Mark Spears – of Yahoo! Sports – asks…Oh, Canada: Has Bosh played his final game?
Although the story begins with Chris Bosh’s injury and the fact he may leave Toronto this summer, the story get interesting (at least for me) when it focuses on Bosh’s current teammates.
The following three paragraphs get at the essence of the article (again, at least for me):
“As silly as this may sound, with Chris going down,” Raptors point guard Jarrett Jack told reporters in Cleveland, “it creates an opportunity for somebody to step up.”
That somebody could be the same somebody the Raptors might have to lean upon should Bosh leave this summer: Andrea Bargnani.
Hedo Turkoglu has been a disappointment since signing a $53 million contract last summer and Jose Calderon’s production and minutes have decreased this season. Bosh’s departure would give the Raptors some salary-cap room to sign another free agent – or bring back a package of players in a sign-and-trade deal – but the team’s best chance to sustain any type of success may have to come from within. At 24, Bargnani is still young enough to reach some of the potential the Raptors saw when they made him the No. 1 overall pick of the 2006 draft.
Before I comment on these paragraphs, let me just present the standard data:
Table One: The Toronto Raptors after 78 games in 2009-10
As Table One reports, Bosh leads this team in Wins Produced (not surprisingly). And the other players on this team have produced less than 23 wins. So losing Bosh is a blow to this team (a blow that would be mitigated if Calderon was playing better).
After noting the problems with Bosh and Calderon, though, Spears makes a couple of observations that I would question. The first involves Hedo Turkoglu, who Spears refers to as a “disappointment.” Turkoglu – who the Raptors signed to a $54 million contract this past summer – was counted on to help turn the Raptors into a contender. This seemed like a reasonable expectation to many since Turkoglu did play in the NBA Finals last year. But this year Turkoglu has only produced 3.6 wins. Such production hardly transforms Toronto into a top NBA team.
Is such production, though, a disappointment? As Table One notes, had Turkoglu maintained what he did last year the Raptors production of wins would only have increased by 1.3 victories. In other words, this team would still be struggling to the make the playoffs in the Eastern Conference. As I noted last summer, Turkoglu’s career WP48 after the 2008-09 season was only 0.105. This year – at the age of 30 – he is a bit below what he did last year and his career average. But the difference isn’t that large, and consequently, no one should really be that disappointed. Or at least, I don’t think one should be disappointed when something expected actually happens.
Turkoglu was not the only major financial move the Raptors made last summer. Toronto also gave a significant extension to Andrea Bargnani. As the Spears article notes, there is still some expectation in Toronto that Bargnani is going to develop into a productive NBA player. In fact, Spears thinks that Bargnani’s age (he is only 24) means there is still time for him to reach his “potential.”
If we look at Bargnani’s performance across this past year, though, it appears Bargnani isn’t improving much. As Table One indicates, Bargnani’s WP48 in 2008-09 was -0.026. This year his mark is -0.012. Yes, that’s better. But it’s far from a mark of a player who is helping a team win games.
And it seems unlikely Bargnani is going to make a significant leap in the future. He has already played 297 NBA games and 8,594 minutes. Furthermore, it appears NBA players generally stop improving around the age of 24 and 25. So at this point, is it likely Bargnani is going to develop into a productive player?
I think the answer is no. Certainly it’s possible. But it’s not likely. And that means, if Bosh departs the Raptors are in trouble. The players the team invested in – Turkoglu and Bargnani – are not likely to produce many wins in the future. Plus, if the Raptors make the playoffs it will not have a lottery pick to build upon. All of this suggests that once again I am Dashing Hope in Toronto (as I did last summer).
Let me close, though, with a tiny silver lining. Last year the Raptors completely changed their roster. Bosh, Bargnani, and Calderon are the only players left from 2008-09. Although the make-over didn’t make much difference this year, the Raptors could try a similar approach again. Eventually, if they keep changing players, they might just stumble upon some players capable of producing wins. Yes, that isn’t much to pin your hopes upon. But sometimes, stumbling around can get you someplace.
- DJ
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.
Shhhh… MLSE thinks I’m a good GM.
