A lot has been said about the effects of the Minnesota Vikings moving to a new stadium, from losing familiarity to selecting a strong-armed quarterback to deal with the wind. The wisdom of choosing a ten-year franchise quarterback based on a two-year stint aside, most concerns are seemingly overblown—the Vikings should be slightly better, maybe, if anything.
According to Chase Stuart at Football Perspective, teams have been one percent better when moving to a new stadium. In his set of 55 teams who have moved from one stadium to another (either permanently or temporarily), they’ve improved their record from 47.2 percent to 48.7 percent.
I had three initial reactions to this:
- So obviously, there’s no effect
- The fact that the record before moving to a stadium was less than 50 percent implies that the sample size is small (which we know to be true anyway)
- The fact that the record before moving to a stadium was less than 50 percent implies that improvement is more likely regression to the mean.
It does, however, fit another study done by Jason Lisk (now at the Big Lead) that specifically looked at home field advantage (against away record) when moving to a new stadium. Depending on what in your opinion causes home field advantage, the results are what you’d expect.
Home field advantage in the NFL is about +16% for a generic team—they win at a rate 16% higher at home than they do on the road. For teams in new stadiums, the first year is slightly higher than expected (consistent with Chase Stuart’s research) but perhaps not meaningfully so at +19.2%. The second year in the new stadium, however, is +22.7%—and it’s not because teams somehow get worse on the road (that record did not meaningfully change).
That difference is not all that extraordinary; an 8-8 team gets pushed to an 8.44-7.56 team. But teams that have a tendency to get into close games (say, like the Vikings last year) could use that additional push to grab a win they otherwise would not have. Using Game Scripts, which calculate the average point differential over every second of every game (and therefore doing a lot to eliminate the effects of garbage time), the Vikings’ “true ability” was that of a 6.5-9.5 team, not the 5.5-10.5 team they really were. Having a small push here or there would have very likely improved their record and the home field advantage in 2015 may make a team like last year’s Vikings into a 7-9 team.
For the most part, this is largely trivia. The fact that teams are better at home is not news, nor is it particularly shocking to learn that teams who visit more often have a smaller road penalty. The familiarity/comfort hypothesis for home field advantage carries a lot of circumstantial evidence in its favor, including the fact that teams that visit more often do better than teams that don’t. Divisional games only carry a 1.8 point penalty for road teams (vs. 3.1 points for non-divisional opponents) and home teams win divisional games 54 percent of the time instead of the 59.5 percent of the time they win against non-divisional opponents. First-time visitors to a stadium do worse than their spread projections by over a point.
The familiarity hypothesis also carries weight because home field advantage is strongest in the first quarter (of multiple sports) and decreases as the game goes on.
The Vikings, in theory, will not be all that familiar with TCF Bank Stadium in their first year, but will be somewhat more familiar than their opponents (aside from, I suppose, the Chicago Bears—meaning Corey Wootton should feel right at home). In their second year at the Bank (I swear I’ve heard that nickname for it before), they will be significantly more familiar with the stadium than the majority of their opponents and should have a more pronounced advantage.
HFA likely comes from a number of things, though I do think familiarity is the biggest cause—things like crowd noise, evolutionary biology, travel time/distance, climate and so forth do not have a large effect (though some do have a significant effect, it’s simply small) or are difficult to test. The most popular thesis, referee bias, usually relies on data (penalty differential) that does not extract opponent performance from actual bias (teams who are behind or players who are uncomfortable may be more likely to commit penalties) and doesn’t hold up to high-sample scrutiny—except in environments where referees may actually fear for their personal safety.
Regardless, it should be fun to see warm-weather teams visit during the winter. The only one on the schedule in 2014 seems to be the Carolina Panthers, on November 30th (they play the Dolphins in Miami).