Is There Really a "Home-Field Advantage?"

One of the reasons that Steeler Nation is so glad about getting the second seed in the AFC is that not only do we get a bye, which we critically need, but we get all of our games at home, at least unless and until we have to play the Patriots.  As everybody knows, it's a big advantage to be able to play in your home field, right?  It just makes sense. You are surrounded by your loyal fans.  You know the field better than anyone else.  You have all the comforts of home, so to speak, unlike what you might get as a visiting team.  (Proper network feed with full replays, anyone?)  Seems like a no-brainer.

But an off-hand remark in one of Michael's posts the other day got me to thinking about whether the numbers really support the idea of a home-field advantage.  Maybe it's one of those things that used to be much more of a factor than it is in recent years.  So I decided to look at the numbers for the past three seasons and see what we find, after the jump.

The numbers are in the table below.  Obviously, "W/L" is win/lose, "H/A" is home/away wins only, and the Cumulative columns are the total number of Home and Away wins for all three seasons, and finally the total number of wins during the three seasons.

Now for some analysis.  (I would put it below the table, but I can't seem to do that - oh well...) 

3 year analysis:

26 teams had more home wins than away wins.

3 teams had equal wins at home and away. All were under .500.

3 teams had more wins away than at home - (2 better than .500)

4 teams had more than 2x as many wins at home as away - (all finished under .500 for the 3 years.)

2010 analysis:

20 teams had more home wins than away wins.

2 teams had equal home/away wins, 1 +.500

10 teams had more away than home wins; 6 had winning records

9 teams had more than 2x as many wins at home as away. Only 1 had a winning record.

So what can we conclude from all of this? Not a lot, as it turns out, but here are a few things that caught my eye:

If you analyze any single year the number of home wins is considerably less than the 26 home wins one comes up with over three years, which just goes to show the power of averaging, but the figure is always 20 home wins or more per season, which indicates a weak but definite home field advantage. I would have expected to find more correlation between a dome team and home field advantage, but it doesn't appear to be a stronger correlation than for an outdoor team. One would have to go a lot farther back to determine whether there is any change over the years in the amount of "home-field advantage." I would be interested to hear whether someone had crunched the numbers for a larger sample size.

The most curious correlation for these three seasons is that between winning more than twice as many games at home as on the road and not being very good. In 2008, 4 teams fell into this category, none with a winning record - 2 teams had losing records, 2 had .500 records. In 2009, 7 teams won at home more than twice as often as on the road, and only a single team had a winning record - one had a .500 record, and 5 had losing records. This season 9 teams won over twice as many times at home as on the road, and all but 1 had a losing record.

Therefore I suppose we could say that there is a strong correlation between winning substantially more at home than on the road and a losing record. Makes sense, I suppose, because if you struggle too much on the road you're going to lose quite a few games, even if you're mostly winning at home. A good example is the '09 Patriots, with 8 home wins and 2 road wins. (That is also the biggest differential, if I'm reading my figures correctly.) They were one and done in the playoffs, which isn't too surprising, as they had to play on the road.

If you look at the 3-year totals, only 3 teams won more than twice as many games at home as on the road over that time period, and none had a winning record over three years.

And finally, we can note that over this time period only 6 teams - less than 13% - won 32 or more out of 48 games (in other words, 2/3 or more of their games.)  One of those teams is of course our own Steelers. Of those 6 teams, 4 of them are in the AFC, and 2 of those are in the AFC North. Which says a lot about our division.