The phenomenon of home advantage in professional American sports is well-documented. A team playing at their home stadium has a greater chance of winning than they otherwise would. In the NBA specifically, the home team has won approximately 58% of regular season games in the past five seasons. Every team in the league has put up a higher win percentage at home than on the road during this span.
The Phoenix Suns exhibited the smallest difference in their win percentage, yet their home winning percentage is still 7% higher than their away winning percentage. That’s certainly not an insignificant difference.
The existence of home-court advantage is undeniable. But what causes it? Well, there are multiple possible explanations.
This study from the University of Pennsylvania found that home-court advantage in the NBA is partially explained by reduced rest for the home team. Less time to rest between games means the players will be more fatigued and likely perform worse. That makes sense.
Maybe there’s a psychological component to it. Some observers have theorized that players expect to have a worse chance at winning on the road which causes them to play poorly as a result. This is known as a self-fulfilling prophecy. In 1951, Minneapolis Lakers guard Kevin O’Shea described the presence of this mentality in some of his teammates: “It wasn’t worth it — you weren’t going to win on the road. You could win some, but you’d have to beat the hell out of that team to do it.”1 Granted, home teams won 74.9% of their games in 1951, so this mentality was probably more common than it is now (if it even exists now). Still, it’s a possible factor.
There’s another way that psychology can have an impact: referee bias. As we all know, referees often make mistakes. After all, they’re human. It’s not absurd to think that may have slight subconscious bias for the home team. I wouldn’t be surprised if referees were influenced by the twenty thousand fans surrounding them who could raise hell at any moment.
There are certain calls that come down to referee judgement. These are the plays in which any bias would likely be most pronounced. Take the shooting foul, for example. A shooting foul is most likely to occur when a defender is attempting to contest or block a shot. We can try to detect a referee bias by looking at the difference in the block/shooting foul ratio between home and away teams.
In every season over the past decade, the ratio of blocks to shooting fouls is smaller for away teams than for home teams. This suggests that the away team tends to get called for shooting fouls more often than the home team.
Another judgement call is the traveling violation. It’s quite difficult to identify travels in real-time, so it’s a situation where bias can play a significant part. Let’s see if the home-court advantage is once again present.
Yep. Throughout the entire decade, travels were more likely to be called on the away team than on the home team.
That’s two judgement calls which are clearly effected by home-court advantage. Of course, there’s one slight problem. It’s technically possible that these discrepancies are not due to biased officiating. Maybe players are just worse on the road and they therefore commit more traveling violations and shooting fouls than they otherwise would. One way to test if this might be the case would be to see if the away team commits more objective fouls as well. The shot clock can’t be biased, so let’s see if the away team is significantly more likely to commit shot clock violations. If they are, then the previous discrepancies are probably not meaningful.
The away team does seem to be a bit more likely to commit shot clock violations, although the disparity is not as large as the two judgement calls. Is it large enough, though? I ran a t-test on the data from the last three graphs and came up with the following p-values:
block/shooting foul ratio: p-value = 0.00045296164933589965 traveling violation: p-value = 0.04804402387699467 shot clock violation: p-value = 0.20863516568666624
There is a statistically significant difference in the home / away data for the two judgement calls, but not the objective call. Just as we would expect.
So far, it does appear that there is a referee bias for the home team in NBA games. However, I think there’s one more place to find evidence. The NBA recently began releasing something called “Last Two Minute Reports.” The L2M report is described by the league as “the league’s assessment of officiated events that occurred in the last two minutes of games that were at or within three points during any point in the last two minutes of the fourth quarter (and overtime, where applicable).” It’s essentially a transparent report card for the referees in the last two minutes of tight games. So, the sample size isn’t huge, but it’s another way for us to see if there’s a discrepancy in officiating favoring the home team.
In the L2M report, plays are either categorized as correct calls, incorrect calls, incorrect no calls, or incorrect calls. An analysis of the plays found that:
The data from the L2M reports suggests that NBA referees tend to incorrectly call fouls on the away team. I used a t-test to find that the disparity is statistically significant.
Previously, I also found that the discrepancy between the frequency of two judgement fouls for home and away teams was statistically significant.
Therefore, I think it’s fair to conclude that there is indeed an officiating bias in the NBA that benefits the home team. Is this bias unintentional? Probably, I’m not sure why the league would want it to exist. It’s also worth nothing that the graphs I posted seemed to suggest a considerable decrease in the disparity between the officiating for home and away teams over the past decade. The difference was far higher in 2010 than it was in 2019. Maybe officiating is actually improving! Imagine that!