You Are Here > > BBR Blog > NBA and College Basketball Analysis

SITE NEWS: We are moving all of our site and company news into a single blog for We'll tag all Basketball-Reference content, so you can quickly and easily find the content you want.

Also, our existing Basketball-Reference blog rss feed will be redirected to the new site's feed. // Sports Reference

For more from Neil, check out his new work at

Which Players Have Played For the Best Offenses?

Posted by Neil Paine on August 20, 2010

Back in June, right before the Finals tipped off, I developed a method to estimate possessions for teams going back to 1951 using the following regression equation:

Possessions ~ -4.05*Wins - 3.96*Losses + 0.97*FG + 0.75*FGA + 0.70*FTA - 1.37*OReb + 0.53*TotReb + 0.31*Fouls - 0.50*Points +0.19*Opp. Pts

For most teams, this method can estimate a team's actual possessions total within roughly one possession per game, so it's surprisingly accurate given the basic nature of the inputs.

At any rate, I went on to use this method in finding the most similar NBA Finals matchups to the Lakers/Celtics clash, as well as in determining the Finalists that improved the most during the playoffs, and ranking playoff defensive performances. Today, though, I want to use estimated offensive ratings as a way to rank the players who have played for the best offenses during their careers.

Here's how it worked: For every team-season, I estimated their offensive ratings in both the regular season and (when applicable), the playoffs. In the playoffs, I also found the strength of the opposing defenses they faced and adjusted their ORtg up or down accordingly (i.e., if you had a 104 ORtg in the playoffs vs. defenses that had a 104 DRtg during the regular season, and the RS league avg. was 107, your playoff ORtg was adjusted up to 107). Then I subtracted the league average to find how good each team was relative to the mean, and found a career average for every NBA player since 1952, weighted by the number of minutes they spent with each team.

As an example, let's look at Leandro Barbosa, who would be the all-time leader had he met the minimum requirement of 15,000 career minutes played:

Year Age Team Type G MP TmOff
2004 21 PHO RS 70 1500 -1.74
2005 22 PHO RS 63 1087 +10.40
2005 22 PHO PO 12 116 +15.18
2006 23 PHO RS 57 1592 +7.70
2006 23 PHO PO 20 632 +9.90
2007 24 PHO RS 80 2613 +9.96
2007 24 PHO PO 11 349 +5.98
2008 25 PHO RS 82 2421 +7.34
2008 25 PHO PO 5 143 -2.80
2009 26 PHO RS 70 1705 +6.30
2010 27 PHO RS 44 786 +9.04
2010 27 PHO PO 16 250 +12.15
Career 530 13194 +7.23

As you can see, over the course of his career LB has played on offenses that were -- on average -- 7.23 points better than the league average, including 5 of the top 11 offenses ever... If you do this for every player since 1952, here are the players who played for the best offenses in their careers:

