A dated snapshot. Current numbers are on the live report, and later model changes go in the changelog. Corrections are at the end.
Injuries have cost NBA teams 17.4 wins this season by my count, and nobody has paid more than the Warriors. The 1-15 Bucks have paid almost nothing, which tells you something about the Bucks.
This is the first Injury Luck Index. Every week I add up what injuries have taken from each team: what each missing player costs on the nights he sits, after the teammates who take his minutes, turned into wins. A star counts for a lot; a 12th man barely counts. The live report has all 30 teams. Data through Nov. 19.
Most hurt
| Team | Record | Wins lost to injury | Player-games missed |
|---|---|---|---|
| Warriors | 8-8 | 2.6 | 52 |
| Raptors | 5-10 | 2.4 | 41 |
| Thunder | 9-6 | 1.9 | 72 |
| Hornets | 4-10 | 1.4 | 44 |
| Pacers | 5-10 | 1.3 | 63 |
The Warriors are the story. Jimmy Butler III hasn't played this season, and Stephen Curry has missed the last 4 games with a right oblique tear. At full strength, my team rating expects them to beat an average team by 6.0 points. With today's injury list, it's 1.0. No team is missing more talent right now, and 8-8 looks a lot better with that in mind.
Kawhi Leonard is the Raptors' season so far. Toronto won both games he played and is 3-10 without him. By my numbers his absence costs about 4.7 points a game, more than any other player out right now. ESPN has him back around Dec. 3.
The Thunder have missed the most games and keep winning anyway. Their 72 player-games missed lead the league, with Chet Holmgren, Jalen Williams and Cason Wallace out 12 games each. That's cost them 1.9 wins, third most, and they're still 9-6. Healthy, they'd have the third-best team rating in the league. ESPN has Holmgren back around Jan. 2.
Least hurt
The Knicks have missed 7 player-games all season, the fewest in the league, and the Celtics 9. Neither has an injury excuse, and neither needs one.
Then there are the Bucks: 1-15, with 10 player-games missed and about a tenth of a win lost to injury. Whatever is wrong in Milwaukee, it isn't the injury report.
The Spurs
I'll admit I checked this one first. Injuries have cost San Antonio less than a tenth of a win, and Dylan Harper's hip strain shouldn't change that much. He's worth 4.1 points a game, 21st in the league by my numbers, but the Spurs' guards cover for him so well that his absence costs about 0.4. ESPN has him back around Dec. 13, and the Spurs page tracks it game by game.
How it's measured: wins lost are each injured player's cost, after the teammates who covered his minutes, for every game he missed, at Basketball-Reference's standard 2.7 wins per point of margin over 82 games. Only injuries and illnesses count, from the NBA's official injury report; "injury management" counts only in runs of 3 or more games. Full methodology.
Correction, Nov. 20: the first version of this post counted each injured player's full worth as lost, which overstates what absences cost because teams adapt. On the corrected basis the league total is 16.4 wins, not 49.5, and the Warriors lead with 2.5, not 6. Two of Stephen Curry's missed games were load-management nights, which no longer count as injuries. The order at the top barely changed. Details are in the corrections log.
Update, late Nov. 20: model improvements the same night (box-score weights learned from on-court results, blowouts capped in team ratings, a test against Vegas) moved the numbers again. The league total is now 17.4 wins and the Warriors 2.6; the Thunder rose to third and the Pacers entered the top five.
Update, evening of Nov. 20: rewritten in a plainer voice, and the evening's data refresh is in: the Raptors' wins lost moved from 2.3 to 2.4, and ESPN's estimate for Kawhi Leonard's return moved from Nov. 27 to Dec. 3.