APBRmetrics

I guess the WOW site has an article on this topic, if interested.

Mike G wrote:First-place votes: George 52, Vasquez 13, Sanders 10, Vucevic 10, Holiday 10, Asik 9, Harden 8; another 8 guys got 1 vote.

Sanders, Vucevic, and Harden made my top 10. Vasquez top 20.
WS/48 has Vucevic and Vasquez stalling, and Harden dropping significantly.

My man Lance Stephenson got 1 nth-place vote.

Revealing one of the major flaws in WS. I wouldn't say Harden has got much better, but he certainly hasn't got worse, and all in all I think he has improved a decent amount.

Interestingly whilst most metrics have George having basically stayed the same, my recently finished metric based on what gets you played (i.e. time-allocation within a team of players) shows George as having made a pretty big stride from around number 75 to about number 30 in the league, according to what I am calling BCR (blind coach's rating). I hope to put up a post on the model soon.

George finished the season badly. In his last 5 games, he shot .283, 6-28 from the arc, and the Pacers lost 4 of the 5 . So he finished with a pretty mediocre .531 TS%, after looking like an elite scorer most of the year.

Still, he jacked up his Usg% from .193 to .235, his Ast% from .136 to .196, and just a nominal uptick in TO%.

His ORtg (104) was just the norm for the Pacers, who were 20th in the league in ORtg.
I see his FG were about 45% unassisted. Between 3 and 22', they were over 70% unassisted.
His eFG% at home was .528, and on the road just .446 . This seems pretty dramatic.
3-ptrs: .413 at home vs .294 on the road! -- about 2 ppg difference right there.

John.
Wall.

That's the most improved player in the NBA, right up there. I'm actually shocked to see him mentioned nearly nowhere in this conversation.

Here are the 6 guys in the NBA whose end of year IPV is more than 2 greater than their prior entering the season. I think this is a good way to frame this discussion as IPV takes into account production, playing time, etc. Anyway, we have Wall at the top of the pack here by a mile, whereas everyone else in our list is very familiar to readers of the final posts in this thread.

John Wall (prior = -1, ipv = 2.2, diff = 3.2)
Larry Sanders (prior = -1, ipv = 1.4, diff = 2.4)
Kemba Walker (prior = -2.4, ipv = -0.1, diff = 2.3)
Eric Bledsoe (prior = -1.5, ipv = 0.7, diff = 2.2)
Tristan Thompson (prior = -4, ipv = -1.8, diff = 2.2)
Greivis Vasquez (prior = -2, ipv = 0.1, diff = 2.1)
note: Lance Stephenson (prior = -2, ipv = -0.3, diff = 1.7) was 3 players down on the list, just under Bismack Biyombo (prior = -4.4, ipv = -2.5, diff = 1.9) and Brook Lopez (prior = -0.2, ipv = 1.7, diff = 1.9).

As I said, I'm really surprised that no one else mentioned Wall. I guess it could be because his raw boxscore stats look so similar to previous years? Or maybe due to the missed playing time? Or maybe the pedigree (and/or being overrated previously) lends him to being looked at as 'too good' for an improvement list? But whatever the reason, he went from not very good yet to awfully darned good during this past season.

I have Wall #35, which is above Paul George.
Does his 'prior' of -1 mean he was a below-avg player last year? Because I think he was well above avg.

He raised his TS% by a modest .019, while shooting a lot more and cutting down his TO. All well and good, but others improved more -- per minute and times minutes.

Per minute, among players with at least 1500 minutes, the clear leader is Andray Blatche; from .68 to 1.59 eWins/484, a gain of .92 . Wall, at +.36, ranks 19th in improvement.

Among players going at least 32 mpg, Wall ranks 6th in per minute jump. Well behind Westbrook's +.49

If he'd played 82 games like he played his last 21 -- including his 36 mpg -- he well may have been #1.

I imagine IPV is too laggy to judge Wall, given his injury prone nature. He has improved, but he was good last year also.

Here are the ASPM/VORP deltas for players of interest:
Any of the top 4 on that list look like good options to me, or John Wall. Five good choices... Paul George is better, sure, but certainly nothing like the improvement of Harden or Curry, both of whom are as good as George.