Rk Player MP TmOff Rk Player MP TmOff Rk Player MP TmOff
1 Steve Nash 35900 6.84 35 Ray Allen 41905 3.11 69 Jerry Lucas 34501 2.43
2 Amare Stoudemire 19621 6.46 36 Rik Smits 25847 3.11 70 Curtis Perry 15202 2.43
3 Magic Johnson 40783 5.65 37 Shawn Marion 33195 3.07 71 Ervin Johnson 18570 2.43
4 Michael Cooper 28379 5.51 38 Ed Macauley 19485 3.07 72 Lucius Allen 21319 2.42
5 James Worthy 35298 5.02 39 Horace Grant 44793 3.05 73 Bryon Russell 22886 2.42
6 Detlef Schrempf 36935 4.75 40 Julius Erving 33965 3.05 74 Dick Barnett 31968 2.40
7 Kareem Abdul-Jabbar 66297 4.40 41 Kiki Vandeweghe 26411 3.05 75 Billy Paultz 15246 2.40
8 Raja Bell 18683 4.32 42 Boris Diaw 17591 3.04 76 Jeff Hornacek 38730 2.39
9 Dirk Nowitzki 38033 4.28 43 Michael Jordan 48484 3.01 77 Ricky Pierce 26321 2.37
10 Oscar Robertson 47559 4.28 44 Bonzi Wells 16232 2.97 78 Elgin Baylor 39373 2.37
11 Scottie Pippen 49174 4.16 45 B.J. Armstrong 20154 2.97 79 Steve Smith 31758 2.32
12 Nate McMillan 22648 4.04 46 Clyde Drexler 43109 2.96 80 Jim Paxson 22464 2.32
13 Larry Bird 41329 4.00 47 Cedric Ceballos 15925 2.96 81 Derek Fisher 32436 2.30
14 Rashard Lewis 31687 4.00 48 Jon McGlocklin 22458 2.93 82 Jon Koncak 17437 2.28
15 Byron Scott 35518 3.97 49 Dale Davis 33268 2.93 83 Chris Mullin 34220 2.28
16 Josh Howard 15795 3.97 50 Shaquille O'Neal 49253 2.84 84 Robert Parish 51881 2.28
17 Danny Ainge 32793 3.91 51 A.C. Green 40671 2.84 85 Rudy Tomjanovich 26755 2.25
18 Charles Barkley 44179 3.88 52 Vlade Divac 37566 2.79 86 Greg Ostertag 16521 2.23
19 Sam Perkins 41383 3.81 53 Bobby Jones 23608 2.78 87 Wayne Embry 23110 2.21
20 Gary Payton 52599 3.67 54 John Paxson 19874 2.78 88 Larry Costello 22669 2.20
21 Adrian Smith 19070 3.62 55 Derrick McKey 31015 2.77 89 Grant Hill 31764 2.16
22 Jerry West 42892 3.52 56 Robert Horry 33889 2.76 90 Bucky Bockhorn 15656 2.14
23 Dan Majerle 34445 3.48 57 Kobe Bryant 45177 2.76 91 Dennis Johnson 42948 2.13
24 Kevin McHale 35834 3.48 58 Wali Jones 16181 2.75 92 Craig Ehlo 22914 2.11
25 Kevin Johnson 28940 3.47 59 Antonio Daniels 21493 2.71 93 Michael Finley 41904 2.10
26 Vladimir Radmanovic 15058 3.44 60 Tom Hawkins 20564 2.71 94 Bob Dandridge 33384 2.09
27 Kurt Rambis 18864 3.39 61 Hersey Hawkins 34563 2.71 95 Kevin Duckworth 19418 2.08
28 Jamaal Wilkes 31074 3.34 62 George Gervin 28738 2.67 96 Gar Heard 21396 2.07
29 Shawn Kemp 32231 3.31 63 Moses Malone 48867 2.58 97 Fred Roberts 15766 2.06
30 Steve Kerr 18204 3.28 64 Happy Hairston 26350 2.58 98 Brian Shaw 24108 2.05
31 Jim McMillian 22949 3.23 65 Mario Elie 21708 2.57 99 Dan Issel 23941 2.04
32 Larry Kenon 16548 3.20 66 Maurice Cheeks 39693 2.51 100 Quinn Buckner 17302 2.03
33 Peja Stojakovic 29105 3.19 67 Cliff Levingston 18009 2.47
34 Mychal Thompson 30472 3.15 68 Jerome Kersey 31509 2.45

It's no surprise that Nash & Stoudemire, cornerstones of the 2005-10 Suns offensive juggernaut, rank at the top; in fact, Nash was also the floor general for the 2002 & '04 Dallas Mavericks, two more teams that ranked among the 25 best offenses in NBA history. Nash has certainly had some talented teammates to work with, but his ongoing presence as the catalyst of some of the most dynamic offenses ever should make us re-evaluate where he ranks all-time -- and not just among offensive point guards, but among offensive players, period. He may not be the prototypical offensive machine like a Michael Jordan or Kobe Bryant, but Nash unquestionably makes his teams vastly more efficient when he steps on the floor. Given the unbelievable offensive track records of his teams, where do you think Nash ranks in the All-Time Offensive Players discussion?


60 Responses to “Which Players Have Played For the Best Offenses?”

  1. Jason J Says:

    I'm not sure how much we can read into this for Nash specifically because if we do a player comparison with Stockton at the same age, a fairly similar player in terms of basic role, we see John's advanced stats all come out ahead, ORtg, OWS, PER, WS/48, (DWS & Drtg by a mile)... and Stock & his teams doesn't even make the list? I get the feeling that this may be more of a system / play style / teammates issue than something indicative of individual ability. Not to say Nash doesn't deserve a great deal of love for his offensive contributions generally. Player comp below:

  2. David Fauber Says:

    Interesting stuff, Josh Howard and Rashard ranking so high are pretty surprising.

  3. P Middy Says:


    To be less prickish about it, I think it would be a good idea to split this into the current era and prior to the new defensive rules.

  4. Panic Says:


    Considering these are all relative to the league average, I'm not sure that such an adjustment is necessary. Today's players all play with the same rules.

  5. The J Says:

    I would think the fact that it's a comparison to the league average is what skews things towards Nash so much. He plays for a run and gun team in an era where few teams play that style. Other teams that played this style historically did so in era's where this style was much more common league-wide.