The real question is this: how many comments will it be before khandor shows up and tells us the real problem with the Raptors is the lineups that the sub-par head coaching staff has chosen to utilize?
I’m guessing it will be fewer than 10.
Bosh is displaying some “contract-year phenomenon” here, right?
He annoys me. He’s been solid on paper, but in real life, in the dying seconds of a close game, the Raptors consistently do the following (and lose):
- call an isolation play for Bosh (since he’s the “star”)
- other Raptors clear out so Bosh can drive the lane
- Bosh makes a lame fake step, settles for a weak 15-foot jumpshot, and misses
- Other team gets the rebound, because no Raptor is available to crash the boards since it was a clear-out for Bosh.
I hope they get some solid players with their salary space in the offseason, and win with a Team approach instead of their constant reliance on over-hyped stars like Vince Carter and Bosh.
$43M for 4 more years? $66M for 3 more years for Rashard Lewis? How productive are the thirty-somethings in the league and how much are they getting paid? How can the Celtics mgmt look Rondo in the eye and pay him less than the big three?
As this article suggests, Hedo has been drawing much noise than he has deserved. The point is the money he gets, it should be 2-3 millions less to be fair.
I think Hedo sold playing in the finals very well. Everything was for his favor including blocking Kobe…
Toronto seems stuck with Bargnani until Colangelo leaves.
A few years back when BC found Anthony Parker, Moon, Garbajosa, prospects seemed bright. Now it looks like we have some rebuilding years to look forward to.
Your model favors big men so Bargnani in negatives really means he’s doing nothing on the floor, I mean I bet he would be in positives with about 4 rebounds per game.
bags fly free//
That’s wrong. dberri’s model favors “productive” big men, but the average big man still produces 0.100 win per game just like the average guards.
So what do you think about the Manu Ginobili extension? On the one hand, he’s underrated. On the other hand, he is getting old.
Ginobili produced 8.2 wins last season in only 44 games. Wins are worth $1.6M each, so Ginobili earned $13.1M last season, but was only paid $9.9M. Interestingly enough, his next contract averages $12.9M/season so he will be fairly paid if he keeps his production anywhere near his most injury-riddled, worst shooting, and lowest minutes/game (since his rookie year) season.
I think that means he’s a pretty good deal, at least for the next couple years.
NBA championships aren’t simply won by great players. They are won by great players who give the team great production for their salary. Kobe’s 12-15 wins every year aren’t worth $30M, they’re worth closer to $20M, and by the 12-13 season years, the Lakers are going to be hurting, not because Kobe is a lousy player, but rather because he isn’t underpaid anymore and the Lakers will be paying $60M for only three players who are all well into their 30s.
Ah man, when I read about the new Kobe Bryant deal… Jesus.
Hey DB,
I have a question. How do derive expected wins from efficiency differential?
I know I can look at a team right now and see they have won X games with Y differential, but that’s looking at wins first. How do I find wins from efficiency differential? Hopefully this is an easyish answer.
I agree that the Turkoglu signing was a terrible mistake.
I also think that Bargnani is overpaid, but I don’t think he’s quite as bad as his WP48 indicates.
He gets compared to Cs, but he doesn’t play anything like most other Cs. He spends more time on the perimeter spacing the floor like a SF. So he’s never going to rebound quite as well or score as efficiently as a C that’s a good post player.
The problem is that if your C plays more like a stretch PF or even a SF, you need to take advantage of his ability to draw big men away from the basket and shoot over smaller men by getting the rebounding and efficient inside scoring from somebody else. Bosh does that to some degree, but it’s not enough.
I think I read that if Bosh leaves, Toronto is considering replacing him with a traditional C and moving Bargnani to PF. If that occurs, Bargnani’s rating will immediate rise a bit even though he’ll be doing the same exact things.
I’m not sure that’s the solution though.
I’ve been saying the same thing about this team all year. Assuming Bosh stays, they need a SF that is a slasher, good finisher around the basket, and an above average rebounder to take advantage of Bargani’s skills and to round out the team skills.