DSMok1 wrote:Here are the ASPM/VORP deltas for players of interest:

Code: Select all

                    2012            2013           Deltas
                    ASPM    VORP    ASPM    VORP    ASPM    VORP    AVG
Lance Stephenson    -4.0    -0.1    0.9     1.9     4.9     2.0     3.5
Eric Bledsoe        -2.4    0.1     1.5     1.5     3.9     1.4     2.7
Stephen Curry       2.7     1.4     4.3     5.0     1.6     3.6     2.6
Kemba Walker        -1.6    1.0     1.4     2.7     3.0     1.7     2.4
Tristan Thompson    -3.1    0.1     -0.5    1.2     2.6     1.1     1.9
James Harden        2.8     3.7     4.8     5.4     2.0     1.7     1.9
John Wall           0.7     3.0     4.3     2.7     3.6     -0.3    1.7
Larry Sanders       -0.1    0.6     1.8     2.0     1.9     1.4     1.7
Nikola Vucevic      -1.2    0.5     0.5     1.8     1.7     1.3     1.5
Brook Lopez         0.1     2.9     2.2     2.6     2.1     -0.3    0.9
Greivis Vasquez     -0.8    1.3     0.4     1.8     1.2     0.5     0.9
Chandler Parsons    0.4     2.1     1.2     2.5     0.8     0.4     0.6
Paul George         2.8     3.7     3.2     4.2     0.4     0.5     0.5
Bismack Biyombo     -2.6    0.3     -1.9    0.3     0.7     0.0     0.4
Jrue Holiday        0.8     2.8     0.9     2.4     0.1     -0.4    -0.2
Omer Asik           1.1     1.3     0.4     1.7     -0.7    0.4     -0.2
Serge Ibaka         2.2     3.1     1.7     2.5     -0.5    -0.6    -0.6

Any of the top 4 on that list look like good options to me, or John Wall. Five good choices... Paul George is better, sure, but certainly nothing like the improvement of Harden or Curry, both of whom are as good as George.

Thanks for posting those. Can't say I think Harden has improved all that much, but certainly Curry has raised his offensive output from 'good' to 'near-league-leading'. I quite like Vasquez as MIP personally.

Mike G wrote:I have Wall #35, which is above Paul George. Does his 'prior' of -1 mean he was a below-avg player last year? Because I think he was well above avg.
He raised his TS% by a modest .019, while shooting a lot more and cutting down his TO. All well and good, but others improved more -- per minute and times minutes.
Per minute, among players with at least 1500 minutes, the clear leader is Andray Blatche; from .68 to 1.59 eWins/484, a gain of .92 . Wall, at +.36, ranks 19th in improvement.

I think that Blatche didnt actually improve nearly as much as those numbers otherwise indicate. I agree that he's a better player than he was in 2010 and 2011, but his 2012 performance numbers were sss/injury biased. Calling all of the diff from his 2012->2013 performance improvement would be very misleading imo. Substantially more of that difference was just mean reversion and not actual improvement.

re: Wall... Yeah, we had his prior at -1.5 on defense and 0.5 on offense. While his offense improved last year, his defense was still awful. That's just not true anymore (visually as well as in the rapm data). We have him at a 0 on defense now. No one is talking about defense enough here imo when discussing Wall (or the other most improved players).

And offensively, he took a *huge* leap up this season. His TS%, USG%, AST%, and TOV% were nearly identical for both of his first 2 years in the league. This year his TS% and AST% went clearly up, his TOV% went clearly down, and all while his USG% jumped to > 29%. That kind of improvement (in those 3 rate stats) occurring along with a jump in usage to something that high (8th in the nba, just ahead of Harden) is pretty powerful. Wall's ASPMo was +3.5 this year (way up from +0.9 the previous season), for the reasons I just mentioned.

Mike G wrote: If he'd played 82 games like he played his last 21 -- including his 36 mpg -- he well may have been #1.

I guess part of this is semantics, but to me how good he was in those 21 games is precisely why I think he was the most improved player in the league this year.