Football picksDecember 17, 2008

The NFL’s (Un)Luckiest Teams

We enter week 16 of the 2008 NFL season with four teams having clinched their division including the Tennessee Titans (12-2), New York Giants (11-3), Pittsburgh Steelers (11-3) and the Arizona Cardinals (at just 8-6). Another division leader, Denver is very close to clinching their division and they are also just 8-6 on the year. Meanwhile, there’s a very good likelihood that the Patriots will miss the playoffs, despite potentially finishing 11-5. Only once before has an 11-5 team missed the playoffs, and that was back when there was only one Wild-Card team.

Knowing that two 8-6 teams, with potential 8-8 finishes on the horizon, will make the playoffs, while teams with much better records will be watching from home, makes me wonder. How much is luck a factor in the NFL? What are the implications? It turns out that luck actually plays a very big role. Just ask any sports bettor who loses a big bet thanks to the bounce of a ball, or a field goal hitting an upright. Given that NFL coaches and players will lose their jobs (or lock in contract extensions), based on the results of the season, why isn’t luck talked about more?

It’s easy for the TV and radio analysts to call the Titans, at 11-3, great. It’s as easy for them to call the Lions, at 0-16, terrible. While certainly there is big difference between these two teams, how much has luck played a role in Tennessee’s success, and in Detroit’s failure? It turns out, quite a bit.

While win-loss records are helpful, they are full of “noise.” A 21-point win counts the same as a 1 point win that only came thanks to a bad call by an official. One of the best ways to cut through the noise is to look at something called Pythagorean wins. Pyth wins aims to come up with a “pure” view of wins and losses – what a team’s record “should” be based on points scored and allowed. It was made famous originally by Bill James for use in analyzing baseball teams. But, we can use it for any sport, including football.

Back to Tennessee and Detroit. Based on Pythagorean wins, Tennessee’s “pure” record would show 1.2 wins lower than it the actual record (i.e. they should be 10-4). Detroit is bad, but not 0-16 bad. They really should have 2.6 wins on the season, sitting at 3-13.

What other teams have been (un)lucky? Let’s have a look at some of the extremes…

Denver is the league’s luckiest team based on their point totals. The Broncos should really be 6-8 instead of 8-6. This is not surprising to anyone who saw them win several extremely close games earlier this year, including a gift from Ed Hoculi in the San Diego game. After the Broncos, the Panthers’ and Colts’ records are measurably better than they should be, both having won 1.7 games more than they really should have with luck removed. The resurgent Dolphins have won 1.6 more games than their point totals would indicate. These four teams are likely overrated in the minds of the press, and betting public. Keep that in mind moving forward.

Who’s the unluckiest team in the league? The Green Bay Packers have posted a record of 5-9, but their point totals indicate they should be closer to 8-6. Their fall from grace since last season has been exacerbated by the fact that they were one of the league’s luckiest teams in 2007. While they went 13-3, their point totals indicate that the Pack should have really been 10-6. Maybe that loss to the Giants in the NFC Championship wasn’t so surprising after all. So, are the Packers really that much worse this year as compared to last? A fall from 10-6 to 8-6 would indicate not so much. So, Packer fan, there is good news and bad news. The good news is that they really haven’t gotten that much worse with the switch from Favre to Rodgers. They are a solid team with both QBs. The bad news is, they were never really as good as it appeared in 2007.

Detroit is the next unluckiest team in the NFL thus far this season, at 2.6 wins below their “pure” win rate. San Diego is 6-8 but should really be 8-6 and Kansas City should be 4-10 instead of 2-12.

I know… Some of you are saying to yourself that I’m full of it. “A team’s record shouldn’t be anything other than what it is! If a team can’t win games they should, then that is on them – they deserve what they get.” Fair enough. Luck is a part of sports – always has been and always will be.

But, regardless of how you view this data, you can put it to some use when making your NFL picks. Consider that the lucky teams mentioned above are likely overrated and will be laying more points than they should in the closing weeks of the regular season and the playoffs. The unlucky teams might be relatively good bets as they are likely to get more points than they should be getting. And, if you happen to find a very lucky team playing a very unlucky team, you might find exceptional value backing the unlucky team (likely as an underdog). We happen to have just this case coming up at the very end of the season when Denver plays San Diego. Keep an eye on that one.