Sunday, September 09, 2007

An NFL over/under system that beat the Vegas line

In a recent post on his NFL blog, Brian Burke comes up with a system that handily beats the Vegas “over-under” line for team wins.

Here’s the system. If a team is predicted to have 9.5 wins or more (that is, go 10-6 or better), take the under. If they’re predicted to go 6-10 or lower, take the over.

If you did that over the last two seasons, you would have gone 24-10. That is, needless to say, very impressive.

Now, it’s possible, and even easy, to mine the data and come up with formulas that pick winners in retrospect. So it could be coincidence that this method worked so well in 2005-06. Indeed, there's some selection bias, because if it hadn’t worked well, Brian wouldn’t have posted it.

But, I think this one might actually have some merit, because it takes advantage of two biases that bettors have. The two principles that fans have trouble with, according to Brian:

1. The NFL is impossible to predict before the season starts. And,
2. Regression to the mean rules the day.

Basically, a lot of what happens over a short 16-game season is luck. And so, consider a team that goes 10-6 last year. It could be that this team is an average 8-8 team that got lucky, or a really great 12-4 team that got unlucky. There are so many more average teams than great teams, though, that the odds greatly favor the “lucky team” hypothesis. But fans have trouble grasping that; they see the actual 10-6 record, and think the team is likely to repeat that performance.

The same is true, in reverse, for the teams with bad records.

So that's a decent explanation for why Brian's system works (if it does): bettors underestimate the amount of luck, and therefore think teams are less average than they actually are.

Still, it’s dangerous to suggest that a strong and liquid betting market is getting these things wrong. All I’m saying is that, compared to other betting systems based on historical results, at least this one has a very plausible explanation for why this happened.

And, intuitively, it just seems right. As of right now, TradeSports is saying that the New Orleans Saints (who are now 0-1) have a 42% chance of going 10-5 over the rest of the season. Given that they were only 10-6 last year, and 3-13 the year before, doesn’t that sound a little optimistic? There are probably personnel changes that are driving some of this enthusiasm (I don’t follow the NFL much, so Drew Brees is the only one I know), but I’d still be taking the under here.

As I said, I hesitate to suggest that sabermetricians are smart enough to beat the Vegas line. But still, if you pointed a gun to my head and forced me to choose one system to play, this would be it.

Labels: ,


At Monday, September 10, 2007 4:00:00 PM, Anonymous Anonymous said...

I'm guessing this is a case of data mining/selection bias with a very limited number of data points. Line-makers and a sufficient number of players are certainly aware of regression to the mean concepts.

Winning 10 games in a year is only a 62.5% winning percentage. This opening weekend alone, 8 of 16 games played had a favorite on Trasdesports at 62.5% or greater. Obviously not all are of that consistent caliber, but the cream of the crop will likely post similar favorite % probabilities all season.

Also, a good team will have a win distribution with a fat left tail, meaning their median (which the bettor cares about in this case) is greater than the mean.

But this is just my guess, more seasons of data would obviously be a help. (Especially because you only get 32 data points/year)

On the New Orleans comment: they were only given a 31% chance of beating Indy anyway, so the market is saying that their realized loss didn't impact their chances of a 10-5 or better season by any more than 8% -- that sounds about right to me.

(Keep in mind I lead a tortured life of believing in efficient markets and gambling at the same time...)

At Monday, September 10, 2007 7:27:00 PM, Blogger Phil Birnbaum said...

The more I think about it, the more I agree with Nate. I perhaps oversold the system somewhat. It should take a lot more than two seasons' worth of data to convince me that the Vegas line is irrational.

BTW, on New Orleans, I didn't mean that the odds were too low in light of the NE game, just that the odds seem too low in general. But, again, the oddsmakers know a lot more than I do.

At Sunday, September 16, 2007 12:46:00 AM, Blogger Brian Burke said...

Well I did some more homework on this. I found a website that had done somewhat similar research and had historical over-under data from 1996-2005.

Using the system I proposed, a bettor would have won 86 out of 148 of his bets, a rate of over 58%.

Although not the 70% rate over the past two years, it would remain a safe and lucrative system. I'm guessing most professional gamblers would sell their soul for a 58% winning rate over a 10 year period.

Data from:

At Sunday, September 16, 2007 9:22:00 AM, Blogger Phil Birnbaum said...

Checking whether 2005 is causing this to happen ...

Leaving out 2005, which was 11-5, the seasons 1996-2004 were 75/132, or 56.8%.

So, nope, it really isn't just an anomaly over the past two years!

At Sunday, September 16, 2007 10:42:00 PM, Blogger Brian Burke said...

I you go with 6 instead of 6.5 for the overs, you'd improve to 61% correct over the '96-'05 period.

At Sunday, December 30, 2007 12:13:00 PM, Blogger Brian Burke said...

Update for 2007. The system went 8-3-1 (assuming SD beats OAK).

Plus more theory on why this happens.


Post a Comment

<< Home