Why doesn't Giannis Antetokounmpo's dominance of the box score match his advanced stats this season?
The two-time MVP is posting the best per-game stats of his career, averaging 31.2 points per game (fourth most in the NBA) and 11.9 rebounds per game (second) while leading the Milwaukee Bucks to the NBA's best record despite playing without Khris Middleton for nearly two-thirds of the season.
As a result, Giannis is a close third in odds to win MVP at Caesars Sportsbook at plus-600, trailing clear favorite Nikola Jokic and Joel Embiid.
Yet, some of the same player value metrics that rated Antetokounmpo as the league's best player when he was winning MVP no longer have him nearly as valuable this season. What explains the disconnect?
Throughout the NBA season, I answer your questions about the latest, most interesting topics in basketball. You can tweet me directly at @kpelton, tweet your questions using the hashtag #peltonmailbag or email them to peltonmailbag@gmail.com.
In addition to the Giannis question, this week's mailbag also tackles the highest-scoring group of top-five players by name and how players' roles change when they go from college basketball to the NBA.
"Why don't all-in-one stats have Giannis Antetokounmpo as one of the league's handful of most valuable players this season?"
-- Jackson
Perhaps the clearest example of this trend is estimated plus-minus (EPM) from Dunksandthrees.com. As recently as 2020, Giannis' plus-8.9 EPM led the NBA during his second MVP campaign. Antetokounmpo was third last season behind Jokic and Embiid. However, this season, his plus-6.8 EPM ranks eighth.
Something similar is true with Basketball-Reference.com's box plus-minus, which only includes box-score data. After leading the league in 2019-20 and finishing second behind Jokic the past two seasons, Giannis is down to sixth. Antetokounmpo is even lower (13th) in FiveThirtyEight's Raptor, granting that he hasn't rated as well by this model historically (just third in 2019-20).
The exception is the LeBron metric from BBall-Index, where Antetokounmpo does rank second, albeit with a lower rating than any of the previous four seasons.
The easiest part of the explanation is that Giannis' production has come in a very different manner than the past four seasons. With Middleton absent much of the season, his 39% usage is both a career high and the highest in the NBA. Yet Antetokounmpo's .598 true shooting percentage (TS%) is his worst since Mike Budenholzer's arrival in 2018-19, the same season as his emergence as an MVP candidate.
Historically, the rule of thumb is that increasing a player's usage rate by one percentage point should cause about a .005 drop in their true shooting percentage, holding ability constant. So Giannis' decline from last season (.035) is bigger than we'd expect simply because of his larger role (.020) -- before we account for the league as a whole scoring more efficiently this season.
Relative to league average, per Basketball-Reference.com, Antetokounmpo's TS% has gone from 12% better in 2021-22 to just 3% this season. As a result, despite scoring at a career-high clip, Giannis has created less value as a scorer than the past four seasons.
The other factor is perhaps more surprising to those not watching Giannis on a nightly basis: His steal and block rates are both career lows. Playing typically alongside Brook Lopez after being Milwaukee's primary rim protector with Lopez sidelined much of last season, Antetokounmpo is blocking opponent 2-point attempts half as frequently as he did last season -- and still more than 33% less often than he did in previous seasons with Lopez.
For most of these metrics, box-score stats are only half the equation. They also factor in how teams do with a given player on the court vs. off, adjusted for their teammates and opponents. That's where you'd figure Antetokounmpo dominates given the Bucks' record and their injuries. Yet this season has actually seen Milwaukee have the smallest differential with Antetokounmpo on the court vs. off (plus-7.3 points per 100 possessions) since 2016-17, via NBA Advanced Stats.
In this case, I think the metrics are undervaluing Antetokounmpo to some degree. The Bucks have shot a bit better on 3s with Antetokounmpo on the bench (37%) than on the court (36%). While it's tempting to say that's because of his own poor 3-point shooting (28% this season), Second Spectrum analysis suggests Milwaukee gets better looks with Giannis on the court.
It's interesting in that context to look at Antetokounmpo's regularized adjusted plus-minus (RAPM) from NBAshotcharts.com. The site features two versions of single-season RAPM, and Giannis does better in the one that incorporates 3-point luck (he ranks seventh in the league, behind only Jokic, among serious MVP contenders) rather than the one that does not (28th).
All told, I think Antetokounmpo's metrics suggest he should be on the MVP ballot, but not quite among the top tier of contenders.
"How differently do players play stylistically when they go from the NCAA to the NBA?"
