The following eight players share the same two characteristics. Each was selected in the 1996 draft. And each has appeared in at least two all-star games (stats are after the conclusion of the 2007-08 season).
- Kobe Bryant: 10 All-Star games, 136.0 Wins Produced, 0.207 WP48
- Allen Iverson: 9 All-Star games, 64.8 Wins Produced, 0.090 WP48
- Ray Allen: 8 All-Star games, 113.1 Wins Produced, 0.168 WP48
- Steve Nash: 6 All-Star games, 122.7 Wins Produced, 0.222 WP48
- Jermaine O’Neal: 6 All-Star games, 52.2 Wins Produced, 0.133 WP48
- Peja Stojakovic: 3 All-Star games, 67.9 Wins Produced, 0.146 WP48
- Antoine Walker: 3 All-Star games, 24.9 Wins Produced, 0.038 WP48
- Stephon Marbury: 2 All-Star games, 68.9 Wins Produced, 0.105 WP48
An average player posts a WP48 [Wins Produced per 48 minutes] of 0.100. So although all of these players made multiple appearances in the NBA’s All-Star game, some of these players (see Allen Iverson, Antoine Walker, and Stephon Marbury ) have not consistently been very productive players.
The Amazing Abdur-Rahim
The idea of an unproductive All-Stars has been discussed previously in this forum. This post, though, is not on this topic. No, I wish to discuss a player from the 1996 draft who was quite productiuon and yet will probably be forgotten by many NBA fans.
This past summer Shareef Abdur-Rahim retired from the game. Abdur-Rahim only appeared in one All-Star game. And he only made one appearance in the post-season (an appearance that lasted just six games). Despite a lack of acclaim and team success, he was actually a more productive player than many of the aforementioned “stars”. In fact, if we review the 1996 draft (something we did in The Wages of Wins), only four players selected twelve years ago produced more wins in his career than Abdur-Rahim (Bryant, Nash, Marcus Camby, and Ray Allen).
The Prospects of the Trailing Pack
As noted, Abdur-Rahim is done playing so he’s not adding to his 82.9 career Wins Produced. After Abdur-Rahim, the next five most productive players from the 1996 draft are Marbury, Stojakovic, Iverson, Jerome Williams (64.4 Wins Produced), and Jermaine O’Neal. Of these five, only Williams is done playing. So can the other four catch Abdur-Rahim?
- Marbury is currently modeling suits on the New York Knicks bench. In other words, he seems stuck at 68.9 career wins for awhile. And since he has only produced 8.8 wins over the past three seasons, it seems unlike Marbury is going to move up on this list even if he does play again.
- Stojakovic is still getting minutes, but like Marbury, isn’t offering much production. The past two seasons his WP48 has been below average and last year he only produced 5.4 wins. If he doesn’t revert to what we saw a few years ago, it will take him three more years to catch Abdur-Rahim.
- Iverson – who recently became one of my favorite players – produced 9.5 wins last year. So he could overtake Abdur-Rahim during the 2009-10 season. Iverson, though, is already 33 years old and he just switched teams. With his contract expiring this summer, he will likely switch teams again. All that moving around at his age probably won’t help him move up this list.
- And then there is Jermaine O’Neal. Relative to Abdur-Rahim, O’Neal has appeared in more All-Star games and has been paid more money. In other words, it’s likely that O’Neal is considered the better player. In terms of Wins Produced, though, Abdur-Rahim offered more. For O’Neal to catch Abdur-Rahim he will have to produce 26 more wins. Across the last four seasons, though, O’Neal has only produced 13.9 wins. So O’Neal is going to need to pick up the pace in Toronto to surpass Abdur-Rahim.
The Lesson Learned
Whether or not the players listed below Abdur-Rahim can eventually surpass his career win total is not really the point of the story. The key point I wish to make is the lesson I think his career teaches.
A basketball player can be very talented and he can be taken very high in the draft. The player can see his talent translate into real production on the court. And a player can stay healthy so that his production of wins surpasses many of his peers. But if a player is not surrounded by quality teammates, the perceptions of a player will suffer.
In both 1999-00 and 2000-01, Abdur-Rahim ranked among the top 15 in Wins Produced. His WP48 in each season exceeded 0.200 (twice the average). This production, though, was with the Vancouver Grizzlies. Because Vancouver was very bad, Abdur-Rahim’s production went unnoticed by those selecting All-Star teams. And a player like Jermaine O’Neal -who has never matched what Abdur-Rahim did those two seasons – eventually earned more money and certainly received more acclaim (primarily because O’Neal played for better teams in Indiana).