    If you think of the league today there are maybe 3 teams playing this style of offense first and defense second (or third or never). Considering one is the Warriors and the other the Knicks, it's easy to see why Phoenix is the strongest Offensive team. Now imagine how these stats adjust if they were playing simultaneously to the Showtime Lakers.

  6. Neil Paine Says:

    The Run & Gun offense makes no difference -- this isn't points per game, it's points per possession, which is independent of pace... The league average today is the same as it was in the 1980s.

  7. taheati Says:

    I'd like to see postseason splits, if/how the numbers change.

  8. Jason J Says:

    The league average points per possession hasn't changed since the '80s? Am I the only one who thinks that's odd?

  9. Anon x 2 Says:

    the hand check rule of course matters. If it was in play, it would hamper everything Nash does severely.

    While you're correct that it's about relative to league average, the point is that Nash's teams would be far closer to league average with that rule in place than without it.

  10. Neil Paine Says:

    Re: #8 - Check it out for yourself:

    League average in 2010: 1.076 points per possession

    League average in 1984: 1.076 points per possession

    Not really sure why that would be odd... They just used more possessions back then.

  11. Jason J Says:

    I don't doubt your findings in the slightest, Neil.

    I guess I figured things had changed pretty dramatically over the years. New rules. Huge shift in the way teams value the three point shot. Emphasis switching from interior scoring to perimeter scoring.

    I expected as the rate of shots taken dropped the quality of shots taken would have increased so that while today we see fewer shots, those shots would lead to more points per possession. I guess the defense kept up?

  12. Anon x 2 Says:

    Actually, you'd think it's the opposite. It's offenses kept up. As time went on we'd imagine defenses would get stronger. And this is what happened. Look at the late 90s to mid 00s. Bottoms out around 102.

    It took the development of 3 point shooting + removal of hand check to allow offenses to catch back up.

  13. Anon x 2 Says:

    one stat that would be interesting to see is league FT per possession numbers.

  14. AYC Says:

    A slower pace helps the defense. Certainly field goal % have dropped over the last 20 years along with pace; I would guess more reliance on the 3 closed the gap

  15. the.maven Says:

    I dunno about the handcheck rule and all, but weren't the '80's the era of Doug Moe, Alex English and the run & gun Nuggets and super-high scoring games?

  16. PZ Says:


    You answered your own question. The point is that Nash is a remarkably valuable offensive player, NOT that he’s good at filling up a box score. That’s essentially been the argument in favor of Nash and his MVPs.
    In 1989, John Stockton averaged 17.1 ppg and 13.6 assists. Yet, despite playing alongside an elite finisher in Karl Malone, the Jazz were actually a below average offensive team in terms of points per possession! Those Jazz teams relied on their defense. (in particular Eaton) Indeed, it wasn’t until the ball was taken out of Stockton’s hands that the team became a great offensive team. The 1998 Jazz were the best offense in the NBA and Stockton played a much smaller role.
    With Nash, there has been a remarkable correlation between his presence on the floor and offensive efficiency. Clearly Nash brings something very special on the offensive end that isn’t captured by conventional statistics. Take note of Joe Sill’s 4+ year regularized adjusted plus minus data. Nash crushed the entire NBA in terms of offensive RAPM. Kobe Bryant is a very distant second. In other words, Nash has been far and away the most “valuable” offensive player in the NBA. Defense is another matter…

  17. roberto Says:

    it´s only me or this is the most absurd and arbitrary collection of stats i´ve ever seen? it looks like you want a result and you start to try with 4,05+0,96-7,63x0,54 until you get the result you want and...taraaaaaaaaaaaaaa!

  18. Jason J Says:

    PZ - Still sounds like an issue of system to me. I think there's a danger is unilaterally assigning team success to an individual when you can't point to what the individual is doing to set him apart, why he in particular deserves all the credit. In the case of APM, that's better for assigning credit IMO than this particular exercise we're examining here, which is crediting for an entire team's performance, so much of which has to do with how a coach runs the game.

    For instance - If Phoenix sets up a high pick and roll btwn Nash & Amare, and Frye is out near the top and the other two smalls take catch and shoot spots at the short corner release points on the baseline, there's a ton of offensive potential in that set. Coach Gentry is happy. On the other hand coach Sloan might not like having his lead guard in a P&R situation with both smalls stationed below the free throw line. It's too hard to get back on defense. So the Jazz will play more conservatively in a similar play call. They will play better defense and worse offense, and it might have nothing much to do with personnel.

    I'm not saying any of that is true. Just that I think it could have a profound effect on a study of this sort.

