In a recent post on LockedOnJazz, David Locke came to the conclusion that turnovers and offensive rebounds are overrated. Unfortunately, his analysis – IMO – is incomplete and his conclusions are wrong. It’s interesting that he chose these two stats, first they are almost the same thing as one could easily consider an offensive rebound a form of turnover, but also because in 18 years of coaching, these were my primary performance indicators; and they told me a lot.
Locke looked at the data from 30,000 feet and noticed that the number of turnovers and the offensive rebound percentage are not related to winning, but this airplane view of the data is quite confounded. Consider turning the ball over more versus a bad team, they are less capable of converting that turnover into points. That is, you could have more turnovers and win as long as your opponent was poor at converting. Thus, we have this contrary relationship confounding the conclusion that turnovers don’t predict losing. In making his point, Locke points out that San Antonio is a team that wins and gets few offensive rebounds. He goes further to claim this is by design.
Well, let’s get out of the airplane for a moment, jump into a helicopter and examine San Antonio from the tree tops. When San Antonio plays teams under 500, they are all over the board in these two stats. And as Locke points out, shooting here determines the winner. Wouldn’t it be nice if we could all shoot like the Spurs. But, in games versus good teams, these two stats were highly predictive of who won. And, in the one game where they lost to a poor team (Portland), San Antonio had more turnovers AND fewer offensive rebounds. These stats are starting to look more important, but this is not science either; this is just a case study.
Let’s climb out of the copter (I like to think we are sliding down a rope) and look at this on the ground. That is, let’s look at this functionally. Just like any other statistics, we can convert these two to points, and points determine who wins. A turnover costs one possession and pays one possession, and given one point per possession a turnover is minus 2 points. To counter this you would have to raise your field goal percentage by 1 ¼ points or so. Offensive rebounds also cost a possession and pay a possession, but the conversion on the payout is higher, so giving up an offensive rebound is closer to minus 2 ½ points. To counter this, you would have to raise your field goal percentage about 1 ½ points. San Antonio regularly outshoots their opponents by enough to offset their turnovers, but I can’t imagine that coach Popovich is any less angry when they occur. To make this point further, in the last four playoff games versus OKC last year, ones where San Antonio did not shoot as well, they turned the ball over 14 more times than OKC. Turnovers cost them two games, perhaps the two that stood between them and the championship.
All things being equal, which we assume at the beginning of the game, turnovers and offensive rebounds are critical to success unless you can guarantee shooting like the Spurs. Can the Jazz make that guarantee? What I’m trying to say is that you can compensate for one stat with another, but some are easier to control than others. I think reducing turnovers and getting offensive rebounds is much easier for teams like the Jazz than creating better and more consistent shooters. If you want to investigate an overrated statistic, it is the blocked shot. Sure, a little is gained by having your non-bigs fly at three pointers and the long twos, but having bigs leave the nest at any time is a poor choice. Likewise, block attempts close to the basket cost you as well (unless you do it like Tim Duncan – straight up). This is why Derrick is becoming a poor defensive rebounder and is leading him the wrong direction (a future Deandre Jordon). That BS “Swat Lake City” thing that Harpring started is working to undermine this team. Defense is about position, getting it, keeping it, and helping your teammates reestablish it. The Jazz would do better if they stopped trying to block shots completely, get rebounds (including offensive ones), and tell Gordo to stop giving the ball away. Those two stats matter, especially to the Jazz.