Bargnani is not star or anything, but IMO he has a rare skill set for a big man that could be effective if the team was built properly.
Right now he’s more or less the same player as Gallinari on the Knicks (who rates higher) except that Gallo gets classified as a SF and Bargnani gets classified as a C.
Speaking of Kobe Bryant, is it possible that skipping college shortens a player’s career due to the additional wear and tear? Jermaine O’Neal is looking like a very old 31. The first generation of high school-to-NBA players is approaching retirement and it might be interesting to compare them to players that stayed in college and played fewer games.
IS: Doesn’t Bargnani guard the opposing teams’ centers on defense? playing on the perimeter might explain why he can’t get offensive rebounds, but his abysmal defensive rebounding is more likely related to the fact that he doesn’t know how to jump and has poor rebounding instincts. although if I was his agent, I’d probably say it’s because Bosh is taking rebounds away from him or something.
Simon and bags fly free:
You’re both wrong, actually. dberri’s model doesn’t *favor* “productive” big men; since dberri already attempts to equalize big men with wings and guards, there should be no favoring of one positional group over another. Rather, the model favors cookie-cutter big men over non-traditional big men like Bargnani.
The problem is inherent in the way WP is calculated, and I hope dberri addresses this sometime soon. Before equalization, guards produce the lowest wp48 numbers, wings a decent amount more, and bigs produce the highest unadulterated wp48 numbers. If Bargnani was listed as a SF, his WP would actually be decent, but because of 1) his below average rebounding rate compared to other bigs and 2) his shot selection being much farther from the basket than most bigs, thus leading to much lower fg% numbers, his unequalized wp48 numbers are closer to those of SFs and not PF/Cs, resulting in a poor WP calculation despite decent shooting %s for someone who takes the kind of shots he does.
For the above to be “okay”, it 1) cannot be arbitrary and 2) must be supported by calculations that one type of big is undoubtedly better than another. Some teams use their bigs to draw defenses in, leaving outside shooters open, while others use their bigs to draw defending bigs out, leaving the lane open. How much is a 3 point shot from a big worth compared to a rebound? If Bargnani was on a team like Houston with Chuck Hayes designated starting C and Bargnani starting SF, does he really produce X more wins than if he was listed as the center, despite the same lineup and strategy? I don’t have Jason Kidd’s numbers for this year, but take a look at his 2008-2009 figures. Is he really better than LBJ, Wade and Nowitzki? No, but his rebounds are exceptionally high for a guard, and thus he gets a bonus the same way Bargnani is penalized for having low rebounds for a big. The difference here is where Kidd is applauded for being multi-faceted, Bargnani is being penalized for the same.
Take a look at Nowitzki to see what I’m saying. Is he really valuable to his team than a Jeff Foster or Dampier? I know a lot of people vouch for their favorite players (e.g. Ben Gordon, Allen Iverson, etc.), who may actually be a detriment to their respective teams, but those are usually the high volume, low-percentage shooters. Nowitzki is actually a very efficient scorer; he just doesn’t rebound or stay near the basket the way Dampier/Foster do. His plus (3-point range) is rewarded by the model, but then penalized right after because 3-point range inherently lowers the player’s fg%.
Suggestion: Equalize the value of each stat beforehand so that the wp48 numbers are in the same range with no need for further equalization (best case scenario: find out how much each stat really is worth, akin to Moneyball statistics in baseball). This would probably lower the weight placed on rebounds (one of the reasons bigs end up with such high unequalized wp48s — rebounds *are* additional opportunities or additional opportunities denied, but unlike a steal/block/etc it probably has a higher rate of overlap i.e. a player getting a rebound denies a teammate from getting that rebound at a higher rate than a player getting a steal/block denying a teammate from the same), while considering the different types of field goals. Considering the stats available already on sites like 82games.com, it shouldn’t be too hard to put different weights on dunks/layups/tip-ins vs mid-range jumpers, for instance. At the very least, the model should be using 2 point shots and 3 point shots, rather than field goals and 3 point shots, to get rid of double-counting.
