Kevin Garnett: Most valuable player of his generation

The 2021 class of the NBA Hall of Fame will be the best class in NBA history. That became solidified on Friday when Kevin Garnett announced his retirement, joining Tim Duncan and Kobe Bryant as guaranteed future inner-circle Hall-of-Famers to hang them up this offseason.

As I pointed out in August, detailed quantitative analysis indicated that Duncan measured out as a bit better and a bit more valuable than Kobe as a player. Both of them, in turn, were significantly more valuable than Amare Stoudemire, another potential future Hall-of-Famer that retired this summer. With Garnett’s retirement, the new question is, where does he fit into the mix? In fact, when you look at the best players of the post-Michael-Jordan era…which star shined the brightest?

Kevin Pelton of ESPN gave his answer to this question, arguing that Garnett was almost as good as Duncan and Kobe but not quite. This followed up his similar article from when Duncan announced his retirement, that also concluded that Duncan was the best of his generation over both the regular season and playoffs.

However, Pelton’s measures are all dependent upon manipulations of information found in the box scores. While the box score is useful, it isn’t sufficient to measure all aspects of the game…there are too many blind spots on both offense (e.g. creating offense, spacing, defensive warping, etc.) and defense (help defense, 1-on-1 defense, pick-and-roll defense, etc.) that the box score just isn’t equipped to cover.

Thus, to really answer the question of who is the best player, we need to go beyond the boxes and also look into the so-called “impact stats” that attempt to correlate a player’s presence on the court with changes in the team’s scoring margins. When defining “Most Valuable Player”, I automatically finish the sentence as “Most Valuable Player to his team’s ability to win.” While the box scores are very useful as a way to estimate/quantify a player’s mechanisms of impact and are more “reliable”, the impact stats such as the family of +/- stats are better at estimating how important a player is to his team’s fortunes and thus the more “valid” stat.

So, let’s take a look at how the best players of the immediate post-Jordan era compared in an impact stat (Regularized Adjusted +/- (RAPM, calculated by Jeremias Engelmann, the creator of ESPN’s “Real Plus/Minus stat.), as well as how they performed according to one composite boxscore stat (PER) from 2000 – 2012.

rapm_vs_per-scatter-red-labeled

This chart contains every NBA MVP winner from 2000 – 2010 and ten players overall, nine of which were drafted in the 1990s and thus fit into the generation featuring Duncan, Kobe and Garnett, plus LeBron James. Interestingly, while Duncan, LeBron and Shaq have the three highest average PER scores over this stretch, it is Garnett that measures out with the highest average RAPM score. This means that, while others may have had more impressive combinations of box score stats in the era, no other player impacted their team’s scoring margin as much as Garnett, according to this 12-year study.

Said simply, by this estimate, no player in the NBA was more valuable from 2000 – 2012 than Kevin Garnett.

Can that be right?

And if Garnett was the most impactful, the most valuable player of the 2000s, how was he doing it? Especially since the box scores have him as elite, but not the best. Why the mismatch?

Defense

The answer to both of these questions lies in the inordinate number of things that Garnett did that don’t show up in the box scores. For example, overall RAPM scores can be broken down into an offensive score (ORAPM) and a defensive score (DRAPM). In the 2001 – 2012 study, Garnett measured out as the best defensive player in the NBA over that dozen years (tied with Dikembe Mutombo) at +6.4, just ahead of Duncan’s score of +6.0. The only defensive stats found in the box scores are defensive rebounds, blocks and steals. While KG was excellent at all three (top-20 all-time in all three by totals), he did not rack up blocked shots at the rate that Duncan (or others) did, nor did he always play on teams with enough talent to post excellent Defensive Ratings. Thus, Garnett’s defensive composite box score measurements (e.g. individual Defensive Rating, Defensive Win Shares, etc.) were excellent but not always at the very top of the league.

drapm_vs_dws-scatter-red-label

But defense is about so much more than blocked shots. And when you look at all of the aspects of the defensive game…from preventing your man from scoring 1-on-1, to helping out your teammates with crisp rotations, to disrupting opponent plays, to knowing the entire defense and communicating it to your teammates on a play-by-play basis, to destroying the pick-and-roll…no player in the NBA was better at the entire defensive package than Kevin Garnett. Garnett’s amazing defensive abilities could be easily noted through scouting, but an impact stat like DRAPM is useful in quantifying how those abilities translated to his team’s fortunes. And in individual defensive impact as measured by correlating with the team’s scoring margin, Garnett was tied for the most valuable defensive player in the NBA over the first 12 years of the millennium.

