APBRmetrics

DSMok1 wrote:I know of Jeremias' "RAPM" for the 90s and love his work, but I would contend that is not RAPM at all (I think Jeremias would agree), since it is not based on PbP data and only uses end-of-quarter scores and total minutes played as inputs--nothing like true Adjusted Plus/Minus derivatives. Because of that, the only solid information we have on Jordan's defense is from scouting/video and from box scores/team results. Note that defensive rebounds are not important in estimating defense, which is well established.

I agree with you to a large extent in this argument, and I am just butting in here, but can you point me to the establishment of this? Everything I have done (and admittedly not released) says that DRBs are important, even at the individual level, for indicating defensive ability. Their linear value isn't huge, but when coefficients are standardised they can be seen to be rather significant.

Yes I would have to say that I disagree about rebounding.

DSMok1 wrote:
"The 1994 Bulls sure looked good without him." The 1994 Bulls were 3.5 points per possession worse than 1993 Bulls, and 9 points per possession worse than the 1996 Bulls.

Edit: Defensively you clearly are wrong there, that's what I was primarily alluding to.

Compared to LeBron leaving the Cavs the 94 Bulls looked great defensively.

I have noted elsewhere that I have intentionally left out the playoffs at this point, because the varying number of games played and SoS are difficult to account for properly. At some point, I will deal with that and add that data in.

It is the most important part of a players legacy if he gets to the post-season often enough.

"I could see an argument made for those three years being comparable, I still don't like your statistic. Wade is an inefficient scorer by a pretty wide degree." In 2009, Wade had 57.4% TS, 36.2% USG, and 40.3% AST. RAPM indicates that USG coupled with AST is more important than efficiency for helping your team score on offense, and his USG/AST combination in that season was unprecedented.

Yes I'm aware of that usage-efficiency ratio, it is still not impressive. I'm looking at the basketball-on-paper figures.

Still not nearly as impressive as the ratio of other players ranked below.

I would appreciate an explanation of what you feel is (mathematically) problematic with ASPM--I'm sure you've seen the derivation at http://godismyjudgeok.com/DStats/aspm-and-vorp/

When my friend was completing his PHD in Economics, he messed up a rent control problem he was working on. The Coefficients were all crazy and such. This does not suddenly mean that econometric analysis does not have value, but it implies his a priori analysis was off.

v-zero wrote:

DSMok1 wrote:I know of Jeremias' "RAPM" for the 90s and love his work, but I would contend that is not RAPM at all (I think Jeremias would agree), since it is not based on PbP data and only uses end-of-quarter scores and total minutes played as inputs--nothing like true Adjusted Plus/Minus derivatives. Because of that, the only solid information we have on Jordan's defense is from scouting/video and from box scores/team results. Note that defensive rebounds are not important in estimating defense, which is well established.

I agree with you to a large extent in this argument, and I am just butting in here, but can you point me to the establishment of this? Everything I have done (and admittedly not released) says that DRBs are important, even at the individual level, for indicating defensive ability. Their linear value isn't huge, but when coefficients are standardised they can be seen to be rather significant.

You are correct, I was incorrect in my statement there. It is a significant factor when standardized--ASPM (overall regression) coefficients are 0.17 for TRB%, 1.26 for STL%, and only 0.28 for BLK%.

MJ 1988: 10.7 DRB%, 3.9% STL, 2.4% BLK: Raw Value would be 7.4
LeBron 2013: 20.8 DRB%, 2.4% STL, 1.9% BLK: Raw Value would be 7.1

So it is at least close when you factor in LeBron's rebounding.

DSMok1 wrote:I know of Jeremias' "RAPM" for the 90s and love his work, but I would contend that is not RAPM at all (I think Jeremias would agree)

I would probably still call it RAPM because the same algorithm is applied to the data, but obviously the data source is not as good for the 90s (fake matchupfiles vs real matchupfiles). Thus, my version of 'RAPM for the 90s' is a lot less accurate than RAPM which uses real PBP. I certainly wouldn't lean on it heavily in arguments of who is better
As a side note, Jordan didn't look to hot on defense in the years where we have PBP, when he playing for the Wizards from '01 to '03, but he was 38 & 39 years old then; so that's not telling us much

I think if your team gets better defensively without you for an entire season, that says a lot. Conversely if your team falls apart, that's probably a good sign.

Anderson Varejao is terrible now looking at his on-court/off-court numbers on defense. Kirk Goldsberry was right about him, he should spend some more time defending instead of roaming for rebounds (assuming he has that capability).

http://www.sloansportsconference.com/?p=10179

http://www.youtube.com/watch?feature=pl ... HX0Ezb-OPE

huevonkiller wrote:I think if your team gets better defensively without you for an entire season, that says a lot. Conversely if your team falls apart, that's probably a good sign.

Anderson Varejao is garbage now looking at his on-court/off-court numbers on defense.

Cleveland has been better on D with Varejao on vs. off every year... not sure what you're referring to.

Past 4 seasons:
(Delta & sample size, where sample is minimum of off or on minutes)
'10: -5.9 (1791 min)
'11: -3.5 (994)
'12: -2.3 (785)
'13: -0.2 (901) (Source: http://www.basketball-reference.com/pla ... -off/2013/ )

He does seem to be declining steadily, that's for sure.

DSMok1 wrote:
Cleveland has been better on D with Varejao on vs. off every year... not sure what you're referring to.

Nope, 82games parses that differently.

Past 4 seasons:
(Delta & sample size, where sample is minimum of off or on minutes)
'10: -5.9 (1791 min)
'11: -3.5 (994)
'12: -2.3 (785)
'13: -0.2 (901) (Source: http://www.basketball-reference.com/pla ... -off/2013/ )

He does seem to be declining steadily, that's for sure.

Varejao:

'13: +0.4 (Post Decision)
'12: -2.6 (Post-Decision)
'11: -3.6 (Post-Decision)
'10: -5.4 (With LeBron)

http://www.82games.com/1213/12CLE20.HTM#onoff

That's interesting. The numbers are significantly different from Basketball Reference.