LOL Khandor would be right. Waive Bargnani/Wright and give their minutes to Johnson/Evans/Weems/Derozan and the team would likely improve by 10 wins by Wins Produced Math.
J//
No, WP DOES favor productive big men since the standard deviation is much higher among big men. It’s supposedly because of the short supply of big people. It has been mentioned countless times both here and the book.
And your claim about 3 point or Nowtizki aren’t really true either since it’s not FG% that matters but rather points per shot. Nowitzki used to be one of most productive players in terms of WP. It’s only in recent years his numbers went down. Guys like Nowitzki fares better than Bargnani because on top of other things like better rebounding, they often simply shoot better than Bargnani when freethrows and 3 pts are taken into consideration.
Kidd is awarded for being multi-faceted but Bargs is penalized because unlike Kidd, who’s an excellent passer and a stand out rebounder for a guard, Bargnani is slightly above average (shooting efficiency, block shot) to below average (passing) to terrible (rebounding) when compared with other centres. In order to be called truly multi-faceted, you have to be good at those facets.
So it’s simple. Bargnani needs to shoot better from 3s like Troy Murphy or learn to earn more freethrows like Dirk. Or even better he can start rebounding and do other things besides scoring. The problem here is that rebounding is one of more consistent stats and it’s hard to see Bargnani suddenly boost his rebounding by a huge margin.
Also if the “overlap” is really that big in rebounding , why shouldn’t the same logic applied in scoring? Scoring has much more “overlap” than rebounding since every shot is taken from a teammate. So Bargnani’s scoring should be weigh less just as rebounding should be?
It sounds to me that when you wrote “find out how much each stat really is worth” isn’t about actual worth of those things but the perceived value of those skills. For instance I’ve seen countless people claim outside shooting big men help others by opening up lanes. However I haven’t seen any empirical evidence, despite the claims, that they actually do improve the productivity of teammates in a way that aren’t captured by boxscore. You would think at least adjusted +/- would show some evidence but it’s far from it. Bargnani for example, has the worst adjusted +/- on the Raptors.
And I think you have fundamentally misunderstood the model. It’s an obvious simplified model for individual contribution that’s far from being perfect but it’s valuable because it has a fairly decent degree of predictability that is linked to team’s wins. Some of the changes you’ve suggested sound interesting, but taken in all, will they actually improve the performance of the model? Or is it just something that LOOKS or FEELS valuable like a Iverson twist-fadeaway jumpshot? (Iverson takes all the hard shots so his teammates do better!… or so they said)
J//
Also, to my understanding, the model takes how good a team is and breaks it down to the players. It says players like Rashard Lewis are rated lower than you would normally think. This is because, in Rashards case, he is playing out of position at PF. DB has shown before that if Lewis played SF for the Magic then his WP48 would be .150+. So it wouldn’t mean that Lewis is a bad player in an absolute sense, but that the way the Magic use him is not the best for the team. I do a much better job of understanding this myself than explaining it I’m sure.
Simon is correct… besides, bargnani, a seven footer, is purposely not utilizing his size. Why be that tall and play like a sg or wing? I doubt Phil Jackson or Stan van gundy would want their centers to shoot threes and play on the perimeter.
Anon,
I agree. Playing on the perimeter accounts for his lackluster rebounding on the offensive end. But he’s definitely not a good rebounder for a man his size no matter what. That’s a big reason why he rates so poorly.
I’m not trying to make the case that he’s a good player. I’m trying to make the case that he’s not as bad as he looks. lol
This is kind of a freaky issue for me.
Like I said before, Gallinari and Bargnani are quite similar in a lot of ways.
I saw a couple of games between the Knicks and Raptors and came away thinking Bargnani was the better player at this stage of their careers. Yet because Gallo plays SF and Bargnani plays C, Gallo rates better.
Something about that makes me uncomfortable because I like to think in terms of skill sets as opposed to positions.
As a prediction though, I think Gallo will be a much better player before it’s all over. Gallo has all-star potential. Perhaps even high level all star potential.
Simon,
I think outside shooters are slightly penalized by just about every model.
If you look at the average TS% from each position, big men tend to produce slightly higher ratings because they are inside more often.
