This post is extremely long; don't read it if you dislike such posts. The NFL is played for the purpose of making money through the sale of tickets, broadcasting rights, and merchandise. The customers are the fans. Thus, striking a balance between playing as many games as possible and not compromising the players' health, and creating drama through single-elimination playoffs, are sensible, because they entertain fans and thus make money. Of course the NFL doesn't play a full round robin and declare the team with the best record the champion; as much of an improvement as that would be in determining the best team, it would be far less interesting to the fans and markedly shortening to the careers of the players; the loss of interest and the dilution of the talent pool would surely hurt profitability in the long run. Quizbowl is played for the purpose of entertaining the participants. The players are the customers, and it is a generally held presumption that a more fair tournament (with no assertions thus far in the argument about what is "fair") is more entertaining and more likely to create new and repeat business from those customers (I'm sure there are some who do not agree with this, but since our precepts differ so much, we have nothing to discuss, so please ignore this message.) The following conditions enter into my definition of a fair format: the same performance creates the same finish for two different teams, won-loss record (and playoff brackets created solely by won-loss record) is the sole determinant of final rank, paper tiebreakers are not used, no game is double-counted (i.e., no head-to-head tiebreakers for advancement or seeding.), and a full ranking (not just a champion) is produced. Under these conditions, the most fair format is a full round robin with all ties broken by playoffs; sometimes, this is modified by requiring a weighted 2-of-3 final if the leading team is ahead by only one game in order to offset one bad packet. This is still fair, because it is mathematically impossible for a team to win the tournament without having both the best overall record and a winning record against the second-place team. In some tournaments, the size of the field makes a full round robin impossible. Some compromise must be reached which takes fairness, available rounds, and available time into account. With the exception of NAQT ICT and Penn Bowl, all collegiate tournaments which are too large for a round robin are still small enough to run preliminary brackets which feed into either a single playoff bracket (which then serves the same function as the full round robin) or into two playoff brackets, which create a fair final: since the teams in the final have not played the same playoff fields, it is viable to assume that no assumptions about their relative abilities can be made, and they can enter a one-game final with a clean slate. The fairness of such a setup is dependent on the preliminary brackets being of equal strength, and it becomes less and less fair as the strength of the preliminary brackets diverges more and more. At Penn Bowl and the ICT, it is not viable to run any of the above formats: even the quickest one, the prelim brackets--two playoff brackets--final format, would take a minimum of, I believe, 20 rounds, and that's allowing for teams to be eliminated from contention with only one loss (which is only fair if one has a level of confidence in the quality of the packets which is so high as to be unwarranted by more than a small handful of the packet sets actually produced each year.) This calculation is an arithmetical one; I'm not skilled enough to generalize a formula for the minimum number of games needed, and I may be off by a game in either direction. So, these tournaments strike a compromise, creating the "least unfair" format with the time and packets available. NAQT's format last year should have provided no gripes. Penn Bowl's leaves something to be desired; while there is nothing unfair in determining the *champion* through single-elimination, it makes no attempt to rank the rest of the playoff teams. This year's format was an improvement over last, where the single-elim was the only stage of play after the preliminaries; only the seven teams which made the final bracket without winning should have a complaint, and that only about the imprecision of their ranking within the 2 to 7 block of the final standings. The problem of Penn Bowl is fundamentally different from the problem of College Bowl, however. I think it can be best summarized as follows: --Penn Bowl determines its champion fairly, ranks team 9 through the bottom of the field fairly, and fairly determines which teams are 2-7. Penn Bowl does not fairly determine the order of teams 2-7. The reason for this is a lack of available rounds and time. --The College Bowl NCT/RCT format does not determine its champion fairly, but does rank teams 3 and below fairly (5 and below in the RR-->four team double-elim format.) The reason for this is a desire to create "showcase" matches artificially for the purpose of maintaining the atmosphere of a televised event. Of course, if the top team entering the double-elim is only one game ahead of the eventual winner, and the playoff bracket plays out such that the champion has a better overall record with all games considered, then double-elim is a fair a format as any. But, consider this scenario: Standings After RR: University of North Tacoma 11-3 Ogdenville Community College 10-4 North Haverbrook Tech 10-4 Shelbyville A&M 9-5 (The tie between Ogdenville and North Haverbrook is not broken because the 2 and 3 seeds play each other in the first round.) Double-elim playoffs: Round 1: Shelbyville d. North Tacoma Ogdenville d. North Haverbrook Round 2: Shelbyville d. Ogdenville North Tacoma d. North Haverbrook (North Haverbrook eliminated) Round 3: North Tacoma d. Ogdenville Round 4: North Tacoma d. Shelbyville Round 5: Shelbyville d. North Tacoma (North Tacoma eliminated, Shelbyville champion) The aggregate records are thus: University of North Tacoma 14-5 Ogdenville Community College 11-6 North Haverbrook Tech 10-6 Shelbyville A&M 12-6 Shelbyville is the champion and North Tacoma is second, even though North Tacoma has two more wins in one more game, and one fewer loss in one more game. This is unfair. In the format used from the early 90s through last year at the College Bowl NCT, the situation was simpler: a 2-out-of-3 final between the top two teams. In 2001, the conditions for a markedly unfair outcome were laid. The round robin finished with Michigan at 15-0 and both Chicago and Texas at 11-4. The Chicago/Texas tie was broken with one of the unfair tiebreakers listed above. Michigan ended up sweeping both finals games, avoiding any problem with the ranking of the #1 team; however, if Chicago had won in any sequence, then they would have had a worse overall record than Michigan. Less rigorously, I assert that College Bowl has, for decades, shown its explicit contempt for the idea of the players as the customer and attempted to undermine other, better tournaments at every turn. Also, in my opinion, their questions are a joke, but I don't wish to create a tangent: even if CBI questions were perfect, their format would still be unfair. So, the reason I don't complain about Penn Bowl is because its fairness problems are as minor as possible considering the resources available, and the champion is ultimately determined fairly. College Bowl obviously has the resources for a fair format (the round robin and playoff packets which it uses anyway) but chooses to create artificial games to indulge the hallucination of a television audience. In the process, the fairness of the entire ranking is compromised. --M.W. __________________________________________________ Do you Yahoo!? Yahoo! Tax Center - forms, calculators, tips, more http://taxes.yahoo.com/
This archive was generated by hypermail 2.4.0: Sat 12 Feb 2022 12:30:46 AM EST EST