ESPN’s smart guy, John Hollinger, released his 2009-10 player ratings for the season today. He had something to say about every Thunder player, but his comments on Kevin Durant are what most folks are talking about. He said (Insider):
Durant projects to lead the league in points per 40 minutes and to log one of the league’s largest PER increases, a sentiment that most observers would echo based on how he performed in the second half of the season.
The fact he will score is unquestioned; it’s his development in the other phases of the game that will define how high his star ascends. Durant exhilarated observers with his scoring ability last season, but his inability to pass or defend didn’t translate into superstar production. If Oklahoma City is going to turn the corner, it has to start with its marquee player becoming more well-rounded.
Wow. Projects to lead the league in scoring per 40 minutes. Last year, KD averaged 25.6 per 40 minutes (which is basically his average). This year he’s got him at 30.6 points per 40. He’s got his numbers at 49.2 percent from the field, 87.5 from the line and true shooting at 59.2. Rebounds per 40 basically the same, assists about the same (2.8 to 3.1), his usage up to 29.1 and his PER going from 20.85 to 23.50. And as we’ve all said all year long, the second KD rounds out his game (he’s just 21!!!), the second he jumps out of this world.
One thing about Hollinger’s predictions though is that he has Russell Westbrook’s PER falling to 14.84 (down from 15.20). Jeff Green’s is only slightly up. Every Thunder player besides KD is under 15, which in turn is under the league average. That type of stuff can make you scratch your head but as reader Dylan so perfectly put it:
I think sometimes there’s a disconnect between what Hollinger thinks will happen and the numbers he attaches to these articles. These projections are based on a formula he has. I noticed his projections have Russ shooting a lower percentage from the field this year than he did last year. Obviously that would be incredibly disappointing, but I don’t think Hollinger really thinks Westbrook is going to shoot 38%, but that’s what the formula said. If it’s anything like baseball sabremetrics, it’s harder to project what young players will do, because there’s less data to base it on.
Yep, pretty much. It’s hard to predict things for a team like OKC where the average age for the starting five is 24. Players are going to naturally progress and it’s hard to calculate that in a formula. So keep it all in perspective.
@dylan
The defining idea is that players improve with time, even in-season. While lots of people like to talk about a rookie "wall," statistically speaking rookies play better at the end of the year as they figure out the league. Hollinger has tried to base his formula around that common sense idea that players get better with experience, it just doesn't always work out. You'll notice also that Hollinger will explicitly state that he reworked the numbers sometimes for players injured near the end of the year. If he doesn't, injured players would be negatively impacted in a significant way because of the formula.
@dylan
Ditto.
I agree, the best way to get his FG% up will be to start converting those inside shots like Durant did last year. Deron Williams also provides a good recent example of a PG roughly his size who converted less than 50% of his inside shots as a rookie and eventually became very efficient inside in his second and third years.
@Keith
excellent post. I've looked into Hollinger's methodology, but I never knew it weighted the last month so heavily. Thanks.
The reason Hollinger's stats predict so poorly for Westbrook is because it puts greater priority on end of year stats for rookies. Statistically, Westbrook had his worst shooting months at the end of the season, so Hollinger's formula tends to think those will be closest to his actual numbers next year. With such an inefficient shot throughout the year, his PER should go down.
In a lot of ways this is more of a problem than his decision making. He will never have Paul or Kidd court-vision, but he can learn where his teammates will be without seeing them. However, there needs to be a huge leap in his shooting/finishing for anything to really move along. I think we expect a lot more from him, and he needs to see Durant to figure it out. Durant jumped 7% in finishing around the rim just by being used to seeing a help defender. Russell needs to make the same adjustment.
Can't wait to see how our guys perform this Wedsneday. My little pocket schedule indicates it should be broadcast on either channel 15 or 52, but I can't seem to find it on their broadcast schedule. From the way Ibaka played this summer, I honestly wouldn't be suprised to see them bringing him off the bench with the main club rather than stashing him away up in Tulsa. Wednesday's game should give us some idea of how ready Coach thinks he is to run with the big boys.
Honestly, I take any projection that uses only one to two years for its sample size with a grain of salt because, and I think anyone would say this, you really just don't know what's going to happen from such a limited sample size on such a mercurial sample group (NBA rookies and sophomores).
Obviously I'll be devastated if RW has such a dip in performance, but honestly, I'll be a little upset if KD's point production RISES that much because it means that he's having to do that to compensate for a lack of production from his teammates.
Any fan would love to see KD explode for a crazy scoring average, but the smart fan knows better, because such a high scoring average means Harden, Green, Westbrook, etc. are not carrying their load and it also means Durant would have to be playing an insane amount of minutes to accomplish that, which then means that the Thunder's bench is also not contributing. And the last thing this team needs is a step back in depth and offensive efficiency.
So thanks, but no thanks, to Hollinger's prediction on points. But I'll take more rebounds, assists, steals and blocks any day of the week.
I think Dylan's quoted comment above is right on; however, I'm still particularly confused that Hollinger's formula didn't project an increase in Russ's PER or TS%.
As Hollinger noted in Brook Lopez's projection and has often been noted on this site, "players that tend to improve the most as youngsters are the ones who have high turnover rates and strong athletic indicators (rebounds, blocks, steals, free throw attempts)." Russ was off the charts for his position in those categories, so I expecting Hollinger's projection to have him making a big leap and was even wondering if he would be projected above Rose based on those factors.
Anyway, I'm glad these are just projections as it will be a long season for the Thunder if Russ takes such a big step back.
As long as durant is, he should be capable of lots of blocks and steals, i mean he is only 35 lbs and 2 inches smaller then garnett is now, but averages 0.7 blocks and 0.6 steals less and 5.7 rebs less. When garnett came into the league he only weighed 220 lbs(about what durant is now) and still grabbed 6.3 rebs. Now obviously they play different positions, but at least the steals durant is capable of matching. Also durants assists will hopefully increase with harden being a better shooter, nothing like passing to watson and him missing
@Jax Raging Bile Duct
I'm the same way.
Basketball is so much harder to quantify statistically than baseball. In baseball, you're talking about what's essentially an individual sport. Each team gets 27 outs, so on and so forth.
But with basketball it's totally different. A point guard getting 10 assists with the Knicks isn't the same as one getting 10 assists with a slower team. A guy might shoot 40 percent from the 3-point line on a team with a dominant post man who draws attention in the paint, but on a squad without that presence, his percentage drops.
What I'm getting at is that I support Hollinger's work, because I love that he's attempting to quantify basketball in a way that goes beyond what can be sometimes misleading stats, but at the same time, you have to look at PER and stats of that nature with a critical eye.
I have a love/hate relationship with John Hollinger articles.
I love that Hollinger will value a player who isn't famous because that player performs well on the court (statistically). Hollinger's tendency to look at the numbers first helps curtail the love-fest that the poster kids get from other authors. This is also why I love the True Hoop network; they get into more detail about players other than Kobe and LeBron.
I hate Hollinger articles because I'm not in love with the PER formula, yet most of his articles/arguments center around it. PER is a good tool. In my mind, it isn't as good as adjusted +/-, but it's a good tool. But to focus as much attention on it as he does is annoying and repetitive, which is ironic since the reason I began reading his articles in the first place is because everyone else just wanted to swoon over Kobe and LeBron (which was annoying and repetitive).
Hollinger's right about KD though. His defense and passing needs improvement.