If you break out the big men between inside players and outside players, the difference becomes larger.
You even see this at other positions. Slashers and post up players among Gs tend to have a higher TS% than outside shooters.
I think even though TS% captures the value of the 3 pointer, it doesn’t capture the value of the mid range shot.
You can argue that mid range shots are bad shots to be taking, but IMHO they are frequently a necessary evil and someone has to take them.
As suggested, some of this get captured in positional adjustment, but not for the outside shooting big man or the slashing small man.
The former tends to get rated lower because of it and the latter tends to get rated higher for his skill even though the team needs both.
But it’s clearly possible for an outside shooter to be very productive. Dirk a few years ago for example, or Troy Murphy for whatever reason. You just need to be good at what you do and not terrible at everything else. I think there might be some validity to what IS says, but I think the advantages of having outside shooters to open up the floor are overrated by a lot, and so it makes sense to me that if you shoot but don’t do much else, your value isn’t as high as it could be.
>Why be that tall and play like a sg or wing?<
That is the crux of the matter.
A lot of NBA people think it's an advantage to have a very tall player than can shoot from the outside.
He'll either draw other big men away from the basket and make it easier for the slashers to get to the hoop and the rest of the team to get offensive rebounds, or he'll have a smaller man on him that cannot defend him.
This is why you often hear players, coaches, etc.. saying that Dirk is the toughest cover in the NBA. They think the combination of his size and outside shooting skills creates value that may not all accrue to Dirk.
Whether that is true and how large a contribution it is is for stats guys to determine.
But other teams seem to be mimicking it because the "stretch PF" is a fairly hot commodity. That's why the Cavs brought in Jamison.
IS//
Like Anon, I agree to the extent. A team cannot take inside shots every time and they’ll have to mix in perimeter shots, but I think the 3pt line mostly negates this and this is exactly why long mid range shots are the worst ones to take. Derrick Rose for example, is not as good as he looks because he hasn’t grasped this simple concept. Mid range shots are OK, but only if they are used to set up other more efficient ways of scoring.
We’ve been here before. We talk about Bargnani and you claim outside shooting bigmen are underrated by this system. Then I mention many other players, like Matt Bonner and Troy Murphy, who are perimeter-oriented that fare better than Bargs in the system. Then you claim Bargs isn’t as good as them but he is still underrated by the system. But at the end it always comes back down to the evidence. Is there empirical evidence, other than the traditional thinking and anecdotal evidences of specific plays, that shows having perimeter shooting big men improves the performance of their teammates? I have yet to seen any.
Even more, I can picture many cases where not having a centre in the middle will be detrimental to the offensive ball movement for cutters rather than benefiting them. Not to mention outside shooting guards will have less shooting opportunities if their centre is operating outside.
At the end, you cannot have all outside shooting players and all inside shooting players. However a “normal” NBA team will always have enough skilled players that perimeter shooting big men aren’t that underrated by the system.
Do mind you that this is one of those things the model obviously doesn’t consider. What if you have a team full of player who are extremely low volume scorers? What if you have a team full of players who absolutely cannot make anything other than layups? (The often mentioned 5 Rodmans/Camby questions) However I think such question is quite meaningless in actual usage as a regular NBA team that consist of guards, forwards and centres who will always have players who are able to increase their usage and some players who will be able to make their shots. I can certainly think of extreme situations but obviously that’s outside the scope of a normal basketball team.
Everyone keeps saying that WP favours big men.
Isn’t it more accurate to say that in general big men are just more productive?
Big men tend to grab more rebounds, block more shots and shoot higher percentages. If we accept these things as conducive to winning games then big men are just better or more useful in general.
We don’t say that the model penalises players who are inefficient, possession eating,
long range bombers do we? No, we accept there is an opportunity cost with each possession and therefore those players arent helping very much.
If anything it is demonstrably the case that big men are penalised by this system and smaller guys favoured in the overall picture, due to the fact that they are compared to others at their own position rather than the league in general.