  19. Raj Says:

    I tend to side with the idea of Nash being one of the all-time (top 10 probably) all around offensive players. If look at his offensive SPM, the only real comparables post-merger (and this includes his Dallas days) are MJ and LeBron. What stands out to me is just how abysmally bad his defense is. He's pretty much the bizarro Ben Wallace.

  20. P Middy Says:

    Nash is amazing, no doubt. Since going to Phoenix, he's shot 50-40-90 for 5 straight years (if you round up FT% for 05-06) and shot a TERRIBLE 88% in 04-05. He's done it in 35 minutes or less per game. And on top of that you've got 11 assists per game. Even with his turnover average in Phoenix over 3.5, he's still astoundingly efficient.

  21. Anon Says:

    @ #18

    Steve Nash is HISTORICALLY great on offense alright. SPM should rank him pretty high on the all-time list - among active players he's 5th in OWS and his 9.22 OWS/48 mins ranks 2nd among active players with at least 80 total OWS; ahead of guys like Kobe, Shaq, KG, Duncan, and Ray Allen. And before people say he's just a regular season player, he's also 6th among active players in the playoffs in OWS and his 8.30 playoff OWS/48 mins is ahead of anyone active with at least 8 playoff OWS total except for Dirk, LeBron, and Chauncey Billups.

    He's not a better player overall than some of the aforementioned players, but is easy to see why everyone LOVES being on a team where Nash runs the point.

  22. AYC Says:

    Nash is the most skilled offensive player in the history of the game; you can't name another player who could shoot, pass AND dribble as well. BUT, his sub-par speed and quickness means he can only excel in a certain kind of system; the type of system that doesn't typically result in championships. Dallas improved after he left....

  23. Anon Says:

    BUT, his sub-par speed and quickness means he can only excel in a certain kind of system; the type of system that doesn't typically result in championships.

    Huh? The man would excel offensively ANYWHERE. If you mean that his drawbacks on the defensive end would be a deterrent to his teams then you certainly have a point. Dallas did improve when he left, but only on the defensive end. Offensively they weren't nearly as good - they were still among the best by virtue of players like Dirk playing like an all-timer on offense.

  24. Jason J Says:

    Anon (and everyone else) - I think I approached making my point the wrong way. I'm not disputing Steve Nash's offensive brilliance at all. Obviously there are more solid numbers to back up what he brings to the table.

    I'm just questioning the validity of approaching the question of who is offensively great by looking at the percentage of teams they've been on that have been offensively great. I took Nash because he was #1 and easy to compare to a similar player (sort of) in Stockton who doesn't even make this list despite being a ridiculously great player. Maybe I should have just gone with Michael Cooper, who is at best a pretty good offensive player, and pointed out that thanks to playing every season with Magic Johnson he gets ranked 4th overall on this list. That may have been less divisive.

  25. Neil Paine Says:

    I'm with you, Jason. And you know I'm the last person to give an individual credit for team accomplishments (like rings, etc.). This post was merely intended to:

    1. Note the players who played for the best offenses over the course of their careers (whether they were the reason or not)

    2. Raise the question of where Nash stands all-time from an offensive perspective, given that he's played for far and away the best offenses of anyone in NBA history.

    Maybe it's a coincidence for Nash, maybe it's the system, or maybe it's teammates. But that's up to you guys to debate.

  26. AYC Says:

    "The man would excel offensively ANYWHERE."

    Anon, how do you know that? We have no proof that Nash would be as effective on a team that played at the pace of, say, Billup's Pistons. I'm skeptical that Nash would have the speed to penetrate to the basket on that type of team. On a fast paced team like the Suns, opposing defenses are spread out, thanks to the fast-break and all those 3-point shooters. The slowest paced team Nash played on was the 99 Mavs, who finished 14th in pace factor. Nash had the worst stats of his career that year, averaging 7.9 ppg and 5.5 apg in 31.7 mpg; Nash shot a .363 from the field and .374 from three, both career lows; his Ortg that year was 99. Every other team Nash has played on has been in the top 10 for pace.

  27. BSK Says:

    System has to come into play, even if the numbers are normalized to the league average and offered on a per possession basis. When guys aren't busting it on D, they have more energy for offense. When they're playing teams unaccustomed to the style of play, they have an advantage over the defense.

    My question would be, can Nash be successful in another system?

  28. Anon Says:

    @ #26

    Using Nash's 3rd season in the league as a young and still developing player in the league to demonstrate your example (based on 40 games/1200 minutes of play) is a bit fallacious. I always disapprove of pointing to season-by-season examples to show a change in stats without using proper statistical analysis - there was a mere difference of two points in pace factor between the '98 Suns and '99 Mavs anwyay. Based on what you're saying, shouldn't that mean Nash should put up the best numbers of his career on the FASTEST paced teams he played on? Of course not, and it's because basketball involves a million other factors that you also need to account for.

