The NBL’s home edge has halved
For four decades the home team started each game about four and a half points ahead. Since 2020 it is about two, and the free throws went first. The league also scheduled out the back-to-back, which cost two points, and swapped it for four times the travel, which barely matters.
The edge has halved and is not coming back. About 4.5 points a game before 2020, 2.0 since, and 2.1 since the crowds returned. Every one of 200 resamples shows a decline.
The whistle left. Home teams earned 2.2 extra free-throw attempts per 100 possessions before 2020; since crowds returned the gap is 0.0. Offensive rebounding, an effort play, recovered.
The back-to-back is gone; the flight barely matters. A back-to-back cost 2.1 points and none has been scheduled since 2016-17. Teams now travel about four times as far, with no clear effect on results.
01 · The edge
Four and a half points for forty years, two since 2020
From the league's first season in 1979 to 2019-20 the home team won 61.5% of regular-season games and started each one about 4.5 points ahead once team strength is allowed for. Since 2020-21 the home team has won 54.6% and the edge is worth 2.0 points (range 0.9 to 3.2). The obvious explanation is COVID, but crowds came back in 2022-23 and the edge did not: it is 2.1 points since (range 0.9 to 3.3).
| Era | Games | Home won | Home edge (pts) | 95% range |
|---|---|---|---|---|
| 1979–89 | 1,765 | 61.3% | 5.1 | 4.0 to 6.0 |
| 1990–99 | 1,871 | 59.8% | 3.8 | 3.0 to 4.6 |
| 2000–09 | 1,686 | 64.0% | 5.0 | 4.2 to 5.8 |
| 2010–19 | 1,160 | 61.1% | 4.1 | 3.3 to 4.8 |
| 2020 onward | 846 | 54.6% | 2.0 | 0.9 to 3.2 |
| Crowds back (2022–23 onward) | 597 | 54.9% | 2.1 | 0.9 to 3.3 |
Single seasons have dipped before, in the late 1990s and 2011-12, and recovered within a year or two. The 2020s are the first sustained stretch below three points. A model that still gives the home team the long-run 4.5 points did worse on the games since 2024-25 (error 14.44) than one that gives it nothing (14.14).
02 · Where it went
The calls left, the effort stayed
Before 2020, home teams earned 2.2 more free-throw attempts per 100 possessions than their opponents and committed 1.6 fewer turnovers. The free-throw gap went to +0.2 in the COVID seasons and 0.0 since crowds returned. The turnover gap is down to 0.2. Together those two gaps are worth roughly 2.5 points a game, close to the whole fall in the home edge in the box-score era.
The effort plays tell a different story. The home offensive-rebounding edge dipped during COVID (−0.1) and came back with the fans (+1.2); the shooting edge moved from +1.0 to +0.5, within the range of chance. What did not come back is the whistle. Crowd pressure on officials is a common explanation for home advantage in sport, and the NBL's recent history fits it: whatever changed in officiating during COVID seems to have lasted.
Travel was never the main source
When visitors travel 2,500 km or more, which means anything involving Perth or New Zealand, the home edge is only +0.4 points bigger than after a normal interstate trip (range −0.6 to +1.2). Derbies are where distance shows: in same-city games the edge is 1.7 points smaller. Perth is a genuinely hard place to win, with +2.2 points a game more edge than the league since 2010-11, but New Zealand is just as remote and gets far less. The Sydney Kings have the smallest home edge of any club (−2.9).
Crowds do not explain it either. Games with bigger-than-usual crowds produced a smaller edge, not a larger one, though marquee opponents bring their own fans. And the edge belongs to the place, not the fixture: in 93 listed home games played outside the home team's state or country, the nominal home team won 48.4%, with an edge of +0.6 points (range −2.1 to +3.0).
03 · The schedule
The league killed the back-to-back and swapped it for the flight
Playing the day after another game cost a team 2.1 points against what the ratings expected (range −2.9 to −1.2, 1,620 team-games). One day off was enough to remove it. When one team was on a back-to-back and the other had two or more days, the tired team won 36.9% of 1,381 games. The penalty grew as the game got faster, from 1.1 points before 2000 to 3.5 in the last seasons before the league stopped scheduling them.
In 1979, 29.4% of team-games came the day after another game. The last back-to-back was in 2015-16; since 2016-17 there have been none. In their place, teams now travel about 1,489 km between games, against 373 in 1979, as the league moved to one club per city from Perth to Auckland. That cost is hard to find in results: distance and time zones do not produce effects clearly different from zero.
The one modern schedule effect worth watching is the long break. Since 2016-17, teams coming off a week or more without a game, usually the international windows, did 1.9 points worse than expected (range −4.2 to 0.2, 255 team-games). That may say more about rhythm than fatigue.
What this does not show
- It is not a causal test. The data shows the free-throw and turnover gaps closing, not why. The officiating explanation fits the pattern, but it is an inference.
- The latest season is noisy. The 2025-26 edge was 0.8 points, with a range of −1.3 to 3.0. One season cannot confirm a trend; the era averages can.
- Some groups are too small to judge. Only 45 games were played in front of fewer than 1,000 people, and 93 "home" games were played away from home, so those estimates are wide.
- Club estimates are shrunk. Each club's extra home edge is pulled toward the league figure by how few games it has played, so small clubs with few games cannot show large effects.
The interactive version
Every chart, the full method and the player and club tables are in the NBL Tracker app that this research comes from.
General information only. Nothing here is betting, gambling, financial or professional advice.