Tuesday, December 30, 2008

Trying to Predict the Playoffs? Forget About Records

by Josh Fyman

A cliché that players, especially those on wild-card teams like to dredge up this time of year is that records are no longer relevant, that everybody is now 0-0. The whole idea sounds like bland motivational speak. After all, wouldn’t it stand to reason that teams with superior records would be more likely victors in playoff games? Well, I’m here to tell you that this is only slightly true. In fact, the cliché that all records are dialed back to zero has more credence when one looks at the numbers.

The first, most rudimentary analysis would be to correlate a team’s winning percentage in the playoffs with their records during the regular season. Since the abbreviated playoffs provide a small sample size, I went back five years and correlated the regular season winning percentages of all playoff teams with their winning percentages in the playoffs. The results are illuminating. There is only a 0.14 correlation between regular-season record and playoff record, which is barely significant. Granted that the data are skewed somewhat by the fact that teams that earn a bye do not get to accumulate as many wins, and therefore, suffer a worse playoff winning percentage in the event that they lose. This problem is erased, however, by comparing wild-card teams to each other and bye teams with each other. When examining only bye teams, the correlation between regular season and postseason winning percentage actually drops to 0.10, which is not significant.

Another school of thought is that the teams that finish the hottest are best suited for playoff success. To test this theory, I correlated playoff teams’ December records over the past five years with their playoff success. Turns out that there is only a 0.18 correlation between teams’ performances at the end of the season and their performances in the postseason. Bye teams’ performances are a little more predictive of their playoff performances, with a correlation of 0.23, but that still is not too much.

To drive home the point, over the past five years, in playoff games in which two teams with different records played each other, do you know the likelihood of the team with the better record winning? Ready? A whopping 51.8% probability. So, if you’re picking playoff games based on which team has the better record, you’re just about as well of flipping a coin.

Not only do teams with better records fare only slightly better in playoff games, but it doesn’t even much matter how much better one team’s record is than the other. In a correlational analysis, there was no significant relationship between how many games separated the two opposing teams’ records and the probability of an upset (upset being defined here as the worse record beating the better record). In short, a game with records separated by one game was no more likely to result in an upset than a game with records separated by three, or even four, games.

Is there no way to predict playoff games with some accuracy? Of course there are models that will provide results better than 50/50. Looking at records, however, offers no help whatsoever.


robert said...

Re: the low correlations. This year the 12 playoff teams average about a 68% win rate. Four teams will have a 0% win rate after the wild card weekend. And the 12 teams will combine for a 50% win rate in the playoffs. Doesn't that guarantee low correlations? Is there any combination of playoff wins and losses that would produce a correlation of 1?

I think correlation is not the right tool to prove your point.

The Football Know-it-all said...

You bring up some interesting points, and I concede that there are limitations to strictly using the first analysis as a proof. That's why I mention that the first couple paragraphs are simply the first step in answering the question. However, in the final two paragraphs, points are made that compensate for this weakness.

For one, the percentage of teams with better records winning could very well be 100%. Finally, in the last paragraph, there is little range restriction in the correlation of difference in wins and outcome of the game. Each game was coded dichotomously as either an upset or a projected outcome. If the probability of an upset decreased substantially as the win differential increased, the correlation would have been much higher than 0.10.

robert said...

My point is that the structure of the playoffs is so different from the regular season that correlation analysis is not justified. I don't know enough statistics to know the best way to analyze the playoffs and regular season records. Simple counting and cross tabs may be a start.

There have been 88 playoff games since the 2000 season. Of those, teams had the same number of regular season wins in 16 games. Of the 72 games were a team had a edge in regular season wins, the team with the better regular season win total won 47 games, or about 65%. Is is more or less the same win percent as teams with better records in the regular season.

Home field advantage is a confounding element because it is based on playoff seeding, which is tied closely to regular season records. Home teams won 52 of 80 playoff games, or 65%. That is higher than the regular season home team win rate of about 60%.

The win rate of 65% for teams with a better record and the 65% home team win rate suggest that regular season records do mean something in the playoffs.

jjbtnw said...

In the last 3 years (2005-2007), of the 6 Superbowl teams, 3 came off the bye, one division winner, and two wild-cards. That's exactly what you would expect if regular season records didn't mean anything in the playoffs.
But if you look at the 5 years before that, 2000 - 2004, of the 10 Superbowl teams, 8 came off the bye. The other two were division winners. No wild-cards at all.
If you ask me, the last 8 years are too small a sample set from which to draw meaningful conclusions.

Brian Burke said...

Tiny correction--In 2000, Baltimore was a wildcard. Tennessee won the AFC Central that year. But the Ravens did host the wildcard game against Denver.

Elias said...

Very interesting article. I'll have to out this process in place and see if I can parlay this newfound knowledge into a little extra cash. Of course I'll be watching the games and results in crystal clear HD. As a customer and employee of DISH, I have access to the most HD channels in the industry, and this weekend they'll all be tuned to Playoff action!

Post a Comment