A great article at Wages of Wins explores the correlation between rookie year performance and career performance. As a Sonic fan, it's not the performance of Jeff Green or Kevin Durant that intrigues me (those guys will be fine, I'm sure), it's the performance - or lack thereof - of Johan Petro, Robert Swift, and Mo Sene.The crux of the article is that there is a 0.67 correlation between what a guy does his first year in the league and what he does the rest of his career. In other words, if Joe Player averages 8 ppg his rookie year, his career ppg is going to be somewhat close to that. Or, in the case of the Three Amigos mentioned before, if you average 2 ppg, well, you get the picture.
Now, as the author freely points out, he did not adjust for 18-year-old rookies vs. college senior rookies, or for players who are in their 1st year of organized basketball vs. players who have been dribbling since they were three. Still, it makes you pause a bit, doesn't it, to think that it's possible that only 1 of the 3 wundercenters drafted by Sund will ever pan out, and even that 1 is relatively unlikely.
At this point, I'd take 1 out of 3 in a heartbeat.
ReplyDeleteBy the way, does the theory say that on average, each player has an 0.67 correleation to his rookie year stats? Or that each player has an 0.67 chance of having a career that closely correlates his rookie year stats? I'm guessing it's the former, but in that case, the "1 in 3" doesn't really hold water...
I know, I should just read the article, but I'm lazy...
It's the former. When I said 1 in 3, I wasn't tying it to the statistics, but I can see how you would get that impression because I didn't word it clearly.
ReplyDeleteI was referring to Swift, who's numbers in limited playing time are decent enough to provide some hope for the future. I've pretty much written off Sene and Petro as non-contenders for a starting spot in the NBA.
The study was based on per 48 minute stats.
ReplyDeleteThe correlation suggests that rooie and career are on average about 70% similar but it will vary in each case and performance can go up or down.
Consistent with what Nuss just said, I see this passaage in the study
"Of the 140 who were below average but positive, 53 – or 38% — went on to be above average performers. The negative performers, though, were not quite as likely to become good players. Only nine of the 81 negative performers, or 11%, went on to become above average players."
In 05-06 Swift was right in middle of the range for guys "below average but positive", Petro was negative. This would suggest by the averages Swift has in neighborhood of around 40% chance to be above average and Petro only about 10%. That sounds about right but I still might give Petro more chance, if he moved to PF and if he started fighting for his next contract to stay in NBA. I think he has enough physical tools and some talent but not enough effort or mental focus.
"I was referring to Swift, who's numbers in limited playing time are decent enough to provide some hope for the future. I've pretty much written off Sene and Petro as non-contenders for a starting spot in the NBA. "
ReplyDeleteRight because as we all know it's usually the big white guy who pans out.
Hey, are you saying that Chris Mihm isn't a superstar NBA center? What about Christian Welp? Those are two big-time contributors right there!
ReplyDeleteThanks for the clarification nuss.
ReplyDeleteRegarding rj's "I think he has enough physical tools and some talent but not enough effort or mental focus," I'm not panning you or anything--just simply curious: Has a "not enough effort guy" ever changed their attitude for more than one year (ie. contract year) in the history of modern team sports? I can't think of a single example...
I just don't trust players that don't try hard. I think it's best to get rid of him while there are other teams that think they can convert him into a hard worker. I think history more or less proves that if he's lazy (or a wuss) as a rookie when he has the most to prove, he'll always be that way...
There are always exceptions, i suppose, but I don't see it from Johan. I don't see him ever being a starting NBA player.
ReplyDeleteStill, he's a decent enough backup, and as he learns the game, he'll have a role for the next 10 years as a 15 minutes a night guy who can get a block, put up some points, and - best - give you a few fouls.
I like Poop!
ReplyDeleteI couldnt come up with good example of effort improvement pure and simple. E Curry and Pryzbilla are mixed cases. But effort and focus may be hampered but not knowing what to do. Coming into league under 20 and /or from overseas or no college are factors. Andray Blatche is 6 months younger than Petro. He hasnt done anything so far. Does he have potential to be average or above average? I am tempted to say yes without knowing him. Petro and Swift have better cases, based on record to date, even if the odds are still against.
ReplyDeleteIf someone is going to mock me while using my handle, then I expect them to do a better job of it.
ReplyDeleteI suggest that you sit back, relax, take a deep breath, think a bit, and come up with a more creative way to take a shot at me.
It shouldn't be too difficult of a task, genius.
Some of the comments about correlation, including mine, weren't on target. I used to know this stuff better. There is a difference between r and r squared. If r= .69, r squared shows it accounts for less than 50% of the variance seen in the data. So this is just a fairly strong predictor, about halfway between what is considered weak and strong correlation.
ReplyDeleteAs Howard U and Nuss both indicated the correlation figure isnt giving straight odds of rookie year matching rest of career, more giving an indication on a scale of how much on average it misses.
