I recently did some research and number crunching on the effect of home-court advantage (HCA) in the NBA. Originally, I wanted to see if I could find any particular teams that had an especially strong (or weak) HCA. I came across some interesting things – and they appear to be part of the explanation of HCA variance in the NBA.
I started by taking every team’s point differential through the first three quarters (PD3Q) of every game over the last five seasons. Though this sounds ridiculous, it actually makes a lot of sense.
Point differential through the first three quarters is a better stat than just point differential. This is because, through the first three quarters of an NBA game, both teams will be playing their standard, point-maximizing style – whatever that may be. Teams will then modify their strategies in the later portions of the game – down by 10 with 3 minutes to go, down by 5 with 20 seconds, etc. – to increase scoring variance. The increasing variance makes usable statistics – like point differential – somewhat misleading. To avoid using muddied, semi-ambiguous statistics, I opted for PD3Q, which discards fourth-quarter statistics entirely. This serves as a better indicator of a team’s true overall ability.
I then subtracted every team’s away PD3Q from their home PD3Q to get a single number – the five-year PD3Q difference between home and away. The teams that scored the highest have the strongest HCA, and vice versa.
The teams with the strongest HCA: Utah (+9.02), Indiana (+8.06), Portland (+7.38), LA Lakers (+7.02), Denver (+7.00), San Antonio (+6.58), Toronto (+6.20), Golden State (+6.20), Phoenix (+6.14), and Oklahoma City (+6.06). Those 10 teams are the only 10 in the NBA that have +6.00 or greater difference between home and away PD3Q. The league average is +5.20.
The teams with the weakest HCA effect: Chicago (+2.40), Milwaukee (+2.54), Brooklyn (+2.60), Minnesota (+2.84), Philadelphia (+3.14), and New Orleans (+3.34). Those are the only six teams with a lower than +4.00 difference between home and away PD3Q. In other words, Chicago outscores opponents by 3.2 more ppg (extrapolated) at home than on the road, while Utah outscores opponents by a whopping 12 more ppg at home than on the road. There is clearly a drastic difference in HCAs across the NBA.
More importantly, there appears to be a range of degrees of HCA, contrary to the standard practice of assuming every HCA is roughly the same. Fascinating in itself, the range of HCAs needed to be explained. So far, I’ve come up with two explanations for the varying HCAs throughout the NBA.
The first is travel distance. Intuitively, teams that are forced to travel long distances are much more effective playing at home than on the road. Teams from the West Coast and relatively remote cities are forced to travel much further on average than other teams, giving them more pronounced home/away differences. Of the top-10 teams with the strongest HCAs, only two – San Antonio and Indiana – are from NBA-dense regions. And of the top-six weakest HCAs, every team is from an NBA-dense region.
The second explanation, which only applies to a couple of teams, is altitude. Everyone knows that altitude affects athleticism. The lack of oxygen at high altitudes limits athletic ability. Athletes acclimated to high altitudes have a significant advantage over those not. Both Utah and Denver play their home games at 4,500+ feet, making them prime candidates for teams that should have a substantial difference between home and away stats. Indeed, they do: Utah and Denver are ranked first and fourth, respectively, in terms of HCA.
There is the possibility that Utah and Denver have strong HCAs because of their remote locations – and thus longer average travel distances. Though I can’t rule it out, the idea that long travel distances are the sole reason for Utah and Denver having such strong HCAs is far-fetched. Long travel distances may compliment the altitudinal effect of Utah and Denver, but they’re far from the sole cause. And for every other team, travel distance will have to suffice as the primary explanation of HCA variance.
Unfortunately, this information does little to explain HCA per se. Some studies have suggested that the main source of HCA is referee bias – the unconscious favoritism of referees to the home team – while others have proposed that “Hawk (home) and Dove (away)” game theory may hold the answer.
Regardless, the variance in HCAs across the NBA is explained primarily by travel distance and, in the cases of Utah and Denver, altitude. These results have serious implications for predicting sports. They entail, for example, that the Karl Malone-John Stockton Jazz teams of ’97 and ’98 held an unnoticed advantage in their title runs with HCA throughout the playoffs. Imagine if people had known how effective the Salt Lake City HCA was for the Jazz?
This intrinsic effect has always been present, dramatically altering the landscape of not just basketball, but all national sports with a home-away structure. And it hasn’t been noticed until now.