This week's mailbag features some MVP talk, testing for NBA prospects' potential, Dirk Nowitzki's debut and more.
You can tweet your questions using the hashtag #peltonmailbag or email them to peltonmailbag@gmail.com.
@kpelton @tomhaberstroh
Has an MVP ever played on a team with a negative point differential (ala Westbrook's Thunder being -0.2)?- Wes Fran (@wesfranson) March 9, 2017
Yes, and in fact it was the very first MVP. Bob Pettit won in 1955-56 playing for a St. Louis Hawks team that went 33-39 with a minus-1.4 point differential. Since then, the closest we've gotten was Kareem Abdul-Jabbar in 1975-76, when the Los Angeles Lakers missed the playoffs at 40-42 but narrowly outscored opponents by 0.1 points per game.
I've mentioned before in the context of Finals MVP that voters (in the Pettit case, the other players) were more open to rewarding players on losing teams many decades ago.
I'm not sure if that's the explanation here or whether Pettit got credit for the Hawks making the playoffs despite finishing six games worse than .500. With parity reigning outside the 45-27 Philadelphia Warriors and eight of the league's nine teams advancing, four below-.500 teams made the playoffs. And you thought this year's race for eighth in the West was weak!
"If you formed a lineup of players who had posted high usage numbers on their own while being average in every other respect, how would such a group fare? Would their aggregate shooting % increase as each player's usage fell? And where do you stand on the value of usage on its own in today's analytics?" - Kevin Fullam
My short answer is "It depends." Not all high-usage players are created equal, so if it's a group of players who are also dangerous shooters off the ball (think Kyrie Irving), the answer is different than if it's a group of high-usage players who need the ball in their hands to be effective (think DeMar DeRozan). It's also worth noting that usage is just part of the difficulty of a player's shot attempts. How often they're assisted vs. unassisted is also important.
Still, we'd expect these players to improve their scoring efficiency almost by default. Think of it this way: a player's efficiency is the average of shots that are so good they'd always take them and more difficult shots they'd only sometimes take depending on the circumstances. Putting a player in the lower-usage role should mean continuing to take the shots they'd always take with fewer of the borderline, or marginal shot attempts.
However, most of the time the reason players end up in high-usage roles is because they tend to see less drop-off taking more marginal attempts than low-usage specialists (think Kyle Korver). So the high-usage players would not benefit as much as a low-usage player would suffer being forced into a higher-usage role in the reverse of this scenario (five low-usage players).
In the early days of the APBRmetrics movement, I think too much emphasis was placed on efficient scoring as a reaction to the way volume scoring was overvalued by preadvanced stats conventional wisdom. Now, I think we've found a better balance.
@kpelton wasn't a buy-out but a late FA signing like Andersen: Bison Dele with 97 Bulls: 0.0 rs vorp; 0.3 po vorp; at least makes the list
- Rod Blunck (@rodblunck) March 3, 2017
Rod was one of several people to reply to last week's list of best buyout pickups with Brian Williams, who subsequently changed his name to Bison Dele and whose mysterious death while sailing off the coast of Tahiti remains officially unsolved.
As Rod points out, Dele wasn't technically bought out. A free agent in the summer of 1996, Dele turned down a four-year, $17 million offer to re-sign with the LA Clippers, according to Jackie MacMullan (then at Sports Illustrated). Finding the market cool after undergoing arthroscopic knee surgery in September, Dele sat out until he signed with the defending champion Chicago Bulls for the veteran's minimum on April 2.
Dele played just nine regular-season games as he worked his way into game shape, but with Chicago backup center Bill Wennington sidelined by a ruptured tendon in his foot, Dele averaged 17.7 minutes per game in the playoffs behind starter Luc Longley. His 12.4 points and 7.6 rebounds per 36 minutes were more production than the Bulls typically got from the Longley-Wennington combo, and you can make a case that the 1997 Chicago playoff roster was stronger than the previous year's team that went 72-10 as a result.
After the season, Dele signed a seven-year, $45 million deal with the Detroit Pistons. He played the first two seasons, averaging 16.2 points and 8.9 rebounds in 1997-98, before walking away from the remainder of the deal to travel the world.
"I think there is some hanging fruit in the assessment of potential through the control of parameters not visible at first sight: visual acuity (as shown in "The Sports Gene"), muscular memory, speed of reactions. I am totally sure that some surprising and underachieving player careers are caused by lack of measurement of these parameters. [Andrea] Bargnani just couldn't understand what was happening on the court. The body control of Isaiah Thomas is what makes him different to the dozens of talented [5-foot-9 guys] that have arrived into the league.
"Wingspan has only become a 'mainstream' parameter to control quite recently. Do you have any input if there are some people trying to study these parameters to assess the potential of players?" -- Daniel Canueto
There was some discussion of these types of tests during the Basketball Analytics panel at last week's MIT Sloan Sports Analytics Conference. Celtics assistant GM Mike Zarren mentioned reaction times and vision tests as examples of things that could translate to basketball. Luis Scola told an interesting story about a testing machine one of his teams had.
"It measured hand-eye coordination and we discovered that whoever's good at this is good at deflections," Scola said. "So I tested myself and we all tested. It kind of matches the deflection chart of the team. Me and a couple other players really did it every day. We ended up becoming really, really good at it. On the table of hand-eye coordination, we climbed. But we did not climb on the deflection list. That's interesting because people ended up getting good at the test. They trained for the test."
This is an example of Goodhart's Law, which concludes that "When a measure becomes a target, it ceases to be a good measure." So I'm a little skeptical of how much utility there is for such measures.
It's also possible that they might tell us something, but not more than we can glean from past player stats. Wingspan is a good example of that. It predicts NBA block percentage better than simply using height, but neither of them predicts nearly as well as college block percentage.
"Dirk Nowitzki's first game was in Seattle (2/5/99)...any chance you were in the building?" - Jonathan Dennis
Sadly, no. While my family had season tickets, I wasn't there. I'm pretty sure I was away at -- of all things -- a math competition. So I missed Dirk's debut.