Over the course of an NBA season, there are all sorts of events that take place both on and off the court that can affect each teams’ level of play. For example coaching changes, player disputes, contract negotiations, injuries, and trades all can have a profound impact on how well a team performs. The question I had about these events was this: Does the betting market accurately handicap these occurrences?
I decided that the easiest event to look at would be injuries. Its possible that losing a role player or a backup wouldn’t have a measurable impact on a team’s performance, but what about a leading scorer or an all-star? Say for example the Golden State Warriors are favored by an average of 10 points with a healthy Steph Curry. Say then, that Curry gets injured. The Warriors should certainly be favored by less without Steph Curry than with him (barring some sort of Patrick Ewing effect), but how much less exactly? 3 points? 5 points? Whether Vegas is, or is not, able to accurately judge the impact that individual star players have on their team is the question I wanted to answer.
Going in to this analysis, I thought there was a good chance that bettors would underestimate the effect of losing a star player. In general, I feel that the NBA is the league where a single player can have the largest impact on their team, as they are able to theoretically touch the ball on every single offensive possession. For an example of this, just look at the Cleveland Cavaliers before and after losing Lebron James: In the 2009-2010 season, they finished with a record of 61-21. The following season, without losing anyone significant from their roster besides James, they finished at an abysmal 19-63. Now, this is admittedly an extreme example, and if anyone wanted to argue that bettors are actually more likely to overestimate the effect of losing a star player, I would be willing to listen.
My first task was to define what I meant by “star player”. I decided that the easiest way to define a “star player” was as a player that was ranked either first or second on their team in PER. Using this as my criteria, I came up with the following list star players who spent time on the disabled list in the 2014-2015 NBA season:
- Anthony Davis, New Orleans – injured for 14 games
- Lebron James, Cleveland – injured for 9 games
- Kevin Durant, Oklahoma City – injured for 55 games
- Blake Griffin, LA Clippers – injured for 15 games
- Dwyane Wade, Miami – injured for 17 games
- Russell Westbrook, Oklahoma City – injured for 15 games
- DeMar DeRozan, Toronto – injured for 21 games
- Carmelo Anthony, New York – injured for 40 games
- Jimmy Butler, Chicago – injured for 17 games
- Kemba Walker, Charlotte – injured for 20 games
- Kawhi Leonard, San Antonio – injured for 16 games
After picking out these 11 players, I simply compared the winning percentages ATS of their teams with them and without them. I found that with their star player healthy, the teams of the above players went a combined 294-306 against the spread, good for a winning percentage of 49%. With their star player injured, the above teams went a combined 109-111 against the spread, good for a 49.55% winning percentage.
Based on the above results, it looks like we cannot assume that the betting market prices injuries to star players inaccurately, and there are no inefficiencies to exploit. Although it is possible the effect of certain players can be over- or under- estimated, it seems that at least for the 2014-2015 season, Vegas does a good job of pricing the impact of star players in aggregate.