Retiring Stars: Breaking Down Kobe vs. Duncan (plus Amar’e too!)
"“Walking Living Legend, man I feel like Kobe!I just dropped 60, man I feel like Kobe!” —Kanye West, “That Part”"
Kobe Bryant announced last year that he was retiring, then went on a retirement tour all season and finished off with a 60-point finale that touched the imagination one more time and cemented his status to the general public as the greatest player of this generation.
Tim Duncan played the whole season and playoffs without comment, held up a finger as he walked off the court in his final playoffs game then announced his retirement via a statement this summer without even attending his retirement press conference.
Amare Stoudemire was unable to secure an NBA slot this season so he retired, only to sign on to play in Israel next season.
All three are NBA legends that retired with the same personality that they wore through their careers. Kobe in the limelight, Duncan in his quiet solitude, and Amare with his never-give-up attitude. But as they ride off into the sunset, the question of their place in history has been the last major story for each of them.
There are many that believe Kobe to be the best player of his generation, the Michael Jordan of his generation, one of the five greatest players in history, and I’ve even seen it argued that he’s the greatest of all-time
On the other hand, there are also many that have Duncan as the greatest of his generation, the best two-way player in modern NBA history, and arguably the third-best NBA player of all-time.
If you follow the links in the above two paragraphs, you’ll notice differences in the approaches used in the articles that help explain the different conclusions. The articles supporting Kobe as the best tend to be very narrative driven (e.g. he’s “an assassin”, “he’s huge in the clutch”, he wants it more than anyone else) or to rely on more “shallow” statistics like a ring count, numbers of accolades, or points scored. Meanwhile, in the pro-Duncan articles you see a lot more quantitative support of arguments along with on-court “advanced stats” like WARP, Win Shares, VORP, and RPM.
In fact, if you build your Kobe search around statistics, you’ll find a bunch articles on the subject…arguing against him. That he was never the best player in the NBA in any season, Michael Jordan was “way” better, that his quantitative footprint just couldn’t match up to the all-time greats, even (HOT TAKE WARNING) that he’s only an All-time NBA great if you don’t do the math.
Hmm.
As you may know, I’m an engineer and a scientist by trade, so I’m very into quantitative arguments (if you clicked on all those links above you even saw me on one of them throwing numbers around). However, I’m also a basketball nerd so I’m heavily into scouting and using the so-called eye-test along with the numbers. And finally, when making a quantitative argument I prefer to be able to look at a full cross-set of the available statistics to form the most robust opinion that I can. So, let’s do it. Let’s jump into this Kobe vs Duncan debate.
While we’re here, let’s see what our methods say about Amar’e as well since his Hall of Fame credentials are a bit more ambiguous.
More from Nylon Calculus
- Nylon Calculus: Reviewing preseason predictions for the Western Conference
- Nylon Calculus: Reviewing preseason predictions for the Eastern Conference
- Nylon Calculus: LeBron James and the slam dunk aging curve
- The Whiteboard: Why has the NBA seen so many 50 point games this season?
- Nylon Calculus: 15 early season predictions for the NBA Western Conference
Before diving fully into the numbers, a quick scouting reports on the mechanisms of action for each player.
Duncan made his impact by combining the stifling defense of the greatest defensive centers of all-time with a diverse offensive skill-set and sneaky athleticism that let him thrive at either big man position. Kobe dominated offensively by putting incredible pressure on opposing defenses that helped make the game easier for teammates, but played a defensive game that was aesthetically convincing (and could ramp up to great on-ball defense) but too-often lacked the focus, energy and rotations to come anywhere near his reputation. Meanwhile, Amare was a devastating finisher on offense but did not generate offense (for himself or teammates) well enough to be a true primary option and was pretty terrible for the most part on defense.
Box Scores Stats
On to the numbers. First, the basic box score stats.
Duncan and Kobe have both been argued as potential top-10 players of all-time, and definitely in the top-20. So here, I gathered the box score counting stats (normalized per 100 possessions) for most of the top players of all-time as ranked in this RealGM project. Possession-normalized data is important because the pace in Kareem Abdul-Jabbar’s day was frantically faster than what Duncan and Kobe were facing in the early 2000s. More possessions means more opportunities to accumulate boxscore stats.
Unfortunately, since pace data isn’t fully available prior to the early 70s, the greats of the 60s and before aren’t included here. I look at the best 10-year stretch for each player (as opposed to full career) so that I can get a handle on who played best in their primes as opposed to a measure of longevity. The one exception was Amare, who was injured so often that he never had a full decade prime, so only seven seasons (six of which were healthy) are included for him. I’ve also included playoff stats for comparison to address any arguments about players “stepping up” when it matters most:
I know, wall-of-text/numbers. I don’t want to spend too much time here, but one of my quick takeaways is that, when normalized for pace, Duncan’s scoring volume suddenly looks awesome, very comparable with all-time scoring leader Jabbar or offensive wiz Larry Bird. Amare’s scoring volume looks better still, up near the premier scoring wings, and Kobe is in the mix with Shaq and LeBron right behind Jordan near the top. In scoring.
