When I replied to a tweet on Mr. Mandel's Twitter account, I wasn't aiming for nor expecting a response from him but rather just sharing my sarcastic attitude about college football's selection committee with no one in particular. However, Mr. Mandel responded and challenged me to explain how the playoff teams could be determined without using subjectivity. I made a brief case for my own idea for ranking teams without using subjectivity in the post below this one. Mr Mandel followed up on Twitter with a generous compliment for me, shared an example of what he views as benefit of the committee, and asked me a question related to the case I made for my idea. It is the latter two that I wish to address in greater detail in this post.
Mr Mandel tweeted: The benefit of committee is it can account for injuries, head-to-head & other factors that aren't easily quantified by W/Ls.
As I see it, injuries are an unfortunate part of the game that require all teams to simply play the hand they are dealt and should have no bearing on a competition's standings. College football needs a committee to account for an issue that is ignored everywhere else in sports except where other committees are used? I recall the basketball committee dropping Cincinnati from a sure one seed down to a two seed following Kenyon Martin's season-ending injury in the conference tournament several years ago. How does that make any sense? Was their season made less valuable due to the injury? Because they had become an arguably lesser team due to the injury and their attempt to win a national title made more difficult as a result, they should be punished further with a lower seed?
As far as I am concerned, teams should be judged based on the value of their season only regardless of which players are available at any moment throughout the season. I suspect the need to account for injuries is tied to the need to identify and validate an imagined best team rather than simply determine a winner based on which team accomplishes whatever the rules in play would have all teams accomplish if they wish to win the competition. That said, how does one begin to account for injuries? Suppose A ranks above B under my point system. Should a committee determine that B would have placed above A if not for injuries and override the objective result based on a subjective assumption? What if A will not have a key player available for the playoffs due to an injury suffered late in the season? Should the committee deny A a playoff berth despite the better season posted by that team? What if the same injury occurred in a semifinal win? Would it make sense to replace that team in the championship game with another team that has no significant injury issues? Of course not. So why does it make sense to do just that between the end of the regular season and the semifinals?
As for head-to-head results, these are only ever used as a tiebreaker in competitions that place teams best to worst record and the winners are only ever guaranteed a higher finish in two team ties. At one time my point system included a head-to-head tiebreaker after net wins (best record) but it rarely if ever came into play given that the first two factors manage to separate teams better than 99% of the time and it is very unlikely that the remaining one percent would include two teams that played. Eventually, I decided that head-to-head tiebreakers are overrated. It is one game among a dozen or more. Where any number of teams share the same record, my point system always favors stronger schedules and still manages to favor head-to-head winners almost as much as the AP poll. For example, since 1978, there have been 33 pairs of one-loss teams ranked in the final AP poll that met during the regular season. The AP favored the winner 23 times. The point system favored the winner 21 times. Under this point system, a head-to-head winner with a stronger schedule would have to lose twice for the loser to pass the winner in the standings but only once if the loser has the stronger schedule. With a head-to-head tiebreaker, the winner's margin of error is always two games where two team ties are concerned. The difference is minimal and if teams know the point system emphasizes overall scheduling over beating specific teams, they have no one to blame but themselves when the advantage gained from a head-to-head win is erased by one loss.
Mr. Mandel tweeted: Very thorough and thought out. But all I've heard for years is how flawed the AP/BCS were, so why strive to match those?
In my previous post I shared how the results between my point system and the AP/CP/BCS are similar and different. Despite my objections to subjective methods, the problem with such has nothing to do with their results being "wrong" relative to a non-existent standard applied to all ranking systems. Different ranking systems can produce different but equally fair outcomes. One of the most annoying things about the BCS era was the annual "did the BCS get it right?" debate. The answer is no and it has nothing to do with which teams were selected for the title game and everything to do with those teams being determined arbitrarily. Earlier I mentioned that competition should be about determining a winner based on which team accomplishes whatever the rules in play would have teams accomplish. Subjective methods never specify what teams must accomplish in order to place higher. Instead, the winning arguments favoring one or more teams over others changes from one year to the next. Ultimately, I don't understand how college football benefits from a committee being able to consider factors that are not easily quantified by wins and losses if we never know when those factors will win the day.
No comments:
Post a Comment