So we see, how you are perceived – and probably how much money you make – is influenced by the quality of your teammates. With the data the NBA collects -which does a wonderful job of separating a player from his teammates – this shouldn’t be the case. Unfortunately, often it’s true. Good players on bad teams are often discounted (while bad players on good teams are often over-estimated).
In closing, we might want to think of a few players that are following in Abdur-Rahim’s footsteps. One that leaps to mind is his former teammate in Sacramento, Kevin Martin. Martin has produced 26.4 wins across the past three seasons. This production surpasses the productivity of Iverson, Ray Allen, Richard Hamilton, and Joe Johnson. One a per-minute basis, he surpassed the output of Brandon Roy in 2007-08. And yet all of these other players appeared in the 2008 All-Star game while Martin (once again) stayed home.
Clearly, Martin fits the Abdur-Rahim pattern. If you can think of another example, please note this in the comment section.
- DJ
The WoW Journal Comments Policy
Our research on the NBA was summarized HERE.
The Technical Notes at wagesofwins.com provides substantially more information on the published research behind Wins Produced and Win Score
Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:
Simple Models of Player Performance
What Wins Produced Says and What It Does Not Say
Introducing PAWSmin — and a Defense of Box Score Statistics
Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.
Was there a link to Abdur-Rahim’s production in there, or did I just miss it?
Rashad,
No link. I didn’t feel like creating a table for him last night (too lazy). Isn’t it good enough for me to say that he was pretty good? (probably not)
I’d very interested in hearing your take on the new adjusted +/- statistic at
http://82games.com/ilardi2.htm
Their conclusions seem at odds with yours on a number of players (Rajon Rondo being a prime example). How to reconcile these differing points of view?
Chris,
I will comment on this in the next book. In the meantime, let’s see if other people can find some of the problems with what has been done. So far I have not been impressed by the ability of people who are on-line to offer critiques of other work posted on-line.
What about Andrei Kirilenko at Utah?
For his career, he is posting a 12.3 Win Score per 48 minutes, compared to a 7.3 average for most small forwards.
He has only one All-Star game appearance and has to defer to Deron Williams and Carlos Boozer on the popularity meter, not to mention averaging fewer than 13 points per game across eight seasons.
However, he has outstanding shot-blocking skills and grabs 1.5 steals per contest.
He doesn’t even have a ring, compared to two for James Posey, the quintessential “glue guy”
Apparently, according to the adjusted +/-, Jose Calderon is not as good as Ray Felton, Earl Watson, Derek Fisher, or Kyle Lowry??? And Chris Paul is one of the worst defensive point gaurds in the league??? I guess steals and rebounds mean nothing at all then…
There are two rather significant problems with the ‘adjusted plus-minus’ statistic. The first seems to be that for a sizable number of players, the error reported is of similar or greater magnitude than the measured effect.
To try to minimize this first problem, it seems that they’ve introduced a second issue. To try to find more independence in their dataset, they have employed data over a window of 5 years. While they say they are weighing more recent data more heavily, this weighting is arbitrary unless defined and measured against a real result. While it provides an “answer”, for this ‘answer’ to have any value, it has to have some level of predictive capacity, else it’s an objective answer to a subjectively created algorithm. Given the large standard error, I do not expect the predictive power would be particularly good.
Oh, and Camby should get at least some consideration in this discussion.
He helped the Knicks become the first (and to date, only) eighth seed to reach an NBA finals. His rebounding was one of the defensive bright spots for the Denver Nuggets, a team vilified for its perceived lack of such. And the Clippers were 0-6 in games in which he played less than 30 minutes (Yes, it’s oversimplified, but still, they needed him out there.)
He has a Defensive POY, but zero All-Star games? And his injury history has hurt his reputation. If anything, this post could have just as easily been about Camby as it was about Shareef.
Thanks Jason, that was interesting. I’m a huge Rondo fan, and an econ major (though I’m rusty). My “something isn’t quite right” sense perked up when they said that they used 5 years of data, but in order to make it timely, weighted the last year inordinately heavily.
Also, while I recognize that any good model should surprise us with the results, I find that some of the conclusions border on preposterous (which, of course, shouldn’t lead us to necessarily reject them).
I had dismissed the +/- stats that they posted last year because of the large standard error, but was worried that they might have got it right this time around, and I’d have to start hating on Rondo.
If we are going on shorter careers, like Martin, how about Okafor?
How about Marcus Camby? He has been a rebounder and shot-blocker with time in Canada, although his most effective time was in Denver. Late bloomers (due to health limitations as much as anything) working in the shadows of stars rarely get recognized.
Speaking of working in the shadows, how about Danny Ainge? He was an all-star once. His numbers appear comparable to Brent Barry, who you commented last season has produced almost 100 career wins. Oh, and Brent Barry.