    You're right in saying that I DON'T know how Nash would do somewhere else though, so I'll amend my previous statement - all of these discussions involve "what ifs" and can't be proven either way. But you can still make educated guesses, and not to mention that it's alot more important (and this IS backed with proof) to look at how much of the team's offense a player is using when looking at his ORtg than to look at the pace of the team he plays on. Nash has usually been a somewhat mid-level player in terms of % possession rate and put up great numbers, so it's obvious that he fins ways to get his points and production at efficient levels whether the team pace is at 90 or 100.

  29. Anon Says:

    ^^^Not to mention that the win shares numbers are normalized for pace anyway.

  30. Neil Paine Says:

    Re: #27 - The problem with that is that there's absolutely no correlation between Offensive & Defensive Efficiency. I took all of the estimated efficiency numbers I used for this post (that is, ORtg/DRtg relative to the league avg. for every team going back to 1951), and checked the correlation between offense and defense... It was -0.03, for an R-squared of 0.00102. This means that literally 0% of the variation in a team's offensive ability is explained by its defensive ability, and vice versa.

  31. AYC Says:

    Anon, the "Nash was young" argument doesn't fly. The previous year, he avgd 14.9 points per 36 min, with a TS% of .566 and Ortg of 111. In 99 he avgd 8.9 p/36m, with a .471 TS% and 99 Ortg. His usage rate was actually higher in 98 too.

    '99 was the slowest paced year of the shot-clock era, with a league-wide pace under 90. It's the only year of Nash's career where he came close to the slow pace of Billup's Pistons, which finished well under 90 in pace every year.

    Neil, what is the correlation between defensive efficiency and pace? I think that's what BSK was trying to get at... I'm of the opinion that a fast pace leads to more points not just because of more shot attempts; I think a fast pace makes it to easier to score on any given possession.

  32. AYC Says:

    PS Anon, Nash's best statistical seasons HAVE come on his fastest paced teams; the suns have had a pace factor over 95 in each of his 6 seasons with the team, the best 6 years of his career

  33. Neil Paine Says:

    The correlation between pace and offensive efficiency is -0.08, for an r-squared of 0.007 (meaning pace explains less than 1% of the variation in ORtg); the correlation between pace and defensive efficiency is 0.13, for an r-squared of 0.017 (meaning pace explains less than 2% of the variation in DRtg). In other words, the relationship between pace and efficiency is so small that's it's almost nonexistent.

  34. BSK Says:

    But Neil, it seems that you are talking about ability as opposed to what actually happened. If a team has a system that essentially encourages dogging it on the defensive end, that would leave them more energy to play on the offensive end. Likewise, teams that kill themselves hustling on the defensive end often can't keep up that same intensity on the offensive end, not for long stretches at least. Now, maybe I'm missing out on something, but that is what my basic understanding of the game (as a fan and player) plus some basic logic seem to tell me.

    Ultimately, it seems we are trying to figure out cause and effect and we simply can't, because there are too many variables. Would swapping Nash with Billups have turned the Pistons into an offensive juggernaut? Maybe, maybe not. We can't know. Would Nash have ever performed as well as he did if he had a coach that demanded half-court sets? Again, impossible to know. The Suns (and Mavs, to an extent) style of play with Nash at the helm suited most of the players there well. Inversely, their skill sets allowed that style of play to flourish. We saw guys leave that team and struggle and guys join that team and struggle. Nash and that system were tailor made for each other, so it is logical that he flourished. But I don't know that we can extrapolate that level of flourishing to other systems and assume Nash's brilliance was the cause of the Suns' brilliance.

  35. Joseph Says:

    I don't think this is so much about how he improves his team (system v. player), but I think there are definite reasons to believe he's an offensive genius. Whether he slacks on defense and has more energy on offense is frankly irrelevant to his abilities as an offensive player. If that's the argument, then those who are great defensively and suck on offense should be docked points for their offensive inefficiency AS A DEFENDER. To me, if you're good at defense, you're good at defense. If you're good at offense, you're good at offense.

    I just used win shares, OWS, and DWS to create a pool of "best players" this past season. I then looked at unassisted field goals made, free throws made, and assists to figure out my own offensive ranking. At a per-min basis, Nash was first, per game, he was third.

  36. Matt Says:

    I have to side with not using Nash's first year in Dallas as a good example. He moved teams and played 10 more mpg than the year before. That was also the weird lockout year, so the sample size is a bit small. In a slow paced game he'd always be valuable as a shooter - and if he wants to penetrate opponents always have to play off him because he's a threat to shoot or pass. He's certainly not afraid to take a hit or two.

