APBRmetrics

[94by50]

From Part 1 of his MVP rankings column today:

[Mike G]

[bchaikin]

despite a low rate of turnovers per touch of just 5% (1 turnover for every 20 touches on offense), simulation shows that among the 40 or so PFs that played at least 1500 total minutes in 07-08, tim thomas is quite possible the absolute worst of the 40 in terms of wins generated on a per minute basis, unless you consider jeff green of seattle a PF...

[Mike G]

This may mean he's the 40th-best PF. But then again, hard to imagine he's playing PF. With Brand out, I guess someone had to try.

Still, in answer to the question: "Is there an NBA forward alive who couldn't play 31 minutes a game, score 12 points, notch five rebounds and three assists,..?
... I submit that there are no more than perhaps a dozen who reach these modest goals in each of the 3 categories, as often as not. Contrary to intuition, not many players (at any position) are as good as average at all 3 tasks.

His 'win generating rate' is probably highly dependent on how you rate him defensively. His production is certainly better than others with similar minutes, who are known to play good defense.
_________________
`
36% of all statistics are wrong

[bchaikin]

hard to imagine he's playing PF...

i'd guess their minutes distribution went something like this:

PG - brevin knight, sam cassell, dan dickau, smush parker
SG - cuttino mobley, quinton ross
SF - corey maggette, quinton ross, al thornton
PF - tim thomas (6-10, 240), al thornton (6-8, 220), ruben patterson, elton brand, nick fazekas
C - chris kaman, josh powell, aaron williams, paul davis

which would put nearly all of thomas's minutes at PF...

His 'win generating rate' is probably highly dependent on how you rate him defensively.

i've seen thomas in over a dozen games this year, and i'd say his defense outside of steals, blocks, and defensive rebounding was poor, based in good part on a seeming lack of effort/interest...

His production...

production and win generating can be two different things, depending on how you define production...

is certainly better than others with similar minutes, who are known to play good defense.

PFs that played similar or more minutes? such as whom?...

looking at all PFs (almost 50) who played at least 1400 minutes (thomas played 1940 min) he was a very poor offensive rebounder and a very poor overall rebounder. his overall shooting or ScFG% was in the bottom 1/5 of all PFs playing at least 1400 minutes, he was one of the very worst PFs for drawing fouls, in the bottom 1/4 for steals, bottom 1/3 for blocks, and one of the very worst in (reb+st+bs)/40min. despite a low rate of turnovers per touch, he's still in the bottom 1/5 of these almost 50-some PFs in points scored per zero point team possession personally responsible for...

so he was not efficient on offense, was a worse offensive rebounder than many SGs and SFs, a worse defensive rebounder than the league average SF, and imho a poor overall defender. not surprising he generated very few wins, and did so at a per minute rate that is among the very lowest of all PFs playing at least 1400/1500 minutes. simulation shows PFs like kevin garnett, tim duncan, shawn marion, and chris bosh generating on a 40 min/g and 82 game basis about 14-17 more wins than does tim thomas...

[Mike G]

OK, Bob; I understand the guy's shortcomings. He's sized like a PF, but he doesn't fit that role. Not every 6-10 player can fill in for Elton Brand.

The big separator between Thomas and others listed as PF would be his assists. Of 39 players listed (at Dougstats.com) as PF, who have played 1400 minutes, he's #6 in assists/min. Now, 2.8 Ast/36 is hardly stellar; but just one PF -- Boris Diaw -- has a rate as much as 50% better; while the median of the 39 is barely half that of TT.

I just find it funny that a broad statement such as the one quoted above can go (almost) unquestioned. Only 4 players in the whole NBA are listed as PF, with better per-minute Pts, Reb, and Ast -- KG, Dirk, Odom, Josh Smith -- in any minutes played.
_________________
`
36% of all statistics are wrong

[bchaikin]

Not every 6-10 player can fill in for Elton Brand.

few 6-10 players (or any PFs for that fact) can fill the shoes of an elton brand...

The big separator between Thomas and others listed as PF would be his assists. Of 39 players listed (at Dougstats.com) as PF, who have played 1400 minutes, he's #6 in assists/min. Now, 2.8 Ast/36 is hardly stellar; but just one PF -- Boris Diaw -- has a rate as much as 50% better; while the median of the 39 is barely half that of TT.

how do those assists relate to winning games?...

I just find it funny that a broad statement such as the one quoted above can go (almost) unquestioned.

who praytell is saying such a statement can't be questioned?...

Only 4 players in the whole NBA are listed as PF, with better per-minute Pts, Reb, and Ast -- KG, Dirk, Odom, Josh Smith -- in any minutes played.

at SF luke walton, ricky davis, and wally szczerbiak got more points and assists on a per minute basis than shane battier. walton, desmond mason, and bobby simmons all got more rebounds per minute than battier. none generates anywhere near the wins battier has this season on a per minute basis - in particular because of battier's overall excellent defense and better than average offensive efficiency, two key areas tim thomas has been deficient in this year...

[Neil Paine]

WARP says Clippers win 1.5 fewer games if Thomas' minutes are given to one of the best NBDL PFs. Using Dougstats positions, TT was the 42nd-most valuable PF in the NBA. Out of the 61 PFs who played 500+ min, Thomas ranked: 29th in Pts/36, 23rd in usage rate, 45th in TS%, 6th in assist ratio, 25th in turnover ratio, 50th in FT/FGA, 61st (dead last) in offensive rebounding %, 49th in defensive rebounding %, 47th in Blk%, and 43rd in Stl%. 82games shows that the Clips' D was 3 pts/100 poss better when he was on the floor, but that he was a poor man defender at PF.

In short, Thomas is not a replacement-level player, and he's not particularly close to being one -- although that doesn't mean he's a league-average player, either. In 2007-08, he was somewhere in between the two, which isn't great for someone with his talent level, I might add. Thomas has a reputation as an underachiever who doesn't always try his hardest (witness how well he plays when motivated for a new contract, or when his team is in the playoffs), and that doesn't exactly earn him respect from, well, anybody. So it's understandable that Simmons sees him as infinitely replaceable. On a losing team like L.A., they certainly "could have finished last without him", as Branch Rickey might say.

[Mike G]

I think it's actually pretty typical that the worst teams have such guys getting major minutes. In Thomas' case, he's filling Brand's minutes; so obviously, he wasn't expected to go almost entirely at PF, if at all.

This may also be an instance of a player whose decline has begun. This is only the 2nd year he's topped 30 MPG. It's also his lowest Pts/min and 3P% ever.

His per-minute stats are very close to the team averages, actually a bit better: