourself into the next hand when your correct Expected Value play goes south, you now need to include into the equation what it means to be possibly eliminated from the tournament and watch from the rail. Or inversely, what is it worth to you to try and chase someone out of a tournament by getting your target to go all in? The balance of Player Elimination in this scenario is pretty compelling; enough to make many a TV show.
Counterstike, the Half-Life mod, is also a darn good example of how Player Elimination ca drastically influence how one plays a game. Instead of the typical, run-around-at-a-breakneck-speed-on-automatic-fire multiplayer experience where you simply press a button and re-enter the game at a spawn point, in Counterstriek when you die, you are dead. That's it. You have to wait. And watch. And it's VERY easy to die in Counterstrike. This influences your style of play IMMENSELY. Suddenly your little 3D model takes great care in hiding, aiming, working with your little 3D buddies on screen, etc. Death means something. It means waiting. And the potential to being eliminated is a great reinforcer to doing stupid things.
Case 2: In games like
Werewolf/Mafia and
BANG! the whole game is about eliminating players, albeit, the CORRECT players. Since players fall like moths near a bug zapper quite quickly, every player is somewhat involved in the decision of who must fall first. The expected value of killing a "good" (in your eyes based on the character you are) player or "bad" player off is balanced between trying to save your own skin, and getting more information regarding what the other player's goals are, and if they are on your side. Much bluffing is involved, and the group dynamics are usually fun to watch. Without being eliminated, however, these games probably fall somewhat flat; and the tension isn't there.
Case 3: Self-imposed Player Elimination!! Try to get rid of your cards faster than other players!!! Hurry, hurry, hurry!!!
Great Dalmuti, a presidents/asshole variant, is a prime example of this. Especially nice amongst this type of game is that the first-one-out feature carries over to the next game, where the faster you get out grants you special bonuses, and the game hands fairly brutal penalties to those who are last out. Woe be the low peasant who stis and watches as all the other players dump the pairs, while the measly peasant sits with 6 twelves! An, oh by the way, I would classify racing games to fall into this category, too.
As an aside, the Great Dalmuti deck is tuned rather nicely. It is much easier using the Dalmuti deck to rise out of peasantdom to merchantdom than using a standard 52-card deck. This is due to the ever-increasing occurance of card sets as the deck works it's way down the values (1 card for the 1, 12 cards for the 12). It's a fairly brilliant use of math to balance the game, but not enough that everyone has an equal footing.
Of course, I am probably stretching the definition of Player Elimination on this final point, since most people would claim that eliminating a player from the game is a lot different than winning the game. Which is probably true. But since I am trying to defend it, and maybe wrap some other ideas around a prospective game designer's noggin that maybe it's not such a bad thing after all, I thought I'd include it. Player Elimination can add quite a bit of tension to a game; it just needs to be carefully balanced.