ReplyDeleteNuss' "maybe" getting 1 good career from 3 young centers given the poor and very poor starts and howards u saying he'd be happy with one look about right by the section of the article I cited which adds information about the distribution of that data, something the correlation by itself doesnt explain or predict. If Swift is a 38% chance of good career and Petro and sene are each 11%, then it is estimated at a 60% cumulative chance at one good career by the recent league averages. And no better than 5% of 2 or more.
Sorry if this was too much stat talk but I had to straighten it out for myself at least.
If you looked at patterns of the very young and those with little prior experienced the chances of good career might look different and maybe hope could be higher, I don't know. The sample size is only modest to begin with, and if you cut it down a lot more the size would be so small it wouldnt have any statistical weight.
ReplyDeleteBy the 3rd or 4th year I would think that what you see is what you get in the future in a very high % of cases, with big men of any background or PGs...
Don't worry, rj, there's no such thing as too much stat talk, at least as long as the stats have to do with the NBA.
ReplyDeleteAll things considered, Mouhamed Sene (Net Efficiency Rating: -11.1; Net +/- Production: -10.9; Net Points Per 100 Possessions: -15.7; Roland Rating: -11.0) and Johan Petro (Net Efficiency Rating: -9.8; Net +/- Production: -7.2; Net Points Per 100 Possessions: -7.7; Roland Rating: -9.1) are busts.
ReplyDeleteWhile Petro has proven himself to be the second-coming of Brad Sellers (i.e., a tall, lanky, and defensively inept power forward whose offensive game consists solely of a face-to-the-basket mid-range jumper), Sene will go down as one of the worst draft picks in NBA history.
Looked at "untradeable" youngster Andrew Bynum. Compared to Petro he scores 5% more per 48 minutes, grabs 20% more rebounds, shots 4 % points better from field and does better on shot defense by the same amount.
ReplyDeleteSene in insignificant minutes against subs had better defensive numbers and equal rebounding.
At a Blazer board they had a thread about combining players into a single guy. Combine Petro offense with Sene rebounding & defense and they'd at least come close to Bynum. But you can't and odds of either figuring out the other half of the game aren't strong.
Sonics under Sund seemed to bet heavily on teaching and either made a bad bet because teaching isn't that powerful or Sonics coaches were bad teachers.
Presti seems to be betting on teaching but with stronger students. We'll see if the teacher is better and if the results are better.
In 05-06 Swift wasn't quite the scorer per minute as the 06-07 version of Petro but the FG% was the same. He probably improves here in return. 05-06 Swift was equal to 06-07 Sene or Bynum on the boards and promises to be better on that if strength adds and weight doesnt hurt jumping and stamina. 05-06 Swift was worse than the worse of the bunch 06-07 Petro on defense. That is the most important test and the most unknown answer.
ReplyDeleteSwift probably has best chance of the three because he is "right" on scoring and rebounding already or 2/3rds of job. Sene you could argue shows some promise in limited minutes at 2 of 3 as well in opposite fashion. Petro could be considered behind, in that he is only at these standards on 1 of 3 skills.
Guys with incomplete game can be worth something as role players, especially when you have both types, in the hands of a master coach but it is tricky. They are usually cheaper to keep.
It is best when your top guys are strong 2 way players. Seeing Durant and Green develop on the defensive end is a big part of whether they were truly great picks and knowing who you can part around them. If both are decent to strong on defense you can go for more offense more easily. If only one, then you definitely need more help on defense somewhere. If neither is above average on defense where they play then most of what you put around them will have to have good defense to get the team to an adequately strong defense to make playoffs and then noise there.
The other day I wrote that "I've always found it odd that post players who are astounding weakside defenders like Marcus Camby, Jermaine O'Neal, and Ben Wallace receieve a lot more press than outstanding man-to-man, one-on-one interior defensive players such as [Jarron Collins,] Jason Collins, Joel Przybilla, and Radoslav Nesterovic."
ReplyDeletehttp://tinyurl.com/3x3eku
Unlike Przybilla, Nesterovic, and the Collins twins, Sene lacks the ability to square off against his opponent and play tough, hard-nosed defense in the post. Instead, Sene's lone skill is that of a weakside help defender who can sporadically position himself to block shots whenever a backcourt player is destroyed off of the dribble by his man (e.g., Luke Ridnour getting burned on a play-by-play basis).
In fact, y'all should check out the following statistics.
Jarron Collins (Opponents' Points Per 100 Possesions: 101.4)
Jason Collins (Opponents' Points Per 100 Possesions: 103.2)
Joel Przybilla (Opponents' Points Per 100 Possesions: 104.1)
Radoslav Nesterovic (Opponents' Points Per 100 Possesions: 103.2)
Mouhamed Sene (Opponents' Points Per 100 Possesions: 113.7)
Undoubtedly, Sene can't even hold his own on defense versus fellow scrubs during garbage time at the end of blowouts. In all honesty, the whole thing is a depressing, yet mildly amusing situation.
All things considered, I'm so low on Sene that I'd presently offer him ($2,105,520) to the Cleveland Cavaliers for David Wesley ($1,750,000) -- whose contract is partially guaranteed for just $250,000 next season, which means that he can be waived at little cost -- solely to save some money.
