THE BOOK--Playing The Percentages In Baseball

Speaking of home field advantage, are you aware of studies that show how the home field advantage/road disadvantage varies over a series or homestand/road trip?

I’ve done a lot of research over the years on HFA, although not lately.

There seems to be a fairly constant wp multiplier of around 1.07 (1.08 in the old days).  So if you are a .500 team, your wp at home is around .535, but if you are a .600 team, you will win around .642 at home.  That is by no mean exact, but certainly the differential is not constant.

Also, teams do not appear to have any special talent associated with the team or park for an extra HFA, other than the following:

Boston has a little extra home edge historically because of the wall.  The home team gets more doubles off the wall primarily because the road team plays the caroms less well.

The Astros in the Astrodome had a very large extra HFA, presumable because the lighting was bad and the road team was not used to it.

MIN has a slightly higher advantage at home because of the ceiling of the Dome.  The road team has a few extra balls get lost in the Dome.

The old cookie cutter stadiums had a slightly less HFA, presumably because of the symmetrical outfields.

Teams with lots of nooks and crannies in the OF (like TEX) tend to have a slightly higher HFA because they get more triples (in those nooks and crannies) than the road teams.

And of course the Rockies have had a huge HFA although that might be changing since the humidor.  The reason is a combination of playing particularly poor on the road (much poorer than a team would play on the road if they didn’t play for the Rockies), the co-called Rockies hangover effect, and the visiting team playing poorly in Coors, because they are not used to the altitude and the funky things that the ball does.

As far as how HFA varies over the course of a series, it is hard to pick up small effects even over many years, because of sample size problems, but it appears, and makes intuitive sense, that the HFA is greatest on the first day of any series (presumably, the road team is not used to the park) and diminishes as the series goes on.  The effect is about .7% (for the home team, over and above the normal HFA) in the first game of the series, nothing in the second game, .5% for the road team in the third game and almost 1% for the road team in games after that.

It appears to have nothing to do with time on the road as a team that is on a road trip keeps losing and then gaining advanatge as they go to each city.  It could be the actual travel itself, or more likely, as I said, familiarity with the park.

Regarding the domes, I had read that another theory about their exceptional HFAs being that the crowd noise there was much louder than at other stadiums.  Could explain part of the increased number of errors you’d see at a place like the Metrodome: distraction.

First of all, they have a very small extra HFA that has diminshed to almost nothing (again, as far as we can tell, given the sample size issues.  From what I recall from my research, it shows up as lightly more doubles for the home team.  The best explanation is lost balls off the dome ceiling and the reason for the diminishment is the Metrodome being around for so long and the unbalanced schedule.

If you’re so certain about this, I’d suggest publishing.  HFA research in journals is sorely lacking - both in frequency of articles and in actual conclusions.  I wrote my undergrad thesis on familiarity in the HFA, and there were maybe 25-30 articles, tops, that I could cite.

One of the biggest components of HFA is a higher K/BB ratio.  I think it would be great if someone used the pitch/fx data to disaggregate that into:
1) home pitchers actually throwing more strikes/fewer balls;
2) home hitters doing a better job of taking balls, swinging at strikes
3) umps giving an edge on called pitches to home team.

Guy is right on.  When you look at the home/away splits, the K, BB numbers is where it stands out.

Yes, the pitch-f/x data would help a lot.  The K, BB, and triples rates are the main components of HFA.

I don’t write for Academic Journals, BTW.  I don’t think I would qualify and have little interest anyway.

To #6 & #7:

IMHO, one of the reasons is home hitters are visually more accustomed to their own hitter’s backgrounds, thus they could have better strike zone judgment.

Ty, great point.  It’s possible that you have different K/BB home advantage based on the familiglia of hitters.

It’s my opinion that using these one-size-fits-all adjustments, while better than nothing, makes it too easy for us to keep looking for answers.  To that end, I’d actually prefer no adjustments.  I’ll pay the price in the short-term, but it will motivate people in the long-term.