Duncan, Gasol and, to a lesser extent, Garnett are all big guys that have a face-up game away from the basket. And their teammates benefit from it. Tony Parker has made a career of breaking defenders down and getting into the lane because Duncan spends time at the high post. High/mid-post play leads to old school give and goes, picks and lanes for drives to either side of the post. Dwight Howard is spending time in the mid-post now, too. He’s facing up and beating guys going left or right—he’s got a left-handed hook shot! Catch-and-shoot guys and cutters will benefit from his play. Right now Orlando has neither. He’ll also need a jumpshot or defenders will just sag some. The Bulls and Pistons won titles with big guys playing away from the basket (Laimbeer, Cartwright). You can also argue that big back-to-the basket guys are more of a hindrance, unless they’re really dominant. Shaq is killing any kind of flow the Cavs would like to have and at times is just in the way. Good gosh, long-winded, I know. Just wanted to point out that there are no absolutes regarding post-play.
todd2//
Interestingly Laimbeer is one player who’s rated very highly by this system and people question his value, illustrating your point that one doesn’t have to play a certain way to be a good post player.
All these position adjustments and teammate effects discussions are still interesting, but I’m somewhat miffed that the level of discussions has’t really improved at all from what they were back in 2006-2007 on the very same topic. In fact I find the old discussions were much more lucid and elaborate since we had much better commentators here than someone like me.
http://dberri.wordpress.com/2007/07/09/adjusting-for-position-played/
http://dberri.wordpress.com/2006/11/19/more-thoughts-on-duck-on-a-rock/
Is it possible that everyone takes wp48 too literally? A great example of why we shouldn’t is a Dennis Rodman type player. I believe he rated at .4+ in the 72 win season for the Bulls. That tells me more about Rodman’s teammates than it tells me about him. I’m starting to think of that number as wins potential rather than wins produced.
Something has to be said for a player who gets his team extra shots but realizes that he has teammates who are better at taking those shots. However without the right teammates those extra shots go to waste.
This takes me to Bargnani. He is the anti-Rodman.
“LOL Khandor would be right. Waive Bargnani/Wright and give their minutes to Johnson/Evans/Weems/Derozan and the team would likely improve by 10 wins by Wins Produced Math.”
QFT!!!
Simon,
I happen to think outside shooters are underrated by every system, not just this one. lol
Even though to a large extent 3 point shooting mitigates the problem and avoiding those long two point mid range shots is partly a learned thing, I think taking some of them is an unavoidable part of putting together a winning team.
I think whoever takes those shots is penalized statistically.
IMO, the evidence is the fact that Cs and PFs tend to have a slightly higher average TS% than Gs. That gap tends to widen when if you break players out as outside shooters, slashers, post players etc… instead Cs, PFs, SFs, Gs etc…
Those mid range shots are typically taken by Gs and SFs. I believe we correct for it partly by the positional adjustment.
However, since not all players fit into a neat cookie cutter positional adjustment we have an issue that is at least worthy of consideration.
The G that is a huge slasher taking mostly inside shots and drawing fouls will look even better than he is because of the positional adjustment that compares him to other Gs that usually shoot from the outside.
The C that is excellent from the outside will look inferior to the C that scores from the inside and draws a lot of fouls because he is less efficient.
Now of course, if you could build a team of all inside shooters, slashers, and great 3 point shooters, we might be on to something. lol But the reality is that some of those poor mid range shots are a requirement from time to time and I don’t think we should assume it has to come from certain positions.
All that said, I don’t think this is a huge issue.
I just tend to think that as great as Wade, Jordan, Magic, and Lebron are, they are not quite as great as we think (they are super duper freaks but not super duper super freaks lol) and as great as we think Bird and Dirk are, both are better than that.
I have no clue by how much.
In my mind, I just adjust marginally and tend to favor outside shooters when all else is equal.
simon,
“Is there empirical evidence, other than the traditional thinking and anecdotal evidences of specific plays, that shows having perimeter shooting big men improves the performance of their teammates? I have yet to seen any.”
Basketball prospectus did a study exploring offensive effeciency of teams with shooters (not just bigs) versus teams without. They found that teams with at least 4 of 5 players being capable shooters are much more effecient that teams without.
