I agree with Matt, though I disagree in others. As far as I'm concerned, I think that anything that follows three general rules can be considered fair: (1) If team A clearly finished ahead of team B in the prelims, then team A should not have a more difficult path to winning the tournament than team B. Conversely, if team A clearly finished behind team B, then team A should not have an easier path. (2) The format and packets should be upset-neutral: it should not make a special effort to create *or* prevent upsets. [As someone else pointed out, if upsets could not happen, there would be no point in playing.] (3) To the extent that it is possible, teams should not be punished for factors out of their control. Almost every format I've seen over the years has met these three criteria. While double-elim (and single-elim, obviously) might not be perfect, teams that do better in the prelims do *not* have a lesser chance of winning the tournament than anyone else: the #1 team will almost always have greater than a 50% chance of beating any other team in the field, so the probability of their winning the tournament is still larger than anyone else's. One example of a format that violated these was an SCT a few years ago where the top team had a one- or two-game advantage on the rest of the field, but had its only loss to the number two team, so the number two team was given a one-game advantage in a best-of-three finals. I would accept the argument that by head-to-head, the two teams should have been considered equal, but I don't think you can make a valid case for giving the second-place team the advantage. With respect to Penn Bowl 9 (where three of the number one seeds lost in the first round), that was because of the way the playoff packets were assembled. I had no control over that, as that was the only one of the last five Penn Bowls in which I had no involvement on the editorial end. This year's Penn Bowl, I think, took away that complaint at least; also, by using the second round-robin, I think it did a better job of determining who the top eight teams in the field were, which made the likelihood of an upset determined by a packet less likely. As far as the ICT goes, my complaint with ladder play, besides its general tendency to create re-matches in the final rounds, is the often magnified task facing teams three through six: the top two teams need win only two games to advance to the finals, while three and four need three wins apiece, and five and six need to win four in a row. This can happen even if all six teams start off with the same number of losses, because of strength-of-schedule issues--again, something over which teams have no control. --STI P.S. As far as ranking all teams go, while I admit there are people who would prefer it--and it would certainly be theoretically possible- -I simply don't think there's any real benefit to holding those matches. But if teams wanted to play--and we had the moderators to read those matches--I wouldn't object. P.P.S. The poll I had a couple years ago found that most teams had no problems with paper tiebreakers, even when needed to eliminate teams from contention for a title. The only tiebreakers I find unconscionable are those that encourage teams to run up the score against bad teams (which is why when I run untimed tournaments point- based tiebreakers never include teams mathematically eliminated from the playoffs).
This archive was generated by hypermail 2.4.0: Sat 12 Feb 2022 12:30:46 AM EST EST