Posted by Johnnymac 8:49 AM
ESPN's "Principled" College Pickem

A few years ago ESPN decided that in response to the "problem" of college gambling, it would no longer be associated with point spreads of any sort for college athletics. Spreads for college games in any sport were no longer posted or mentioned anywhere on the ESPN.com website and part of that change involved revamping the popular "College Pickem" game on the website as well. One can argue about the legitimacy of ESPN's stated reasons for making the change, but one cannot argue that their changes to the College Pickem game have been pretty useless. Why? Because the point spread is one of the most elegant and efficient gambling devices ever conceived for betting on football.

As most of us know, the point spread exists in football betting to encourage equal dollar volumes of betting on both teams involved. And as action disproportionately increases on one side of the spread, it's adjusted to a new equilibrium point that better reflects equal sentiment on both sides. By definition, it's not necessarily intended to be an accurate predictor of the outcome of football games - just a reflection of what bettors think the outcome will be - but as any student in a Freshman Statistics course can tell you, with a sufficiently large pool of bettors the point spread will move around until it does indeed approach the actual score of the game. As I have said many times, I love gambling and poker because of the way that it's a reflection of economical theory applied to an everyday situation. The point spread is a perfect reflection of this - anything other than the "true" score of the game presents an arbitrage opportunity that is eventually corrected through a form of market activity.

(David Sklansky has a very good book that, among other things, explains the theories behind sports betting, if you are interested)

So what's the point with ESPN College Pickem? By changing the format of the game, it's inherently inefficient.

This post is specifically about the current incarnation of the game, which is the third. The second version, which was in place up to last year, was completely bogus in such a way that the results of the individual games didn't really matter - so long as you picked one or two big underdogs correctly each Saturday you were disproportionately rewarded with more points than you would have received by correctly picking six slight favorites in the other games of the day.

The current version is slightly different, but is quite illustrative of my point that the point spread is an elegant invention, because by trying to avoid the point spread ESPN has created a game that is easily arbitraged by players who use the spread's efficiency as an indicator of each game's result. The new rules require the player to pick each game heads-up and then rank each game for confidence from 1 to 10. Points are then awarded equal to the confidence level for each correct pick.

Now, I have never been a big better on college football, but I follow it a little bit more than I used to. Nonetheless, I don't have the time or desire to follow every single team out there year in and year out and I can't tell you off the top of my head if Florida is going to beat LSU this weekend and whether or not I feel more confident in my pick in that game than I do about my pick between Oklahoma State and Colorado. But I do care about scoring more points than my buddies and having bragging rights in the game, so how can I accurately pick the games and assign accurate confidence levels? By using the point spread.

To do well at this game, all one has to do is look at the point spreads for the week, see which teams are favored to win by more points, and then correspondingly pick the bigger favorites to win with higher confidence than the teams that are favored by fewer points. While the point spread is designed to turn a lopsided matchup into a 50/50 bet, the scoring system in the ESPN game turns a 50/50 matchup into a lopsided matchup. It's like shooting ducks in a barrel and is the very definition of "insider information". Economics in action.

So I guess I have to compliment ESPN on being principled, but I can't say that somehow their principles are more important than a competitive game.

0 Comment(s):

Post a Comment