Two anecdotes they gave (ugh) were the Cavs remaking a big jump in offensive effeciency after adding Mo Williams; he’s not an outstanding player, but can really knock down open looks. Miami adding Chalmers was another example they used.
And yes, there were numbers behind the article.
Out of curiosity, can anyone think of a good jump-shooter NOT helping a team’s offensive effeciency? Even if he’s only average from a TS% or EFG% perspective.
Prof,
I hope it is ok that I post this link.
http://basketball-statistics.com/howtheshootingabilitiesofpointguardsaffectoffenses.html
This guy studied 3 point shooting ability of point guards and how it affected offenses. He did that for each position. But, the results seem to suggest that is important to have guards that can shoot 3s. Which I think agrees with IS’s hypothesis.
Phil//
I don’t see how that article’s particularly applicable to the discussion of outside shooting big men being systemically underrated. I think there’s certainly some merit to it, but only to the extent that a team probably can’t live off one extreme on inside or outside. You want balanced attack. Also let’s not forget that for the guards, it’s often hard to be productive without being a good 3 point shooter unless you can really contribute in other ways or draws a lot of fouls.
So what would this mean? A perimeter shooting big man “may” help more than the stats suggest if a team had guards who cannot shoot at all but only can drive and post up. Does that sound like what would happen in an ordinary NBA team? Not really to me. Also that ignores the point that in such cases centres’ contribution should partly be reflected by some extra assists, which isn’t true in this case anyway as Bargnani is below average in passing.
And yes, I can see how a centre playing in the perimeter can actually stagnate ball movement and restrict plays, and it’ll also take opportunities away from perimeter players who’ll have to stay on the weak side or inside.
However the crux of the matter is this. In a regular setting, I don’t think a player is really that much more or less valuable than what is shown through boxscore with the exception of defense.
On a related note I want to ask a question. Using the same logic that outside shooters make their teammates better, does this mean inside-oriented centres or guards who cannot shoot, Dwight Howard, Wade, Rondo, Rose, etc, make their teammates worse even though it’s not captured by stats somehow? Shouldn’t that be the case if outside shooting had such drastic effect? ;)
If I follow the argument is that having a big man who can shoot from the perimeter, while damaging his own efficiency, will actually benefit the teams offense – presumably by opening up the floor?
I can certainly see that outside shooting skill could be beneficial for drawing out defenders, thus freeing up lanes for cutters, enabling more successful pick and rolls etc.
Indeed if a big can keep his defender honest by draining an occasional three, and then used this threat to execute the plays mentioned above, the offense would certainly benefit.
This would of course however show up in that big mans box score assist numbers wouldn’t it?
Do we have any evidence that perimeter shots are correlated with assists for frontcourt players?
Michael//
That type of reasoning has problems though. The other side of the coin is that post-oriented big men attract double team and thus creates spacing for other players even if it doesn’t show through assists from centres. Imagine a play where a centre has to be double-teamed near the basket, he passes it out, defense frantically scrambles to catch up, then boom! either a wide open 3 point or a quick cut& layup off a extra passes occurs. So in that hypothetical play, the post player created spacing and helped his teammates, but in a different way a outside shooter would
However this goes right back to the point I have already mentioned. . There are an endless amount of anecdotal scenarios one could imagine in order to show superiority of one style of centre over another, and likewise, statistically I just don’t think the effect in general is all that significant.
There certainly could be exceptional cases where jump-shooting big men do improve their teammates’ productivity without doing much on the boxscore. However looking at the best known perimeter big men, such as Nowitzki, Bargnani, Bonner, Okur, etc, and their teams, I doubt any of them really is known to significantly improve the productivity of their teammates in the way that suggests they are underrated by boxscore-based systems.
Anecdotal scenarios and conjecture are one thing, but I am suggesting that you could quantify any potential positive effect of perimeter shooting skills in big men by looking at their assist numbers.
Since those numbers would be credited to that big in the box score it would also remove the notion that WP doesnt give these types of players fair credit.
Michael //
Yes I agree and I have said something similar about assists somewhere above between my verbal outbursts.
