Extra Points
News and commentary from around the Web

Home Field Advantage in the NFL

We've done research on home field advantage in the past at Football Outsiders, but I haven't looked at it for a few years. Here's some good research by Robby Greer that backs up some of the findings we've made in the past, and some other findings made by others more recently. Among them: home field advantage is lower now than it used to be, roughly 2.2 points instead of 3 points. Greer also finds that while individual teams might have periods of elevated home field advantage, there's nothing sustained or predictive about this. This is what I've found in the past, except for Seattle. Seattle was the one team that seemed to have a real larger home field advantage. But Greer finds this also isn't true, or at least hasn't been true the last few years. So perhaps no team truly has a larger sustained home field advantage than any other. The other interesting finding here is that home field advantage seems to be lower for divisional games. I may have to study that one myself in the offseason.

View Full Article


12 comments, Last at 27 Jan 2020, 11:57am

1 That's interesting, and…

That's interesting, and surprising to me, as I've had pretty good success in pick'em pools, and especially in DFS, by putting extra emphasis on HFA. That strategy has served me particularly well early in the season, when we really don't know for sure which teams and/or players are good or bad, but the one thing we do know for sure is who has HFA.

2 Tie Home Field Advantage to Referee, Not stadium

Its been proven that home-field advantage is due to the Referee's themselves: http://freakonomics.com/2011/12/18/football-freakonomics-how-advantageous-is-home-field-advantage-and-why/

Therefore, if we're trying to predict HFA, we should tie it to Ref, not stadium/team.

4 I don't know if such…

I don't know if such statistics are available, but I'd bet that HFA is a function of what % of attendees are fans of the home team. The two ends of that spectrum would be, say, a sea of cheeseheads at Lambeau, and... a sea of cheeseheads at Dignity Health Sports Park, formerly known as StubHub Center.

At some point, there will be drones that can zip around the stadium and and instantly classify every fan in the stands as "home", "away" or "neutral", but AI image processing ain't quite there yet...

5 How much of the decline in…

How much of the decline in HFA is due to the London Jaguars and SD and StL now playing 16 road games per year?

6 So much has changed over the…

So much has changed over the past decade for the NFL which off the top of my head include:

- more London games and at weird times

- more franchises moving to new homefields

- CBA restricting player practice time and coaching

- Thursday night expanded to full schedule in 2012

- rookie wage scale pushed QB salaries up leading to more player movement at other positions?


Interesting that Denver is no longer seen as having HFA. Given there is a known physiological advantage to playing at altitude, I question whether the methodology is sound. Comparing 8 home games to 8 away games seems like small sample size

9 London & Mexico City games…

London & Mexico City games should be excluded, as well as Chargers games from 2017-2019. I would also look at the results w/without Thursday games. I did not see how they defined HFA, simply winning or adjusted for relative strength (e.g. vs ELO prediction).

8 One of the hardest parts of…

One of the hardest parts of modeling HFA is - how do you separate worse performing offense vs better performing defense? Its a joint measurement problem.

10 Yeah, any study which…

Yeah, any study which porportedly shows that Denver has no advantage from practicing and playing at 5000 feet is one deserving of some skepticism.

11 In fact, another recent…

In fact, another recent study which measured performance compared to the betting line found that the Broncos had the greatest HFA in the league from (I think) 2006-2015. Not surprisingly, the Nuggets and Rockies were #1 in their leagues too, and the Utah Jazz were #2 in the NBA.


However, at least in football, it doesn't look like it's that significant of an advantage - look how closely bunched everyone is - and the study linked here by Aaron appears to use a much larger sample. And really, even in the NBA, where Denver and Utah appear to stand out more from the pack, their HFA only appears to result in +2 wins over the course of a season compared to the team with the worst HFA.

This isn't as rigorous, but the Broncos have a crazy good home W/L record in the month of September but I believe it's nothing special during the rest of the season. Maybe the benefits of training at altitude are most pronounced earlier in the season - away teams struggle more when they're not in as good of game shape yet, or perhaps the grind of the season and accumulation of injuries help to mute the altitude advantage.