113 is bad for Sene and whole team.
ReplyDeleteThose other 4 guys mentioned are among the best lane cloggers though.
Other young guys
Bynum 109
Turiaf 111
Pachulia 107
Collison 110
Garnett and T Chandler are 107
101 is about the best there is. Ben Wallace is 101. Yao is 100 and change.
Duncan & Spurs with him at 99. If you consider him a center.
ReplyDeleteCamby and B Haywood at 108.
A good center can have impact but harder on a bad defensive team.
A salary dump might be an option if Presti and PJ don't believe in Sene. But that would be pretty emphatic. Tougher call might be whether they give him the 3rd year team option before season starts. I'd take him all the way up to Oct. 31 on that, make him earn it in camp and preseason games and if the performance is weak, not give it to him.
Although Ben Wallace (Opponents' Points Per 100 Possesions: 101.4) is the best interior weakside help defender in the NBA, it's P.J. Brown's (Opponents' Points Per 100 Possesions: 98.4) outstanding one-on-one defense in the post that made the Chicago Bulls a formidable defensive team.
ReplyDeleteAs it is, that's proven by the following statistics.
P.J. Brown (Opponents' Net Points Per 100 Possesions: -3.5)
Ben Wallace (Opponents' Net Points Per 100 Possesions: +2.4)
Defensively, the Bulls actually performed better whenever Wallace was on the bench, while the team was more successful whenever Brown played in the game.
Along with Jarron Collins, Jason Collins, Joel Przybilla, and Radoslav Nesterovic, P.J. Brown is one of the most undervalued players in the league.
In all actuality, Marcus Camby and Brendan Haywood shouldn't be mentioned in the same breath.
ReplyDeleteFirst and foremost, Haywood undeniably fits the aforementioned mold of a top-notch interior defender who's stuck on a dreadful defensive team. Irrefutably, Haywood (Effective Field-Goal Percentage Allowed: 50.7%; Opponents' Points Per 100 Possessions: 108.1; Opponents' Net Points Per 100 Possessions: -6.6) outplayed his backup and noted rival, Etan Thomas (Effective Field-Goal Percentage Allowed: 52.4%; Opponents' Points Per 100 Possessions: 112.9; Opponents' Net Points Per 100 Possessions: +1.6), on defense last season.
Yet, with regards to Marcus Camby -- who's really a power forward, as he normally mans the high-post on offense -- he's much more similar to Ben Wallace than Haywood on defense—there's no doubt about it.
Analogously, Camby (Effective Field-Goal Percentage Allowed: 50.1%; Opponents' Points Per 100 Possessions: 107.7; Opponents' Net Points Per 100 Possessions: +2.1) is to Maybyner "Nenê" Hilario (Effective Field-Goal Percentage Allowed: 49.3%; Opponents' Points Per 100 Possessions: 104.9; Opponents' Net Points Per 100 Possessions: -3.6) as Wallace (Effective Field-Goal Percentage Allowed: 48.5%; Opponents' Points Per 100 Possessions: 101.4; Opponents' Net Points Per 100 Possessions: +2.4) is to P.J. Brown (Effective Field-Goal Percentage Allowed: 47.5%; Opponents' Points Per 100 Possessions: 98.4; Opponents' Net Points Per 100 Possessions: -3.5).
In conclusion, P.J. Brown and Maybyner "Nenê" Hilario deserve the credit that's been received by 2006-2007 NBA Defensive Player of the Year Marcus Camby and former NBA Defensive Player of the Year Ben Wallace.
Unfortunately for them though, solid man to man defenders don't get the glory that weakside defenders do. It's much easier to pile up stats and make flashier, highlight reel blocks coming from the weak side and stoping gaurds.
ReplyDeleteI'm not completely writing off Sene, simply because he's got practically zero experience compared to other guys his age. I see Sene's shot at being a contributing player at about 25%, Swift at 50%, and Petro at 15%.
ReplyDeleteThose are all completely unscientific numbers, but it's a feeling I get from watching these guys. One thing we can say for certain: We will know definitively - barring injury - if any of these guys are NBA-ready by the end of this season.
It is true that Camby and Wallace didn't have great raw team +/- results last year and are overrated.
ReplyDeleteIn 05-06 Wallace was 3rd best in league on adjusted +/- (offense and defense and it was mostly from defense from him) while Camby was barely above average. PJ surprisingly with a small negative with the Hornets.
Wallace signing probably wasnt a very good call in the end, given th eother uses that money could have been used for then , now or in the future. Nuggets willingness to trade Camby is understandable given age and team results and Nene and hopefully KMart's return. Nene is a bit of an enigma but he had a good year.
PJ Brown shows what you know matters a lot in being a good interior defender. It isnt easy to teach quick, precise reads. Safetys in football are among the highest IQ football players. Centers need to be too. Outstanding physical attributes are nice but unless they are above average students of the game, the better offensive minded basketball players will find ways to score on them. Defense is a lot about getting to right spot and to some degree on the inside having stoutness to hold it. Eliminating easy shots saves as many or more points than takeaway blocks and steals.