Mar 1 2014, 9:12am CST | by Forbes
In 2014, college football gets its long-awaited playoff. Four teams, two rounds, with a special selection comittee charged with naming the deserving schools. Their job will be, in a word, hard. Being annointed No.4 (last team in) instead of No. 5 (first team out) by the commitee can be worth a ton of presitge to school, not to metnion more money (the bulk of the revenue from the 12-year deal with ESPN that pays an average of $500 million annually gets doled out equally across the FBS conferences, though some will be carved out to the participating teams. Not to mention – making it to the big final four football dance figures to do wonders for donor dollars and ticket sales).
So when six or seven teams look to be in the mix at season’s end, who do you pick? Exerting a lot of influence behind the scenes will be a staff of advanced metrics experts at ESPN, the powerful sports network that is now embracing analytics to such a degree that it’s become the title sponsor for the annual MIT Sloan Sports Analtics Conference in Boston, which just kicked off its 2014 two-day forum on Friday. The ESPN stat geek team didn’t exactly exude confidence in the committee’s expected ability to get it right.
While the 13-member committee is dotted with big names and lots of football pedigree – Wisconsin AD Barry Alavarez, former Nebraska head coach and AD Tom Osborne, retired Air Force Acadameny superidentent Lt. General Michael Gould and even former U.S. Secretary of State and Stanford professor/super fan Condoleezza Rice – the ESPN numbers crunchers voiced concern that all are over 50. That means they grew up in the college football world of polls and eye tests, not sophisticated sports data. How can we trust them to get it right?
The ESPN staff has no formal control over the committee, but they’re planning to submit their numbers in order to help them out. And of course it’s their network that’s airing the games, so expect them to have some influence. And all in all, that’s probably a good thing. Here’s the gist of what the panel presented in Boston:
Picking the four best teams means going beyond the basics of won-lost record and strength of schedule. It means breaking down each game, quarter by quarter, to assess the level of domnence. For example, last year’s national champion, Florida State, would have been a deserving high seed in a playoff (some thought hey didn’t due to their less-than-stellar ACC schedule) not only for going 13-0 but by winning every game by at least 14 points. And particularly telling: the constant recalculation of a team’s odds of winning as a game progresses. In FSU’s case, the ESPN staff figured their average in-game odds of 84% for the season, thanks to so many big leads and statitistical dominance early in games. That means they typically had an 84% chance of winning at each point in each game during the season. That metric eliminates the need for teams to run up scores against helpless opponents. Example: in one game last fall Florida State jumped out to an early 35-0 lead over North Carolina State before calling off the dogs and winning 49-17. The game gets classified as a “blowout win” regardless of the final score, thanks to the constant in-game tracking of the odds of winning.
Add up all the advanced stats for FSU and their opponents for the season, and the analysts assert that an average team would have won seven games playing the Seminoles’ schedule, making their 13 wins genuinely impressive. Maybe even more interesting – the ESPN crew had numbers to support the view that Alabama, No. 1 and undefeated all year until a flukish, last second loss to Auburn in the regular season finale, didn’t deserve a spot in the final four. Given their so-so competition – strong SEC teams like Georgia and Missouri weren’t on the Tide’s schedule in 2013 – the advanced numbers show that a typical Top 25 team had a 36% chance of duplicating their record against the same schedule. That puts Alabama a tad behind Michigan State (30%) and Stanford (29%), who would have ostensibly joined Florida State and Auburn in a four team tournament.
As one ESPN panelist put it, no sport cries out for advanced analytics like college football, given its plethora of teams and conferences and its disparate schedule. Thannks to the coming playoff system, the battle of the new breed stat head vs. the traditionalist, brewing for awhile now in Major League Baseball, is about to hit its next frontier.
Source: Forbes Business
Forbes is among the most trusted resources for the world's business and investment leaders, providing them the uncompromising commentary, concise analysis, relevant tools and real-time reporting they need to succeed at work, profit from investing and have fun with the rewards of winning.
blog comments powered by Disqus