Feb 6, 2015; Oklahoma City, OK, USA; New Orleans Pelicans forward Anthony Davis (23) handles the ball against Oklahoma City Thunder forward Serge Ibaka (9) during the second quarter at Chesapeake Energy Arena. Mandatory Credit: Mark D. Smith-USA TODAY Sports
With the All-Star break having just finished, I figured it’d be a good time to look at which players have improved from last year and which ones have seen a dropoff. Specifically, we’ll look at improvements/declines in RPM[1. It’s important to note that last year’s RPM was multi-year with RAPM and box score priors while this year’s RPM only has box score priors. So there are going to be larger increases/decreases in RPM.]. Let’s first look at RPM this year compared to last year:
Ideally, you’d like to be on the right side of the graph (means you have a positive RPM). Many of the players in the bottom right quadrant are younger players who were/are likely to improve. In the top left quadrant, we can see some of the players who’ve gone from making a positive impact with their teams to a negative impact. Obviously role and team play a big part in this. But in some cases, such as long time RAPM star Nick Collison, age is just as good of an explanation. Collison’s MPG are roughly the same as last year but he has a career worst 47.2% TS% and his DRPM is not as high as in years past. Others, such as Vince Carter and Channing Frye have changed teams from last year. Both players are shooting worse and it’s possible that the change in teams has negatively affected them[2. age would also be an appropriate reason for Vince’s decline- though even that seems hard to explain nearly a 13% drop in 3PT%]. Let’s take a look at the biggest risers and fallers in RPM:
Middleton is the biggest riser in RPM followed closely by Elliot Williams. Williams was on a really bad 76ers team last year[3. The 8th worst in NBA history.] and made the move over to the Jazz this year where he has played a whopping 42 minutes. He’s still a negative player and the large increase in his RPM is due to 1) being THE worst player in the NBA last season and 2) that low minute total. But enough about Elliot Williams[4. I challenge everyone to find any article in existence that has spent this much time talking about Elliot Williams.], let’s talk about some players who have made some ACTUAL improvements like the previously mentioned Khris Middleton. Middleton is just 23 years old[5. How about the young Bucks? #OwnTheFuture] and has been improving every year he’s been in the league. His 3.83 DRPM leads all SGs and per SportVU, opponents are shooting just 39.6% when he’s defending the shot. Opponents are shooting worse versus Middleton at each type of location. And according to Synergy stats, he’s in the 80.7 percentile for isolation defense. Meanwhile, on offense, he’s shooting 42.3% from downtown and provides excellent floor spacing for the Bucks. At the moment, he might just be the best 3-and-D player in the league[6. There aren’t many players who combine 40+% 3-point shooting with very good defense. What does that list look like right now?].
Other players improving like Kyrie Irving, Anthony Davis, DeMarcus Cousins aren’t surprises at all. But how about Tyreke Evans? His box score numbers are very very similar to last year’s and yet he’s had one of the large improvements in his RPM. So what gives? Well, a little digging and you find out that Evans has played the most minutes on the Pelicans with Anthony Davis[7. The Tyreke-AD 2-man combination ranks 29th in total minutes together.]. So that ought to help a bit.
Another interesting name — Meyers Leonard. An honorable mention in Zach Lowe’s excellent piece on big men, Leonard’s TS% has risen by nearly 20 percentage points points. 20![8. Granted, he only played 355 minutes last year but there was clearly a reason for that.] Additionally, if we stretch the boundaries on what should qualify for minimum Opp FGA at the rim, Leonard is 2nd in Opp FG% at the rim. And he fares well in Seth’s rim protection metrics too. At just 22 years old, he should continue to improve.
One last note, you may have noticed that there’s a general linear trend in the second visual. The R^2 isn’t strong (0.20) but RPM is statistically significant. This upward linear trend suggests that the league has improved. Notice that there seems to be more players in the upper right quadrant than the lower right one. Although the same isn’t necessarily true for the left side of the graph. This suggests good players are getting better while bad players are both improving and declining.