Offense

Garnett’s offensive RAPM from 2001 – 2012 was also strong, one of the top-30 scores out of the almost 500 players in the study. Garnett’s +2.8 ORAPM score from that study is lower than most of the elite perimeter players mentioned in figure 1, but higher than Duncan’s +2.5 to rank him behind only Shaq, Nowitzki and Pau Gasol on offensive impact among all big men in the study.

While the box scores do a better job of covering offense than defense and would paint Garnett as great but perhaps slightly further down the list, there are still offensive contributions that aren’t measured. And again, Garnett is excellent at those things. He was the primary offensive initiator for the Timberwolves from a big man slot, and he also had an excellent mid-long-range jumper that spaced the floor and opened things up for his teammates. Historical trends show offense initiation and spacing are both very high-impact, particularly when coming from big men as evidenced most recently by Draymond Green’s unexpectedly high impact on the offense of the 2016 Warriors.

orapm_vs_ows-scatter-red-labeled

At his peak in the 2003 and 2004 seasons, Garnett was one of the most impactful offensive players in the league. Pelton points out that using ESPN’s real plus-minus (RPM) and WAR, their impact stat of choice that is also heavily influenced by the box scores, Garnett has the two best seasons on record (2003 and 2004) of all NBA players since RPM was first calculated in 2000-01, just ahead of Duncan’s 2003 and LeBron’s 2009. Garnett reached such a lofty impact in those seasons because while his defense was still dominant (ranked #7 in Englemann’s PI-DRAPM in 2003 & #3 in 2004), his offensive impact was actually right there with Shaq at the very top of the NBA (#2 in PI-ORAPM in 2003 and #1 in 2004). And again, a thorough scouting of Garnett’s offensive performance in 2003 and 2004 demonstrates the validity of his top-of-the-league offensive impact scores.

Playoffs

The one area that many hold against Garnett in these types of comparisons with his peers is the playoffs. Garnett only played in 143 playoff games in his career, 77 fewer than Kobe and 108 fewer than Duncan, so he did not have as many opportunities to rack up counting box score stats in the postseason.

Also, and importantly, Garnett’s scoring efficiency in the playoffs dropped slightly compared to the regular season while Duncan’s and Nowitzki’s stayed roughly constant or even slightly increased. This is used by many to argue that Garnett wasn’t quite as good in the playoffs as in the regular season, and that some of his historic-level peers like Duncan separate themselves from him in the postseason. You can see this in the per-100 possessions 10-year prime chart from the Duncan/Kobe/Amare article:

top-20-per-100-boxscore-stats-final

However, again, it is important to remember that the box scores barely cover the defensive half of the game and, even on offense, misses glaring high-impact aspects as well. Garnett has a documented history of huge defensive performances in the postseason, and his offensive impact is also extremely resilient and largely independent of his scoring efficiency. Thus, it is important that we also consider the available impact stats for the postseason. The best playoff impact stat available is on-court/off-court +/-. Let’s re-examine the on-off +/- scores from both the regular and post-season first introduced in the Duncan/Kobe/Amare article:

pm-scatter-primes-red-labeled-kg

The clear weakness to postseason on/off +/- is that it is a noisy stat that requires a lot of minutes across different circumstances for both on- and off-court to reduce that noise. But with the playoffs being a limited sample it is difficult to get all of the noise out. However, the marks displayed here are cumulative over more than 100 games for each of Garnett, LeBron, Duncan, Kobe and Nowitzki which helps bring down the noise enough to see some trends.

As pointed out in the previous article, Duncan maintains his clear advantage over Kobe in this postseason impact stat while opening up separation over players like Nowitzki. However, the trifecta of players that clearly measure out the best here are Shaq in his Lakers years, LeBron and Garnett. And even among that trifecta, it is Garnett whose playoffs on/off +/- is the most impressive. Some evidence that he continued to be the most valuable player of his generation in the postseason, in addition to the regular season.

The 2021 NBA Hall of Fame Class will be the best ever. Kobe was the most famous player of this generation, and Duncan the most decorated. But no player’s efforts in the 20+ years for which we have +/- data was more correlated with their team’s success than Garnett. He did more to lift his team to greater success than any player on record, and he did so with amazing longevity in a career spanning an NBA-record tying 21 seasons. While the Black Mamba was a monster, and the Big Fundamental was better still, it was the Big Ticket that was the most valuable.

This article originally appeared on