Study: NBA refs are biased towards home teams, underdogs

[mopper8]

Blog post over at The New Republic: (emphasis mine)

At The Study, though, we prefer our evidence in academic form, so we ask: are pro basketball referees biased?

Yes, say economists from BYU, Johns Hopkins and Oregon State, but not in favor of "bigger" teams. Using play-by-play data from ESPN for regular season and playoff basketball games from 2002-2008, the economists looked at "discretionary" turnovers (traveling violations, offensive fouls, and other turnovers caused by "active ref behavior") and fouls called against both teams, though they caution that the latter measure is "only suggestive, though still important." They found "evidence of three biases: favoritism of home teams, teams losing during games, and teams that are behind in a multi-game playoff series. All three biases are plausibly profit-enhancing for the league." The authors calculate that, during the regular season, the turnover biases "equates to win probability changing by approximately 2.2% when a team switches from away to home status," and a further 2.5% if fouls are included. In the playoffs, the biases do not appear to affect fouls, but the effect on turnovers becomes nearly doubles, keeping the probability change close to 5%.

Here is a link to a PDF of the study itself.

I just browsed it myself, but at first glance, the methodology seems sound, at least for the discretionary turnovers. I have not given it a lot of thought though; posting here because I'm curious if you folks agree with the underlying assumptions about how one could measure referee bias, if the methods are sound, etc etc. I know a number of posters on here who are better-equipped to handle that stuff than I am.

From their abstract (I thought this was worth highlighting):

We also find that the biases do not increase in situations where their direct financial benefit to the league would be greater, and conclude that the biases are likely of an implicit nature.

I'm also mulling over which cognitive biases could plausibly explain this, and would be interested to hear any arm-chair psychological explanations.

Basically, I'm adding little-to-no original content beyond the study itself, and am hoping to gleam something from the responses of those of you more intelligent than I.

[andre316]

This might be naive of me, but I imagine in 50-50 situations a ref defaulting to "well, the other team is up anyway, so I'll err on the side of caution", which in practice means shrinking a deficit and letting the teams continue to sort it out. Keeping a game competitive probably FEELS fairer than going the other way, all else being equal.

Not sure about home team bias, I wonder whether there's a subconscious crowd noise effect. Refs are still people, and we're wired to like hearing cheers and hate hearing boos, so maybe even a theoretically impartial ref can sometimes be influenced by the collective emotion of a home team rally, and make calls that elicit a positive response.

Then you have the "LeBron refs" who seem to like playing the role of villain...

[penbeast0]

I remember a few years ago (quite a few by some standards), the Philly stat guy who had been there forever used to publish stats comparing calls favoring the home team v. calls favoring the away team and found a distinct bias in favor of the home team with some refs well over 10% more likely to call a foul on the away team. The only ref who I remember having a bias in favor of the away team was Mendy Rudolph. The league ordered Philly to require him to stop publishing the data.

For what it is worth . . .

[floppymoose]

This might be naive of me, but I imagine in 50-50 situations a ref defaulting to "well, the other team is up anyway, so I'll err on the side of caution", which in practice means shrinking a deficit and letting the teams continue to sort it out. Keeping a game competitive probably FEELS fairer than going the other way, all else being equal.

Bavetta has been accused of this, and there is some statistical support for the claim. Personally, I don't really mind it if true. Obviously you want the game to be fair, but if it's a situation where you need to call a foul (huge contact) but it was a 50-50 call between charge and block, I can see the case for just taking the "refs are screwing us" arrow out of the quill of the losing team.

[Paydro70]

I recall reading a study that tried to get at the nature of home court advantage... as in why teams win more at home. The answer was "almost totally the refs."

At the Sloan conference there was also this:
http://www.sloansportsconference.com/pa ... advantage/
and then
http://www.sloansportsconference.com/pa ... analytics/

[mopper8]

I don't think refs alone can explain home court. For one thing, a decent percentage of the time re as play on the road, they're playing in a different time zone, which will mess up their body clocks. When the Knicks get into the 4th q against the Lakers in LA, it's past midnight by their body clock. That's a huge disadvantage. Sleeping in a hotel instead of at home....flying through the night...extended road trips through multiple time zones.

It just seems to me like there are far too many other things going on when NBA teams travel that affect the body that it seems like HCA goes beyond the refs almost prima facia.

[mopper8]

I'll check those videos tomorrow Paydro...thanks for the links

[floppymoose]

Some players have career home/away shooting % splits that suggest they shoot better in the home court. While I can't prove this isn't a side effect of the reffing, I can certainly understand how it might be easier to shoot on the familiar court, especially since there is a mental (or confidence) aspect to shooting.

I've seen too many guys hit wide open jumpers better at home than on the road to think it can be all the refs, even if it is mostly the refs.