I haven't worked it out explicitly, but I'm pretty sure the probability that a given category decides a match is just that category's share of the total distribution. Figure that the average NAQT match goes to 24 TU, 22 boni...that's 46 questions all told. Off-the-cuff (ignoring tossup-bonus coefficient thingies), for two equal teams, ~2% of matches should be decided by a category that only appears once in a round -- and I don't think anyone here's arguinging that 1/0 or 0/1 is too high a trash fraction in an NAQT round. Now, figure that a medium-big tournament (16 teams) will involve over a hundred matches. That means that there's a greater than 90% chance that at least one match in the tournament will be decided, in the balance, by a microcategory. So after you've been to 10 or so tournaments of decent size -- that's one spring semester for some teams -- there's a 50-50 chance that your particular team has lost a bitterly-fought round because of "that damn question about duck à l'orange". So we can state this principle: Any given question in a match, no matter what category, can decide a match. This is probably something one should keep in mind when writing questions, more than a theorem of distributions. Another off-the- cuff figure: just *one* out-of-left-field question in the whole tournament, for a 16-team tournament, has about a 30% chance of causing an upset somewhere. For heaven's sake, write good questions. ERS --- In quizbowl_at_y..., "Stephen Webb" <sdwebb91984_at_y...> wrote: I > don't mind a little trash here and there, but NAQT has allowed it to > become a major field that can decide matches, which should never be > the case for anything that isn't a TRASH tournament (for obvious > reasons).
This archive was generated by hypermail 2.4.0: Sat 12 Feb 2022 12:30:46 AM EST EST