Tournaments, Time Limits, and a Chess Story
Over at the geek, it looks like there will be a Con. And the organizers asked “Tournaments, yes or no?” I’m mildly pro-tourney, but for the most part I skip them. Playing the same game three or more times in a row bores me (with a few exceptions). Tournaments tend to bring out the worst in people (and sometimes me). Also, I’ve done it. Five or ten years ago, I played in lots of tourneys.
Interestingly, I noticed that the War of the Ring’s tourney at PrezCon just posted it’s rules. A four hour timelimit followed by an artificial tiebreak system. All I can say is “ugh”. Time limits are a necessary evil in tournaments (especially for games that have a highly variable game length, like War of the Ring), but I dislike the artificial ending. If you are going to have a time limit, then make it a hard time limit with clocks. Running out of time means you lose. Yes, this introduces another way to lose, but this gives everyone an incentive to play quickly and actually finish the game rather than stall and win on tiebreakers. You’d need a way to clarify who was on the clock, but you avoid all sorts of “He’s winning on tiebreaks and stalling” accusations (true or not).
I feel strongly about this because our Shadowfist league times the games and ends when the buzzer sounds. For fluid multiplayer games (where turns can be interrupted by other players), individual clocks don’t really work so this is the best we can do. But when the clock’s at a few minutes you can sometimes see the “Ah, I should make this play to improve my tiebreak points” light go off. [Fortunately, this is a small group, so social pressure works fine to prevent gaming the tournament system. But that doesn’t work as group size grows.] I still don’t like tiebreaks, though. I’d prefer adjuication, but that requires an acceptable evaluation function, which just doesn’t exist (for many multiplayer games). Ah well.
I’m reminded of a story in one of my chess books. A grandmaster plays a simultaneous exhibition. As time winds up, there’s still a game with a young boy (12). [The book doesn’t say, but presumably the GM won most, if not all, of the other games]. The boy is up a pawn in a complicated rook and pawn endgame, but the grandmaster and assembled experts analyze the game and agree that it’s drawn (as many rook + pawn endgames turn up).
The boy fires off a letter to the experts giving a detailed analysis of the final position. The Grandmasters agree with his analysis, overturn their decision, and award him the win. The boy’s name? Garry Kasparov.