Who tanks in the NBA?

Tanking: intentionally losing in order to improve draft position.

After my PhD buddy Chris and I circulated our findings that NBA teams tank a lot, we’ve been asked a few times, “Which teams are tanking?” Today I offer a quick look at teams that have likely tanked.

First, a refresher: we measure tanking by comparing performance before and after playoff-eliminated teams “clinch” their lottery spot. In the last couple games of the season, many teams lock in their spot, so they no longer have an incentive to lose. Those games act as our control. The problem with doing it this way is that some tankers may keep trying to lose even after they clinch their spot. This could happen because teams shut down star players because of “injury” or just because teams develop a habit of losing.

So, the big caveat with the results below (and the results in our paper) is that we are almost certainly missing some tankers. Some teams don’t appear to tank because they look bad before and after clinching. Others don’t even clinch until the last game of the season, meaning there’s nothing to compare. We can only peg a tanker for sure if they prove they can play better once the draft position incentive goes away, so we’ll miss teams. That said, here are winning percentages for teams that played at least three games after clinching (ending with the 2008-09 season):

Only the Mavericks (1992-93, 1993-94) and the Timberwolves (1992-93) played 5 or more games after clinching. The Timberwolves didn’t seem to tank that year, but the Mavs were better after they clinched, especially in ’93-94. Their winning percentage was only 9% in the 11 games in which they could improve their draft spot (after playoff elimination). This jumped all the way to 44% after they clinched the worst record in the NBA. That’s a pretty shocking improvement!

Perhaps Dallas was bitter since they had the worst record the year before but the Magic won the lottery. The Magic had the lowest odds! The NBA shifted the odds in favor of the worst teams after that year. Dallas was not rewarded for their tanking, though. They lost again in ’93-94, this time to Milwaukee. At least they got Jason Kidd with the second pick.

I don’t want to hammer the other teams too much, since the sample sizes are pretty small. However, every team in this list except for the ’97-98 Nuggets played better after clinching their spot. This means that tanking is about more than shutting down star players. There must be some active effort (by coaches and/or players) to lose.

Three other teams do stand out: the ’95-96 Grizzlies, the ’00-01 Bulls, and the ’06-07 Grizzlies. All had winning percentages in the teens before clinching, but the ’06-07 Grizz managed to win their last three games! That’s not very likely if 18% was really the best they could do over the previous 11 games. The other two teams won two out of three to close the season.

These numbers make me wonder how much tanking is really out there. To make sure we get a clean estimate, we’re definitely missing some teams. If we want redistribution of talent through the draft, there have to be incentives to lose. Still, I’m surprised that players themselves seem to be putting in less effort on the court.

5 responses to “Who tanks in the NBA?

  1. Long time reader, first time commenter here.

    Could this dataset be expanded to include teams that tank to protect a traded pick (e.g. GSW this year)? Perhaps it’s not as common as tanking to get the first pick, but I’ve seen fans and journalists assume that it happens. Also, it feels like many of the teams mentioned above don’t start tanking from the outset; does it make sense to try to exclude early-season games before they appear to be tanking? I know that’s difficult to figure out, but it might be interesting.

  2. Thanks for the comment, and thanks for reading! Indeed, protected picks and traded picks in general would be helpful to add. Like you say, I would imagine teams tank even harder when they have traded a protected pick. If they’ve traded an unprotected pick on the other hand, they might not tank at all. I think I’m going to run this same analysis for hockey first (and maybe the NFL), but this is on my agenda.

    On your second point, I should have been a little clearer in the post. For precisely the reason you state, we only look at games after a team was eliminated from the playoffs. Some teams probably quit on the season even before they are eliminated, but that’s a safe cutoff to use after which they have nothing to play for except individual goals, pleasing the fans, and improving their lottery position.

  3. Ah, that makes sense. Looking at the numbers, it’s actually fairly clear. I should have paid closer attention when reading!

  4. Pingback: The Tank Watch (4/11/2012) | Causal Sports Fan

  5. Pingback: The Tank Watch (4/17/2012) | Causal Sports Fan

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s