clock menu more-arrow no yes mobile

Filed under:

Humans Out-think Computers?

Or so it would seem in college football.

The two main polls at the end of the regular season in college football both had Southern California ranked #1, and Louisiana State ranked #2. Oklahoma was ranked #3 and Michigan was #4. However, the BCS computer programs had Oklahoma as #1 and Southern California as #3, and the averaging formula put Louisiana State and Oklahoma into the BCS Championship game, and left Southern California to play Michigan in the Rose Bowl. As fate would have it, the humans were "correct" and the computers were wrong, Southern California and Louisiana State both won.

In a sensible world. these two teams would play next Saturday for the national championship, but, alas, the world of college football is hardly sensible. Instead, two national champions will be crowned, and the question of which is the better team will never be resolved.

There were some flaws with the BCS approach, most notable computers were not allowed to take into account margin of victory. Strength of schedule is seemingly highly weighted, even though teams have no control over the quality of their conference opponents, and when schedules are made years in advance, teams that should be good may have bad years. However, tweaks with the BCS system will not address the basic flaw of the BCS approach: There are usually more than two teams who can successfully challenge for the national championship.

As basketball fans, we're lucky. Having a 65 team tournament allows a national champion to truly show their mettle over the course of three weeks. First, they'll need to beat a strong, minor-conference opponent (the conference champion). Second, they'll probably end up facing one or two strong mid-major programs, then a hot program or two from a major conference, and round it off with playing the best. While it is true that maybe 16 teams have a realistic shot at winning it all, having the full 65 team playoff provides the eventual champion with a series of different tests of their skill. While the selection process isn't perfect, the best are invariably selected.

In football, there are maybe 4-8 teams who could realistically think they had a shot at the title, but the system currently allows only 2. Two to six teams that could win the national championship never even have the chance.

The only sensible solution is to have a playoff. Invite 16 or 24 teams, include all the conference champions, include the "hot" teams, and let them decide who the real national champion is. Force the teams to go through the different tests: The small-conference champion, the hot teams, and the best. Reward the best teams in the regular season with home games, at least until the quarter-finals. Allow teams that lose before the quarter-finals to play in bowl games. Have a real champion crowned, like is done in basketball, soccer, lacrosse, volleyball, baseball, softball, and all football divisions except 1-A.