  37. AYC Says:

    I picked 99 because it was the slowest paced year of his career; is it a coincidence that that is also his worst offensive season by far? And how does more minutes hurt Nash? Typically, more minutes help a player's productivity. I think there's a correlation between the record slow pace that year and his underperformance.

  38. andy Says:

    Quick reply to the Stockton comparison. I don't want to pile onto everyone's criticisms, the Stockton comparison is thought provoking and valid. It seems to me that since ORtg is based on absolute (points and assists total), rather than relative (against league average, adjusted for pace, etc.), numbers, the results are going to be skewed because of shifting standards. I think everyone's on board with that, and that was the specific problem this post was meant to address directly.

    Regarding the larger question of individual versus system, this seems like a problem that is not unique to Nash but rather applies to the sport itself. I can't imagine any type of statistic that results from an individual play, everything's a reaction to one's teammates or opponents' decisions, whether active or passive. Assists are the most obvious example of this, and you could easily turn Stockton's numbers around and say if he didn't play in a pick-and-roll offense with Karl Malone, he'd have far fewer assists (or, conversely, Malone couldn't score so many points without playing alongside the man with most assists in NBA history).

    So what's the converse of "system, not individual"?

    My guess if someone were pressed to give an example of "individual, not system," they'd say MJ, but MJ's style changed throughout his career, too: Collins' scattered playbook vs. Winter/Jackson's triple-post; wild fluctuations in 3PA and 3P%; never winning DPOY again after Pippen became an equally awesome defender and distributed defensive responsibility; etc.

    Regarding the '99 season, a lot of other things could factor, too. New system and teammates; lockout caught him off guard (if Nash's 1999 season winds up being remembered with an asterisk next to it, it would only be joining a far wider trend across the league); he was injured a lot as a younger player; more minutes per game. It's not necessarily about pace. But there's another possibility: what if instead of positing that because A) pace was slow therefore B) he struggled, you flip it and say A) he struggled and therefore B) pace slowed? In other words, the mature Nash's successes/failures are always measured by his ability to increase the pace against opponents and thereby beat the other teams by scoring more efficiently. Other teams could stop the young and first-time starting point guard at the time and thereby slowed the whole team. Ever since the early 2000s, however, when it became clear that Nash's teams' goal was to beat you by playing their game no matter the opponent, other teams have designed entire strategies around slowing down Nash's teams but almost no teams have succeeded thus far. So in reply to the implicit hypothetical scenario "if the pace slowed, Nash would struggle," one could return with "but as long as Nash plays his game, the pace will not slow."

    This thesis was the crux of a lot of revisionist basketball history being written this past spring: "where earlier we all thought that Nash's teams played too fast to win important playoffs games, now we understand that Nash's teams should have been playing even faster in the playoffs and that as long as the Suns win playoffs games, then we need rewrite the laws of how whether or not speed can win important games."

  39. AYC Says:

    Neil, what is the correlation between pace and FG%?

  40. Neil Paine Says:

    Correlation between pace and FG% is -0.029, for an R-squared of 0.0008... Again, pace literally explains 0% of the variation in FG%.

    As much as you guys want there to be a relationship, it doesn't exist. The independence of pace and efficiency is one of the central cornerstones of APBRmetrics -- I mean, why do you think we make such a big deal about efficiency, and about measuring everything on a per-possession basis? It's because of results like the tests I just ran.

    Say it again, with feeling: there's no relationship between pace and efficiency.

  41. AYC Says:

    Hey, I'm not arguing the point; I'm just a lay-person trying to get an idea of how these things relate. But I think what you're saying applies on the team level, not the individual level. Certain players are better suited to a given pace than others. Eddie Curry can't play D'Antoni ball....

  42. pemba Says:

    playing with magic every year made micheal cooper 4th, that IS the point of this sysytem
    playing with nash evry year made amare stoudemire 2nd on this list, lets see if that keeps up in nyc, i dont think it will. nash and magic are the two best offensive players in terms of making their teams better, again the point of this system/story.

  43. andy Says:

    Also, isn't this post also a surprising testament to the efficiency of Detlef Schrempf, who is the only non-teammate of Nash or Magic in the top nine? What explains that one? Lots of 90s Sonics on there, warms the heart.

  44. Luke Says:


    why are you comparing nash and stockton by there age? what is the logic in that? I wonder who was better nash at 12 yrs old or stockton at 13yrs old? It doesnt matter at what age players are great all that matters is that they are great. What makes nash great aren't his stats. watch him play and you will see what makes him great. Kevin Johnson is a better comparison of stats with stockton than Nash.

