Manu Ginobili: A Shot Chart Commemoration

Manu Ginobili retired last week after a 16-year, 4-championship career that will likely see him land in the HOF and was undoubtably great. When a player has a run as long and as durably impactful as Manu’s, there’s a natural tendancy to remember him as you last watched him play and to forget the explosive, luxuriously maned 25-year-old rookie. To remind myself of the player that Manu was been throughout his career, and to get a better understanding of his trajectory as a scorer, I’ve put together a statistical summary of where on the court Manu got his shots, how efficient he was from different spots, and how this changed over his career.

First, a few preliminaries. Manu’s greatness defies a cursory statistical glance: he famously spent much of his career coming off the bench, he never averaged 20 points per game in the regular season, and his other raw boxscore stats were similarly solid-but-not-spectactural. But he was a dazzling passer and great defender, and he played his best basketball in the most meaningful playoff moments. Relevant to this post, he was also an extremely efficient scorer, with a career TS% of .582 and multiple seasons over 0.600. (See Figure 1 here for comparison. The average player has a TS% of around .53, while the average elite scorer is around 0.57; Manu holds up very well.)

Field Goal Distributions: Detailed Breakdowns

To get a better feel for how these shot patterns translate into scoring, Figures 2 through 5 show in greater detail where on the floor Manu got points and how efficiently he did so (made again with ggplot2 and ggExtra::ggMarginal() to add x-axis histograms).

Manu's Detailed shot data, 2004-05 season; main panel shows points scored on every shot attempt, by distance; top panel shows number of shots taken at each distance. Source: stats.nba.com API

Figure 2: Manu’s Detailed shot data, 2004-05 season; main panel shows points scored on every shot attempt, by distance; top panel shows number of shots taken at each distance. Source: stats.nba.com API

Figure 2 shows Manu’s 2004-05 season, a breakout offensive year and one of his most efficient scoring campaigns. The plot here corresponds to the trends we just saw above: Manu got a ton of shots at the rim, which he converted at a high rate, and with relatively few shots from distance.

Manu's Detailed shot data, 2007-08 season; main panel shows points scored on every shot attempt, by distance; top panel shows number of shots taken at each distance. Source: stats.nba.com API

Figure 3: Manu’s Detailed shot data, 2007-08 season; main panel shows points scored on every shot attempt, by distance; top panel shows number of shots taken at each distance. Source: stats.nba.com API

Skipping ahead to Manu’s 2007-08 (Figure 2), we see his most statistically impresive scoring season, in which he had career highs in points per game, points per 36, and TS% (he put up astronomically efficient numbers in 2011-12, but only played 34 games that year). He achieved this by still getting a large number of high-percentage looks at the rim, taking a few more long mid-range shots that he managed to hit at a relatiely high clip, and taking substantially more threes than he had in prior years, hitting them at a career high .401.

Manu's Detailed shot data, 2010-11 season; main panel shows points scored on every shot attempt, by distance; top panel shows number of shots taken at each distance. Source: stats.nba.com API

Figure 4: Manu’s Detailed shot data, 2010-11 season; main panel shows points scored on every shot attempt, by distance; top panel shows number of shots taken at each distance. Source: stats.nba.com API

Moving ahead to the 2010-11 season (one of his two All-Star seasons; Figure 4), Manu’s shot selection continues to drift noticeably outward, with a much higher proportion of his looks coming from three, and fewer of his two-point attempts clustered near the rim. While his three-point percentage dipped a bit relative to prior years, his greater proportion of field goal attempts from three still had an upward effect on his overall efficiency, and his two-point percentage also stayed high, resulted in a still-excellent 0.581 TS%.

Manu's Detailed shot data, 2016-17 season; main panel shows points scored on every shot attempt, by distance; top panel shows number of shots taken at each distance. Source: stats.nba.com API

Figure 5: Manu’s Detailed shot data, 2016-17 season; main panel shows points scored on every shot attempt, by distance; top panel shows number of shots taken at each distance. Source: stats.nba.com API

Finally, in 2016-17, Manu’s second-to-last campaign, his drift to the perimeter reached its maximum, with a majority of his looks coming from three, and his two-point attempts largely scattered away from the rim (corresponding to a career-low two-point field goal percentage). As part of this perimeter drift, he also had a career-low free throw rate of 0.244 (compared to a high of 0.567 in 2004-05, and a career average of 0.398). Still, even in the least efficient year of his career, Manu’s 0.392 three point percentage kept his true shooting percentage at a respectable 0.532.

Field Goal Distributions: Comparison Players

As we’ve seen, Manu’s career as a scorer was characterized by a hyper-efficient Moreyball emphasis on threes and forays to the rim. How does this compare with other top scoring guards during Manu’s career? A useful approach to answering this question is by making a “ternary” plot, placing individual player-seasons according to the percentage of field goal attempts that came from the restricted area, outside of the restricted area but inside the three-point line, and outside the three-point line.

Ternary plot showing Manu (red) and comparison (blue/gray) seasons, with position governed by proportion of fg attempts from 1) within restricted area, 2) outside of RA but inside 3-pt line, 3) from 3. Source: stats.nba.com API

Figure 6: Ternary plot showing Manu (red) and comparison (blue/gray) seasons, with position governed by proportion of fg attempts from 1) within restricted area, 2) outside of RA but inside 3-pt line, 3) from 3. Source: stats.nba.com API

Figure 6 does this, made with ggtern package. All of Manu’s seasons are shown in red, compared with the 48 top-scoring guard seasons since Manu entered the leage in 2002 (LeBron included). Some specific comparison players that are especially like or unlike Manu in their distribution of shots are highlighted in blue. What we see is that Manu has always been a particularly Morey-friendly scorer. His seasons fall in a band from contemporary LeBron, through Harden and Lillard, to Steph – players with varying inclinations to be very far from or very near to the rim, but who all take less than a third of their shots from an intermediate two-point range.

At the opposite end of the spectrum, we see players like Wade, Iverson, and Kobe – guys whose transcendant skill allowed them to thrive as scorers, despite living off the advanced analytic gruel of NBA shot attempts. And at the very opposite end of the scoring spectrum is DeMar DeRozan, who in his peak scoring year of 2016 took nearly three quarters of his shot attempts from mid-range. This presents a particular irony given his trade to the Spurs – many of Manu’s 1300 minutes from last year will be going to possibly the least similar scoring guard in the league.

Conclusion

Manu’s extraordinary efficiency as a scorer is reflected in his shot chart; he was ahead of his time in clustering his looks where he was most efficient – near the rim and from three. While the proportion of these looks shifted toward the perimeter as he grew older and lost a bit of explosiveness, he maintained a distinctly modern split between shots at the rim and shots from three, and he remained a very solid scorer up until the end. I’m going to miss watching him play.