Jeff Hornacek (probably better than Ainge and Barry), Buck Williams (outstanding rebounder), Mark Jackson, Vlade Divac (excellent defensively regardless of flops), Mark Eaton (in a one-trick pony sort of way), and Horace Grant (2/1 assist/turnover ratio). They all played second/third banana to the point that few respected their value.
Also, the Villanova star never quite getting his due, Ed Pinckney (career Paws/min of .060).
Could David West wind up being the next Abdur-Rahim? Even though he’s on a good (great?) team like the Hornets, he is a relative unknown.
I’m surprised you didn’t mention Iverson, as he’s a much less deserving candidate than Roy. I guess you really are a fan.
When it comes to all-star selection, players whose strength lies with scoring tend to garner votes.
However, like the guards chosen ahead of him, Martin is a prolific scorer. So what gives? A lot of things, actually. He played (and plays) for the, Kings, small-market team, alongside Artest, the most maligned player in the league. The Kings were out of the playoff race, and it’s extremely players from a losing to have a be all-stars, especially And Martin is not generally the type to come up with thunderous dunks, dazzling fade-aways, or miraculous circus shots after ill-advised trips to the paint; he is rarely highlight reel material. He also has really improved a lot over the past three years, and isn’t on a lot of people’s radars.
Part of it was that the West was a lot stronger than the East, and had several candidates more deserving of being an all-star (though I do not know if total WP or WP/48 min confirms the latter point). Martin wasn’t going to take Johnson or Allen’s spot, or vice versa.
Whether justified or not, “Stars” of a team tend to reap the rewards (in this case, selection to the all-star team) when their team wins, especially if that team exceeds expectations, and are often treated as an afterthought when their team stumbles, irregardless of their own part in that success or failing. Boston, Atlanta and Portland were all more successful than the Kings, and it’s unsurprising that all-star slots were granted accordingly, even though Martin arguably is better than those chosen ahead of him, and is a prolific scorer.
As for players that compare to Martin, there’s a huge list of players that deserve all-star berths (at least, according to WP), but don’t score enough points to typically merit inclusion in most peoples’ minds. Still, there are players that score prolifically and produce wins, but miss out on all-star berths.
Richard Jefferson and Manu Ginobli fit the criterion. But Jefferson played for a mediocre team alongside all-star fixtures Jason Kidd and Vince Carter, and there was no way a third Net was being selected. Likewise Ginobli is going bald, so he clearly isn’t all-star material.
At some point, though, don’t you have to acknowledge the fact that if Abdur-Rahim is the alpha dog of your team, you’re not going to be very good? Iverson, Bryant, and Allen have all carried rather mediocre supporting casts to playoff berths as the go-to guy, while teams with Abdur-Rahim as The Man were always bad.
Everyone always makes excuses for Abdur-Rahim (just like they did for Mitch Richmond), but the plain truth is that the mainstream fans & media basically got it right — this is a star’s league, and if your team’s #1 star isn’t really a true star (which Abdur-Rahim wasn’t), you’re not going to win. This isn’t baseball, where Albert Pujols can have an amazing season on a non-playoff team and it’s not really his fault (he doesn’t pitch, he only hits once out of every 9 times, etc). In basketball, the stars have a far greater say in how well their teams do, and the blame is largely theirs when the team is bad.
Fred, I’m guessing you’re a new reader. What you wrote there is exactly the misconception dberri talks about: Abdur-Rahim, while definitely not one of the greats, produced a lot more wins than people give him credit for, while players like Iverson did not do as much but got a lot of kudos for his team’s success.
In other words, what dberri claims is that many of the “mediocre” supporting players for likes of Iverson were actually better than many people believe, and the reverse is true for likes of Abdur-Rahim.
A related example would be the KG vs. Duncan debate, where Duncan often was thought to be a much better player because his team kept winning. dberri argued that’s simply because Duncan always had better teammates around. dberri.wordpress.com/2007/05/15/speeding-up-time-for-bill-simmons/
Oops, the proper link should be
http://dberri.wordpress.com/2007/05/15/speeding-up-time-for-bill-simmons/
Abdur Rahim just simply wasn’t very good, nor was he abyssmal. This may be a case (like Jason Kidd since his knee surgery) where the context of his teams and how bad they were and the fact that he is a big (with Jason Kidd it’s the fact that he’s one of the best rebounding point guards of all time and been on teams with no interior presence) where the statistics imply something that simply isn’t true.