-- Kevin
This question popped up when I was watching college tournaments this week for scouting purposes. So many NCAA offenses remain post-centric because post-ups are relatively more efficient given the size mismatches at the college level and the limited floor spacing for pick-and-roll basketball.
Using Synergy Sports tracking data, I decided to look at how the distribution of plays finished by this year's rookies changed from last season in college. Although the results are predictable, they're interesting, nonetheless.
Naturally, rookies tend to have less opportunity with the ball in their hands in the NBA. The percentage of plays they finish as a pick-and-roll ball handler, in isolation and in the post all decline, while plays as a finisher (most notably cuts, putbacks and spot-up opportunities) all increase relatively.
By far, the biggest decline comes in the post. On average, this year's rookies (minimum 100 plays finished in the NBA) finished 10% of their plays on post-ups last season in college. That's down to 2.8% as NBA rookies.
Among this group, eight players (Paolo Banchero, Jalen Duren, Christian Koloko, Keegan Murray, David Roddy, Jabari Walker and Jaylin Williams) finished at least 16% of their plays in the post in college. No rookie has reached that mark in the NBA. Murray has seen perhaps the most dramatic change in his usage, going from 21% of his plays in the post and 11.5% on isolations to just 4% of his plays total between the two in the NBA.
This doesn't just affect post players. Ball handlers play a different role in college too, spending more of their time entering the ball to the post instead of playing spread pick-and-roll as they will in the NBA. Keep that in mind when you're scouting players over the next few weeks.
"Is Kevin the first name with the highest-scoring top five?"
This question was inspired when I noted that Kevin Durant had passed Kevin Garnett as the all-time leading scorer among Kevins back in November and posted the top five, all of whom have at least 14,000 points.
1. Kevin (102,128 through Thursday)
Kevin Durant 26,764; Kevin Garnett 26,071; Kevin McHale 17,335; Kevin Willis 17,253; Kevin Love 14,705
As it turns out, Kevins are the leading top five in large part because of Durant and Garnett -- the only two players with the same name in the all-time top 40 in scoring. Despite the fact that there have been relatively fewer Kevins than other names near the top of these rankings, there's also strong depth. If Love can get to 15,000 career points, Kevins will be just the second name with at least five players at that mark, joining the runner-up name.
2. Bob/Robert (99,209)
Robert Parish 23,334; Bob Pettit 20,880; Bob Lanier 19,248; Bob McAdoo 18,787; Bob Cousy 16,960
Including all versions of Bob and Robert, this is the only name with five players over 15,000 points -- actually six, with Bob Dandridge left out of this group -- and two who have scored at least 20,000. Alas, with just three active players (Robert Covington, Bobby Portis and Robert Williams III), this group is unlikely to add to the total any time soon.
3. Michael/Mike (94,089 through Wednesday)
Michael Jordan 32,292; Michael Finley 17,306; Mike Mitchell 15,016; Mike Conley 14,777; Mike Bibby 14,698
Because of Jordan and the ubiquity of Michaels, this would probably be most people's first guess as to the name scoring leader. Alas, there's a huge drop-off between Jordan and Finley, who's barely scored half as many points. Despite that, Michael could move into second if Conley can score at least 1,202 more points in his career. That would require another couple of seasons at his current pace.
4. John/Johnny (86,268)
John Havlicek 26,395; John Stockton 19,711; John Drew 15,291; Johnny Newman 12,740; John Long 12,131
The Johns are a top-heavy group, with Havlicek and Stockton both in the top 51 in all-time scoring but just three players in the top 200. To move up, the Johns will need John Wall (currently stuck on 12,088 points, 43 behind Long) to resume his NBA career. Otherwise, it's going to be a while for John Collins to join this group.
5. Paul (86,246 through Thursday)
Paul Pierce 26,397; Paul Arizin 16,266; Paul George 16,238; Paul Millsap 14,536; Paul Westphal 12,809
With George still going strong, the Pauls are a group with upward mobility despite limited representation. Reed is the only other active Paul, but George should be able to score the 1,166 points necessary to pass the Johns.
6. Chris/Khris (85,276 through Thursday)
Chris Paul 21,545; Chris Mullin 17,911; Chris Bosh 17,189; Chris Webber 17,182; Khris Middleton 11,449
Along with the Kevins, Team Chris is the other one with four players over 17,000 career points. There's a huge gap after that, though Middleton continues to close it with his play.
7. James/Jim/Jimmy (81,770 through Thursday)
James Harden 24,529; James Worthy 16,320; James Edwards 14,862; Jimmy Butler 13,369; Jim Jackson 12,690
With two of the top five still active, this group could also catch the Johns if Butler and Harden stay healthy and extend their careers.