Why I'm not a WP fan

[UGA Hayes]

^ Thats fair enough and I agree with your supposition. What I would say is that you could be right about Lee and I could be right about Lee. I just don't think its so black and white that a guy is going to have the exact same productivity from year to year, and that Lee's having a down year isn't necessarily a concrete argument against the system. Berri has noted that guys will occasionally have a wildly variable year, but as a rule most guys are pretty consistent under his sytem. Now if Lee has another year that is stastically similar to the one he had this year, then I think that would be a more rock solid case that the Lee/Love types are inflating their numbers on bad teams. I mean if you look at Lee's log he has too many holy **** games to just totally dismiss his impact and there to me at least isn't a strong reason why his numbers were different since he was playing on a stylisticall similar team to the one he was previously on. LEe was shooting in the 55% range before this year but only shot 50% this year. Given the number of shot attempts he gets thats a huge difference and iin my mind can more than account for 5-6 game swing in the number of games won not including the ones he missed. So the question becomes did you believe he just had a bad shooting year or do you believe teams adjusted to him say b/c they took the Warriors more seriously or something along those lines?

[floppymoose]

So I'm saying this part without looking at the numbers: I think Lee shot fine. He had a spell when his elbow was hurt where he couldn't shoot, and that may have pulled down his efficiency. But other than that I thought his shooting was exactly like he had been in NY. He just got fewer shots on a GS team that was loaded with offensive talent.

It's actually his rebounding that felt like it was down to me, not his shot. I chalk that up to playing some PF alongside a good rebounder (Biedrins) instead of being a full time center like he was his last two seasons in NY, and maybe also due to facing better teams in the western conference.

[mysticbb]

With a defensive adjustment, most of the box score metrics will correlate strongly with team wins. At the team level, of course. That doesn't mean they're doing a good job of evaluating INDIVIDUAL contributions, however.

We actually showed that already in this thread. viewtopic.php?f=344&p=26748754#p26749885

We created a "metric" called FM48 and the correlation coefficient to winning goes up to 0.97 from 0.68 while using a defensive adjustment.

[mysticbb]

Another good way to test WP is to use the WP48 values and see how well they correlate to actual performance of different team lineups. If that correlation is poor, that's a red flag. I believe mystic did tat earlier in this tread.

Indeed, I did that. http://bbmetrics.files.wordpress.com/20 ... ineups.gif

That's for the first half of this season. I've done that for the all Top50 lineups from 2007-08 on, and the result is similar. I also added PER to that. WP48 and PER are finishing nearly the same, roughly 0.1 points less in the correlation coefficient.

Especially with the way Berri is acting, like using WP48 for suggesting minutes distributions, the result is embarrassing. It shows that WP48 is the worst stats to suggest minute distributions.

[Idunkon1stdates]

So I'm saying this part without looking at the numbers: I think Lee shot fine. He had a spell when his elbow was hurt where he couldn't shoot, and that may have pulled down his efficiency. But other than that I thought his shooting was exactly like he had been in NY. He just got fewer shots on a GS team that was loaded with offensive talent.

It's actually his rebounding that felt like it was down to me, not his shot. I chalk that up to playing some PF alongside a good rebounder (Biedrins) instead of being a full time center like he was his last two seasons in NY, and maybe also due to facing better teams in the western conference.

His career TS% .590, but this season he was .545. His career FG% is .550, and he was just over .500 this season. He also rebounded quite well alongside Zach Randolph a few seasons ago, with Randolph and Lee both playing over 30mpg IIRC.

[mysticbb]

His career TS% .590, but this season he was .545. His career FG% is .550, and he was just over .500 this season. He also rebounded quite well alongside Zach Randolph a few seasons ago, with Randolph and Lee both playing over 30mpg IIRC.

His ORB%, DRB% and TRB% were also below his career numbers.

[Idunkon1stdates]

His career TS% .590, but this season he was .545. His career FG% is .550, and he was just over .500 this season. He also rebounded quite well alongside Zach Randolph a few seasons ago, with Randolph and Lee both playing over 30mpg IIRC.

His ORB%, DRB% and TRB% were also below his career numbers.

Yes, he was down across the board. His rebounding numbers were unaffected next to Randolph, though.

[mysticbb]

Yes, he was down across the board. His rebounding numbers were unaffected next to Randolph, though.

Pace adjusted per 36 numbers for Lee in 2007/08 and 2008/09:

in games with Randolph (ORB/DRB/TRB): 3.6/7.0/10.6
in games without Randolph: 3.2/8.6/11.8

Lee grabbed more offensive rebounds, but he clearly shows a trend down in terms of defensive rebounds.

[Idunkon1stdates]

yeah, 'unaffected' was clearly inaccurate. he was still a good rebounder alongside randolph, though. this was his worst season by far in terms of rebounding and scoring efficiency overall.