  45. Luke Says:

    Nash is not slow. If you have seen his work over the last 10 yrs you would know that with the ball he is one of the quicker Guards in the league. His dribbling ability allows him to blow by good defenders and lay the ball in before they can recover to block the shot. Take for example the spurs put george hill (quick, long) on nash in the 2010 playoffs as a defensive specialist. but nash blew by him like 5 consecutive times before they called a time out. He regularly gets to the bucket on chris paul and deron williams. And williams might be the best defensive PG in the league.

  46. AYC Says:

    By the standard of NBA point guards, Nash IS slow. And you can't easily separate what Nash does from the system he plays in. He plays on a running team that shoots alot of threes; that means opposing defenses are spread more thinly against PHX than typical. Compare his stats in PHX to his stats before; there's a very clear improvement in his production playing D'Antoni ball. (His worst season in PHX was 09 thanks to the addition of Shaq, and new coach Terry Porter's emphasis on D over running)

  47. Joseph Says:

    So slow he set the record at the Skills Competition, had it broke by a "faster" guard, then beat the faster guard at the same competition...right? It's just the system, every team has a system. What good is speed if it's not part of the system that leads to points and wins? You can be as fast as you want, but if you can't shoot, pass, or dribble, you're sitting on the bench. Barbosa was the fastest Sun, why didn't he start?

  48. AYC Says:

    Are you unclear on what the word "skill" means? Here's a hint; it's not a synonym for "speed". Apparently you missed post #22, when I said "Nash is the most skilled offensive player in the history of the game"

  49. Joseph Says:

    Yeah, apparently I did miss it, AYC, but I appreciate your condescending tone. Next time, before I reply to a post you make, I'll verify you haven't addressed it in any earlier posts, check definitions, etymology, semantics and syntax, and try to make sure I get it just right for you. Sound good? Great.

  50. AYC Says:


  51. Anon Says:

    "I think there's a correlation between the record slow pace that year and his underperformance."

    Seeing as you have NO sound statistical proof for your claim, a rational person would certainly not give any credence to this statement until you can properly verify it.

    It's silly for you to assume this without accounting for a myriad of other factors that could go into his '99 performance besides pace. There's a SCIENCE that you can use to determine if your claim is valid, and you're not using it.

    "PS Anon, Nash's best statistical seasons HAVE come on his fastest paced teams..."

    Not quite the point I was trying to make. Rather, I wanted to emphasize that it's naive to think that there's a nice relationship where faster pace = better Nash performance, as if bumping up the pace factor automatically gave you a better Nash. Nash has had plenty of years where a slower pace yielded better play than years where he played at a faster pace.

    Clearly, you can't go year-by-year and relate directly performance to pace without controlling for other relevant variables. And once again, YOU DIDN'T DO THIS!

    "But I think what you're saying applies on the team level, not the individual level."

    I'm pretty sure Neil's findings were based on the population of individual NBA players.

  52. AYC Says:

    "The man would excel offensively ANYWHERE."

    Anon, I used the words "I think" to acknowledge that I was expressing an opinion; but since you've decided that we're not allowed to express opinions that aren't grounded in "SCIENCE", can you please provide the empirical support for your above statement (which started this whole line of discussion)? I tried to find some evidence for you, based on the tiny sample that we have of Nash at a slow pace, but it didn't turn out so well for your position did it?

    "But I think what you're saying applies on the team level, not the individual level."

    I'm pretty sure Neil's findings were based on the population of individual NBA players.

    Please note that Neil didn't disagree with my comment. Do you think all players are equally capable of playing in all types of systems? That's what your comment implies... but I don't think many people would agree with you.

  53. Anon Says:

    "Anon, I used the words "I think" to acknowledge that I was expressing an opinion; but since you've decided that we're not allowed to express opinions that aren't grounded in "SCIENCE", can you please provide the empirical support for your above statement (which started this whole line of discussion)? I tried to find some evidence for you, based on the tiny sample that we have of Nash at a slow pace, but it didn't turn out so well for your position did it? "

    Please reread post #28. Beginning of the 2nd paragraph.

    "Please note that Neil didn't disagree with my comment. Do you think all players are equally capable of playing in all types of systems? That's what your comment implies... but I don't think many people would agree with you."

    That's not I implied, nor is that what is implied when you run a regression relating efficiency to an independent variable (pace). But you really don't know too much about how these things work, do you?

    But since you like to post past quotes over and over again without reading what was written afterward, let me try:

    "Hey, I'm not arguing the point; I'm just a lay-person trying to get an idea of how these things relate."