For example, the statistics say Jason Kidd was one of the best players in the world last year and yet it’s obvious he hasn’t been able to defend for several years now, which is one of the reasons he got torched in the playoffs (Chris Paul being the other, though he would have been torched by any west point guard except Fisher) and one of the reasons he didn’t play that much during the olympics. Yet he still puts up numbers during the season while being no more than average player at his position.
Another problem with Kidd is that he can’t shoot, so by being on the floor, he creates a situation where perimeter defenders can jam all non baseline drives making it harder for your team to get the rim and more difficult for slashers to play their game. Not as much of a problem when he was quick enough to get to the rim at ease and pressure the defense, but now it is. Of course, he still does other things that are great like run the break and outlet the ball (probably still the best in the league at that) but he’s no longer really a winning player, and is actually a losing one whenever paired against an elite offense point. Even shooting 45% from three, the fact that he no longer gets to the free throw line says a lot about his effect on the game.
I’d bet that Abdur Rahim had many of the same negatives that negate some of the positive things he did. Like other current or former hawks, Joe Johnson, Stephen Jackson, if you spend your whole career being the best player on a bad team, perhaps it’s for a reason.
Kevin Martin is a better player than Abdur Rahim ever was. So is Kirelenko. But I’m not sure if he’s underrated. Look at what happens in the playoffs. Low 40s shooting percentage every time. So he’s not a frontline scorer or even that efficient when the defense gets tougher, and since the injuries he’s no longer near the perimeter defender he used to be or the shotblocker (on and off ball). I think for a couple of years he may have been the best defender in the league. But he’s not a frontline player, especially not with his current abilities though you could win with him.
David West is also better than Abdur-Rahim ever was. Al Jefferson could be the next Shareef Abdur-Rahim, as his defense is pretty bad, his interior passing is not good, and he always plays on bad teams, but he’s a better rebounder and scorer. Still his numbers are his numbers, they don’t add to those of his teammates at all. We’ll see this year if being paired with Kevin Love (who’s going to be great) and Mike Miller help at all. Probably the best two teammates he’s ever had, and both have good peripheral games for their position. Though Mike Miller can’t play d. And neither can Telfair. And they still have no great shotblocker which would help negate some of Jefferson’s inability to defend in any way. (Bad at pick defense, on the ball defense, and interior help.)
Walker has consistently been himself every time he has been allowed on the court. Fact is since he left Boston he hasn’t been given the court time to put up the stats and it has not been about him playing badly.
When he was allowed to play in Dallas and Minny for at least 30 minutes he got his stats. I remember one game in Miami where he was 3/4 and Riley still took him out of the game.
Walker also should have been an all star at least 5 times. In Dallas and at least once in Boston when his stats out produced the many picked in front of him.
The all star game is a popularity contest. Even when the coaches pick the players
Dear Fred,
Please see Garnett, Kevin.
I love watching Skip Bayless eat it
It’s insane to hear you guys compare Shareef to Garnett… It’s like comparing Maggie Gyllenhaal to Jessica Alba on looks. Before 2004, Garnett at least proved he could carry a team of nobodies to the playoffs (he couldn’t get out of the 1st round, but at least he got there). In 2004, you surround him with some decent talent, and he takes you to the WCFs. Anybody with eyes could tell Garnett was one of the best players in the game, if not the best.
So I don’t get why WoW people have been acting like Garnett was Berri’s own personal secret before last season. We get it: KG was/is really good. Thanks for stating the obvious.
Thanks also for proving my point about Abdur-Rahim. He was nowhere near as good as KG – while KG alone was good enough to at least get you to the playoffs, Shareef alone was good enough to get you a pretty high lottery pick. Like I said, it’s a star’s league. Shareef + scrubs = awful team. Garnett + scrubs = low-seeded playoff team. Proof that middling “stars” like Shareef are a dime a dozen.
I still can’t believe you even compared Shareef to KG. That’s just nuts.
anyone whose seen Abdur-Rahim play knows his stats are inflated from being on a bad team.
never seen the guy get a “tough rebound” in traffic. he just picks up the ones where there are 2 of his fellow teammates around and nobody from the opposition is within 5 feet.
his assists did not produce lay ups or close shots for that matter – just Mike Bibby, Glenn Robinson, or Jason Terry jumpers. he just happened to be the guy who through them the ball in screen and roll.
horrible big man defender in the mold of Zach Randolph and Eddy Curry (not size wise). if “changed shots” were a stat, he’d be at the bottom. and talk about slow rotation.
seriously, if this guy had “good” teammates, he’d never get the ball to put up those #’s. at best, he’d probably be an unathletic man’s version of what McDyess is now.
Shareef and Juwan Howard, two of the most overrated players of the 90s and the new century. they’d be 10 pt 5 reb bench guys on a good team.