Point Spreads and Efficient Markets

I’ve been playing around with the database on Pro Football Reference. They have a database of NFL games going back over 30 years with point spreads. Give me any open database, and I’ll spend time looking through it.

What fascinates me is inherent biases in any type of market. Just as value investing has been proven to hold up over time, there are also small biases in the market of football point spreads. It may seem odd to think of it this way, but a point spread is a market just like any financial market.

For example, favorites fail to cover slightly more than they do cover. This doesn’t appear to be an exploitable bet since bettors have to pay the vig (the bookie’s cut). I suppose that people like to bet on winners so they may overpay. That could be the same reason why so many stock market superstars are overpriced.

Also, favorites playing on the road don’t do a good job of covering the spread. This is particularly acute during the latter weeks of the season. Also, wider points spreads (over 10 points) seem to be unjustified as a whole. These biases are well-known to gamblers and have been documented by academics, but they were new to me.

I think there’s something about numbers, odds and statistics that people just don’t get. If you were to say to a roomful of intelligent people that men are on average taller than women, I can guarantee you that someone will say, “That’s not true, what about Tracy?” Yes, It’s frustrating as hell, but take some solace because this is why we’ll always see market mispricings.

Posted by on November 20th, 2012 at 10:31 am


The information in this blog post represents my own opinions and does not contain a recommendation for any particular security or investment. I or my affiliates may hold positions or other interests in securities mentioned in the Blog, please see my Disclaimer page for my full disclaimer.