    From #41 - your post. So is there really any point in debating you on this matter?

  54. AYC Says:

    Anon, I reread #28 and I agree; as long as the pace has been over 90, Nash has been an excellent offensive player.

    "That's not I implied, nor is that what is implied when you run a regression relating efficiency to an independent variable (pace). But you really don't know too much about how these things work, do you?"

    I know enough to see through your BS. If that's not what you were implying, then what point WERE you trying to make? You didn't state your point explicitly, probably because the implication you tried to draw is nonsensical.

  55. Neil Paine Says:

    I would say a broad implication of the team-level finding is that most NBA players can function at the same "true efficiency" level regardless of the pace. However, the population of NBA teams is much more homogeneous than the population of the players who make up those teams, so there will naturally be more outlying players at either end of the distribution who have some physical characteristic that makes their efficiency more dependent on pace than the average player. Whether you think Nash is one of those players is certainly up for debate, although I'd warn against drawing sweeping conclusions based on a sample consisting of one lockout-shortened, 40-game season.

  56. AYC Says:

    Thank you, Neil. Just to set the record straight, I'm a Nash fan; I just think he's a little overrated by certain stat-heads. I said all along that we have a small sample of Nash games at a slow pace; but the fact is, he did play unusually poorly that year; furthermore, there does seem to be a general correlation between his production and the pace of his team in any given year. People can draw their own conclusions, but don't shoot the messenger!

  57. Anon Says:

    "People can draw their own conclusions, but don't shoot the messenger!"

    I'm not shooting the messenger at all, I'm just pointing out a flaw in your reasoning. In his rookie season, Nash played on the 3rd fastest paced team in the league and had the worst production numbers in his career (and don't use the "it was his rookie season" response; you didn't seem to have a problem doing the same when you used Nash's 3rd season as an example despite the fact that you were using a season with a minuscule sample size and one in which he was still a developing and erratic player in the league). He then played for a SLOWER paced team the following season and put up BETTER numbers - but the thing you keep overlooking is that Nash has had plenty of seasons he playing at a slower pace yield better numbers. His '02 and '03 seasons were better than his '01 season offensively; his '03 was better than his '08 (the fastest paced team he was on), and also his '09 and '10. His '09 was actually among the fastest paced teams he ever played on, and he had his most pedestrian production numbers since his early years in Dallas. The following season the Suns took a step back in pace and he put up better numbers. As a matter of fact, his '08 and '09 seasons were the fastest paced team, but he had plenty of better offensive seasons nonetheless. His 2010 season is among the fastest and he still had better seasons on his "slow-as-molasses" (in comparison) Dallas teams. And these are just some of the examples where Nash played better at a slower than faster pace.

    Like I said before, you can't simply look at year-to-year numbers and draw a conclusion about Nash and pace without controlling for other variables, such as stating that Nash played his best years on the fastest paced teams - which ALSO happens to coincide with the prime production years for some NBA guards.

  58. Matthew Cornwell Says:

    How much credit should Nash receive for helping to create the pace that benefits him and his style? Surely the paces of his teams were not thrust upon him.

    Many people love to do two things - discredit Steve Nash and push win shares/PER/+/- etc. Lets not forget that Steve Nash is on pace to finish with near 120 Win Shares after next season, and only needs a few more after that to be ranked in the the top 30 all-time. Nash also ranks 6th all-time among all point guards in win shares right now and 5th in career PER too. It seems that PER and W.S. are actually tools that would benefit Nash's status and perception. Lets not forget his prodigious placements in Roland's Ratings every year.

    As far as the MVP's, of course he didn't deserve the first one and probably not the second one either. Of course Nash isn't as good as Duncan or Shaq or Kobe or LeBron. And of course he isn't as good as Stockton or probably Payton or Kidd - but he doesn't need to be to be viewed as a legit HOFer.

    The 2-MVP backlash has gone too far.

  59. Anon Says:

    Co-sign on the above post.

    AYC is entitled to his opinion, but the claims he make are often presented without rigid proof. With Nash it's faster pace = better production, even though this isn't the case alot of the time even if you just look at year-by-year stats without controlling for other variables. With Eddy Curry it's the suggestion that he doesn't play better at faster paces even though his best per 48 minute production have actually come on the faster paced teams he's played on.

  60. Ricardo Ohlinger Says:

    Hi-Ya. I suggest you have produced some compelling content and well worth sharing with my buddies. Its so hard to find somebody who understands what they are talking about. Would you please guide me to any similar websites you may publish in your response ... or if I can push my stay and seek some of your personal favorite sites online. Thank-You.