However, Duncan is also dominant in several other categories, primarily defensive ones like rebounding and blocks, while Amare and Kobe don’t particularly shine outside of scoring. This wall of numbers, if anything, helps illustrate how diverse but overall awesome the stats are for the best players in history. It would be hard to rank the players based on what we see here.
Next, composite boxscore stats. While each stat is different, there are some commonalities across them, such as the tendency to heavily reward scoring efficiency. Essentially, these types of stats attempt to break that wall of numbers from Table 1 into a more easily consumable 1-number stat. Let’s choose a couple of these stats and plot how our NBA greats compare versus each other, again in both the regular season and the playoffs.
This approach can be illustrative, and can help with a quantitative sorting process. For example, we see that Duncan is among the top-7 in both regular and postseason of each of PER and Win Shares when compared to other top-20 GOAT candidates and that he outpaces Kobe (and Amare) in all three. If these measures are trustworthy, this would speak very well to Duncan’s candidacy on the GOAT lists and ascendancy over Kobe within their generation (similar story to the quantitative articles in the links above).
However, it’s not that cut-and-dry. Take a look at how Kobe and Amare rank in those stats. In Win Shares they almost overlap in both the regular and the postseason, while in PER Amare’s postseason mark is actually HIGHER than Kobe’s…and Magic Johnson’s, and Julius Erving’s, and Larry Bird’s…and MUCH higher than teammate Steve Nash’s. Conservatively, nobody has Amare Stoudemire as a better player than any of those guys, and most had Nash as the most important Sun. So, either the composite box score stats are showing us that we’ve all underestimated Amare for years, or there is more going on. I argue the latter.
Because while the box scores are useful, they don’t completely span the game. On offense there is no boxscore which fully measures creating offense (both points scored and assists are lacking in identifying who created the scoring opportunity), there is no generally accepted stat for generating spacing, there is no stat for putting pressure that warps an opposing defense and makes it easier on teammates. And the problems with measuring defense output are even larger!
The +/- stats
Thus, we need to move into some of the non-boxscore quantitative approaches to flesh out our analyses. The +/- stats are the most developed of the so-called impact stats, with available information even back to the mid-90s. Basketball-Reference has on-court/off-court +/- stats available for both the regular season and the playoffs back to 2001, covering most of the prime years for all three of our players under consideration. Figure 3 shows us how the on/off +/- averages of our current list of players compares to what we saw with the boxscore stats:
Interestingly, Duncan maintains his clear advantages over both Kobe and Amare in both the regular season and the playoffs. Duncan’s regular season on/off +/- is right there with Dirk Nowitzki’s, but he blows Dirk away in the postseason and trails only the triumvirate of LeBron James, Lakers Shaquille O’Neal and Kevin Garnett in both the regular and postseason. Similarly, Kobe seems to find his level as well, right in the midst with Nash and Dwyane Wade in the regular season both surpassing both (as well as Dirk) in the postseason, just behind Duncan.
But this impact-based analysis really hammers home how out-of-place Amar’e Stoudemire is in this company. While his scoring volume and efficiency boosted his boxscore marks into the inner circle Hall of Fame range, his lack of defense and inability to generate team offense shows up clearly in his +/- footprint. Amar’e was well back of the rest of the group in the regular season and even further behind in the postseason. Here, we see the numbers to support the generally held position that Nash was clearly the more important Sun to the team’s success, and that Amar’e just isn’t on that mega superstar level.
This effect is only solidified when one looks at the more polished, regressed versions of the +/- stats. Jeremias Englemann’s 12-year RAPM study from 2001 – 2012 (had Duncan tied with LeBron for second in the NBA over that stretch with a +8.5, Kobe a bit further back but still near the top at 12th overall with a +5.5, but Amare well outside of the rankings with a score of only +1.5.
Conclusion
Kobe Bryant is a living legend, as probably the most popular and polarizing player of his generation. Amar’e Stoudemire did stand tall and talented as a perennial All-Star when healthy that may someday have a Hall of Fame plaque in Springfield. But of the three that hung it up this offseason, it was clearly Tim Duncan that played at the highest level. He was more individually productive in both the regular season and postseason in the box scores AND his presence correlated with more team success than either of the other two. While the Black Mamba was a monster, the Big Fundamental was even better.