However that is not the discussion. IS and Phil, AFAIK, are arguing that the perimeter shooting big men aren’t getting credits through assists or anything else that shows up on boxscore but the players are still making others better through spacing and thus they are systemically underrated.
But back to how this all started, Bargnani is a below-average passer. And as a Raptors fan who have seen the Bargnani & Kapono experiments where fan forums were super excited about all the spacing those players would provide,(and obviously failed) I think I’ve seen my share of anecdotal stories on this matter. :(
Simon,
I think we actually agree.
We certainly agree that a balanced attack is probably superior to either an all inside or all outside attack.
We also agree that it’s sort of rare to have a C or PF that is a very good outside shooter and guards that only slash and draw fouls etc…
I’m probably making more of this than it deserves. I’m just uncomfortable with a purely positional adjustment because not everyone fits into the cookie cutter.
So when I identify someone that does not fit, I tend to think the team would be best served by trying to balance the attack with another player that also does not fit and for basketball analysts to try to understand the contributions of each in a non traditional way.
Hence, IMO, even though Bargnani does not fit the stereotype for a C, he does have some valuable skills.
Since he brings outside shooting but is not a good rebounder, IMHO the Raptors should be looking for SF that is an above average rebounder, slasher, post up player without worrying as much about his outside shot the way the typical team would.
D. Berri
In the future when you provide stats on players I hope you will consider providing them both with and without positional adjustments.
It might help with some of the conversations along these lines.
On the assist matter, I can’t say I’ve given it a lot of thought or read much on the subject either.
I would point out that the typical starting SF does not get a lot of assists per 36 minutes. It’s somewhere around 2.75 +/- a little. Rebounds are about 5.75 per 36 +/- a little.
Since assists are statistically worth about 1/2 as much as rebounds, even if playing on the perimeter helps some of these stretch Cs and PFs get an extra assist here or there, it’s probably not worth a lot because the offense does not run through them very often and the assists are not super valuable statistically.
I want to be clear that I think at worst this is a marginal issue, somewhat rare, and it impacts every model I’ve seen.
If this model explains 95% of everything, then perhaps buried in the 5% is something along these lines that is worth less than 1%, but makes for good conversation.
Simon,
The article discussed players for bigs and guards alike, those to examples just stuck with me. For the record, I’m still on the fence. But let’s look at those four teams.
Utah has consistently had one of the better offenses in the league. Ditto the Raptors. Acquiring Bonner has coincided with having the best offense of the Duncan era for the Spurs. The Mavs have ranged from good to historically excellent for Dirk’s entire career.
Looking at the top 15 offensive teams in the league, I count 10 which employ a perimeter-oriented big. The five exceptions, in my estimation: the Hawks, Nugs, Cavs, Celtics and Grizz.
Only five of the bottom 15 employ perimeter-oriented bigs: the Knicks, Hornets, Pacers, Kings and Wizards (Jamison no longer plays for them, but he spent the bulk of this season playing there).
http://espn.go.com/nba/hollinger/teamstats
Is there any causal link?
I certainly haven’t demonstrated one, and I’m not sure there is one.
I’d wager that the Suns being a full 3 points/possession in front of the second best offensive team in the league has more to do with Nash than Frye, that it’s Howard putting up 20 points/game on 60 TS% which helps the Magic more than Lewis’s outside shooting. And Murphy’s excellent from beyond the arc hasn’t lifted the Pacers to lofty heights.
Heck, Utah was a lot better offensively in the late 90s, playing with – gulp – Ostertag. But I hear they had a couple of good players at the 1 and 4 spots.
Having a perimeter oriented big does not guarantee offensive efficiency, nor is one necessarily vital to a good offense. But it may be a contributing factor, beyond simply the points and points/possession of that player.
I absolutely agree that bigs who draw double-teams open things up for their teammates. But not all bigs can do that. No one’s doubling Perkins or Dampier on the block, even though they’re efficient scorers. Their offensive game is predicated on what their teammates do with the ball. Most teams rely on one big like that. But it’s rare for a team to be effective with two.