Tuesday, March 3rd, 2009 at 10:02 am  |  62 responses

Tweet

Is Durant The Anti-Battier?

Can +/- expose great players’ flaws or do our eyes deceive us?

by John Krolik

For nearly a month now, the thing du jour for basketball junkies like me has been the fantastic Michael Lewis article in the New York Times on Shane Battier that brought, among other things, the +/- stat into the mainstream. +/-, for those of you who don’t know, is the measure of how well a team plays with a certain player on the court versus how they play when that players off the court. It’s pretty interesting stuff, and helps to measure the value of guys who coaches love but don’t have contributions on the stat sheet, like a Ben Wallace, Jamario Moon or Andrei Kirilenko, all of whom have great +/- ratings.

By and large, there appears to be a real validity to the statistic and is a whole new way to look at which players are contributing without needing a number-crunching formula or the word of someone WHO DOESN’T CARE WHAT SOME STAT SHEET SAYS BECAUSE THIS IS THE BEST PLAYER IN THE WORLD, PERIOD.

The list of the top-five players in the League according to +/- tells you that it both has some validity and is far from perfect:

1. LeBron James
2. Chris Paul
3. Dwyane Wade
4. Lamar Odom
5. Jason Kidd

Okay. James, Paul and Wade are the kind of names you’d expect to see at the top of a list. LeBron and Wade are in most people’s top four for MVP, and Paul is having one of the best seasons a point guard has ever had. So far, so good. But Lamar Odom? Not many people have him as one of the best players in the League. In fact, I’d go so far as to guess he won’t even get the most MVP votes on his own team. (We go out on limbs like that here at TBF.) And Jason Kidd isn’t quite MVP-caliber anymore either.

+/-, both in the raw form used by NBA.com and the adjusted form used at 82games.com, has some serious flaws. (In fairness to Lewis’ article, he acknowledges this flaw and says the Rockets use an adjusted form of +/- whose formula is not available to the public.) If a player plays on a terrible team who gets crushed when he sits and only loses by a little when he’s in, his +/- can be higher than a guy whose team is good when he sits and better when he’s in. The worse your backup is, the better your +/-. Maybe a defensive stopper is only in while the other team’s best scorer is playing. Maybe a guy always comes in with another player who is either great or terrible. There are just as many holes in +/- as there are in any other statistical measure. But does the fact that +/- cannot be accepted as dogma mean it should be completely disregarded? Of course not.

So to help try and ballpark +/-’s actual value, it’s important to find case studies. And you’re not going to find a better one than Kevin Durant of the Oklahoma City Thunder. Durant appears to be a player both scouts and stat geeks can agree on. He’s 6-9 with a wingspan that’s even more impressive, he handles like a shooting guard, has one of the purest strokes in basketball and understands the game at a level far beyond his years. His moves from the perimeter are calculated and gorgeous to watch, from sweeping crossovers for smooth finger rolls to jab-step set-up moves to long-range jumpers that find twine time after time. He can even put on bursts of acceleration and throw it down on unsuspecting defenses. He is poetry in a prototype body.

Statistically, he’s showing all the makings of being one of the best scorers in NBA history. He’s averaging 26 ppg on a fantastic 59 percent True Shooting Percentage, which is actually better than, believe it or not, LeBron and Kobe’s career high in scoring efficiency. He appears destined to be not just a great shot-creator but one of the budding great scorers to ever pick up a basketball, and his rebounding and passing are even coming around. When the Thunder came to town, I made sure not to miss the chance to see Kevin Durant live.

But then why is his +/- an abysmal -6.4 on one of the worst teams in the entire league? Can a player as obviously good as Durant be secretly bad? (If you’re wondering if any other team’s best player has a negative +/-, the answer is no, unless you count Derrick Rose.) The issue here is Durant’s defense–he makes the Thunder 2.7 points per 100 possessions better offensively, but Durant costs the Thunder a whopping 9.1 points per 100 possessions on defense. This is consistent with last year, when +/- found him to be the worst defender in the League.

My first reaction? Bull-Pucky. Players on bad teams who play a ton of minutes can often have wonky +/- scores, and a player as good as Durant clearly cannot be a detriment to a team as bad as the Thunder. Durant’s backups could be playing amazing in spurts, the Thunder might be better in garbage time, other players’ best players might get benched with Durant, something. Besides, I can’t think of a way a wing defender could be so absolutely horrendous defensively that he couldn’t be hidden, especially not one as good offensively as Durant. (Case study: Steve Nash maintained one of the best overall +/- ratings in the League during the D’Antoni years.)

But I decided to see if the Thunder actually did play better without Durant over full games instead of in the short spurts than made up Durant’s odd +/- rating. And they actually have, albeit briefly so far. When Durant went down, the Thunder were on a seven-game losing streak despite a string of amazing scoring performances from Durant. The game he was injured, the Thunder took the Mavericks to overtime, and have rolled off a quick two-game winning streak.

Now, is it time to toss out Durant after two quick wins over the Grizzlies and Mavericks? Of course not. A million times no. And even if the Thunder play inspired ball during the two weeks Durant is out, all it proves is that he has to work on his defense, and Scott Brooks has to work harder to hide him on that end of the floor so that he can use his phenomenal talents in a way that can fully benefit the team. But this is something I’m going to keep an eye on as someone curious about the value of +/- and the way players are valued in the NBA.

Perhaps Russell Westbrook, the rookie with an brick-happy perimeter game, dubious finishing abilities, and a below league-average TS% but a staggering +10 +/- rating, is the true prize of the Thunder’s young duo. On the flip side, it’ll be interesting to see how the Cavaliers play without Ben Wallace, who appears more than replaceable on the stat sheet but had a fantastic +10.6 rating. If injuries to our favorite players are to occur, we may as well take the opportunity to try out a good, old-fashioned hardwood science experiment.