96-97 (NPI) RAPM is out!

[ceiling raiser]

http://ascreamingcomesacrossthecourt.bl ... m.html?m=1

Apologies if this has already been posted. Infinite thanks to the poster (valued contributor here) for his hard work!

It looks like he'll be putting the data out for the remaining three years (and perhaps the others as well, since they're all over the place). Tremendous stuff.

[E-Balla]

I guess people evaluated talent better in the 90's because those rankings seem super clean.

[colts18]

Worst defenders on the list:

dominique Wilkins
Brian Shaw
Antoine Walker
AI
Van Exel

Looks like the RAPM of 97 was definitely valid.

[semi-sentient]

Not sure how anyone can take this stuff seriously.

Defensive +/-

Code: Select all

Ron Harper       3.21
Tim Hardaway     2.65
Michael Jordan   2.59
Clyde Drexler    2.27
John Stockton    2.13
Byron Scott      1.47
Jeff Hornacek    0.78
...
Scottie Pippen   0.20

How in the world is Pippen that low? He played a full 82 games and was still a great defensive player. Stuff like this makes me want to continue ignoring RAPM whenever it's brought up in player comparisons.

[colts18]

Not sure how anyone can take this stuff seriously.

Defensive +/-

Ron Harper - 3.21
Tim Hardaway - 2.65
Michael Jordan - 2.59
Clyde Drexler - 2.27
John Stockton - 2.13
Byron Scott - 1.47
Jeff Hornacek - 0.78
... tens of other players ...
Scottie Pippen - 0.20


How in the world is Pippen that low? He played a full 82 games and was still a great defensive player. Stuff like this makes me want to continue ignoring RAPM whenever it's brought up in player comparisons.

In 1998, the Bulls played significantly better defense in the games Pippen missed than the games he played.

[semi-sentient]

In 1998, the Bulls played significantly better defense in the games Pippen missed than the games he played.

Assuming that's true, it probably has more to do with the fact that missed half the season with a severe foot injury and had to work himself back into condition. We're talking about '97 through -- before the injury when he was still an all-league defender.

PIPPEN'S FOOT INJURY is worse than people know. He basically has not played in half a year, not since he hurt himself in the championships last June. Our strength coach, Al Vermeil, worked Scottie out today, and was stunned to find his vertical leap is only 17 inches. That's the same as Joe Kleine's. Scottie's used to be 30 inches or so. His leg strength is shot. So now we're looking at another month or more of rehab.

[Durins Baynes]

Man, it makes me so angry to realise now that Karl Malone stole Terry Mills MVP.

I will never understand how people take these numbers as literal assessments of players. At best they're one other indicator, which provides a broad "gist", and which is sometimes wrong. Since we can't know when it's wrong and when it is right, it's a horrible tool to justify similarly scored players being better or worse than each other based on their APM. Yet KG fans continue to trump his APM (not his NPI APM of course, because it's not as good, so somehow invalid) as the test of his being better than players with comparable APM's (like Duncan or Shaq). Silly. He was much worse than both those guys.

[Winglish]

Tremendous, indeed! I've long thought that Greg Ostertag was a far better player than Hakeem Olajuwon. Now there is incontrovertible proof!

[SideshowBob]

Awesome job Justin, can't wait to continue see you take this forward.

Not sure how anyone can take this stuff seriously.

You can't really, its just encouraging that we're going to have data like this available for this time. It's non-prior informed, and its unstable. Once we get a version that uses some sort of prior (as well as once we start getting into the next few years) we'll have something that aligns with the J.E. single year studies that everyone around here quotes frequently.

[Ryoga Hibiki]

Instead of delighting us with this low level sarcasm, you might want to try to understand if there's anything into Ostertag's game that went unnoticed that season, like maybe Terry Mills'. This stat is not a final ranking, only its detractors seem to believe it.

Sent from my Nokia 3210 using Tapatalk 2

[mysticbb]

It is great to see someone having taken the time to clear up the mess the NBA.com pbp provide for the most part. Seeing such numbers gives some hope to get the matchupfiles for 1997 to 2000 at one point in the future. Anyway, I took the liberty to merge those numbers with my SPM by calculating the appropiate coefficients via a regression on the team's game by game performances.
Here is the result for the Top50 players of that season:

Code: Select all

Name                  Tm    Rat
Michael Jordan        CHI   9.2
Karl Malone           UTA   8.6
Shaquille O’Neal      LAL   6.6
Grant Hill            DET   6.2
Tim Hardaway          MIA   6.2
Scottie Pippen        CHI   5.7
Patrick Ewing         NYK   5.5
Mookie Blaylock       ATL   5.4
Christian Laettner    ATL   5.3
John Stockton         UTA   5.2
Terry Mills           DET   5.2
Alonzo Mourning       MIA   5.1
Jeff Hornacek         UTA   5.0
Gary Payton           SEA   5.0
Hakeem Olajuwon       HOU   4.8
Charles Barkley       HOU   4.3
Arvydas Sabonis       POR   4.2
Toni Kukoc            CHI   4.2
Chris Webber          WSB   4.1
Clyde Drexler         HOU   4.0
Detlef Schrempf       SEA   3.9
Hersey Hawkins        SEA   3.6
Anfernee Hardaway     ORL   3.5
Vlade Divac           CHH   3.3
John Starks           NYK   3.0
Rasheed Wallace       POR   3.0
Kenny Anderson        POR   2.9
Ron Harper            CHI   2.8
Dikembe Mutombo       ATL   2.8
Tyrone Hill           CLE   2.8
Shawn Kemp            SEA   2.8
Greg Ostertag         UTA   2.7
Kevin Garnett         MIN   2.7
Reggie Miller         IND   2.6
Terrell Brandon       CLE   2.5
Bo Outlaw             LAC   2.5
Joe Dumars            DET   2.4
Byron Scott           LAL   2.3
Kendall Gill          NJN   2.3
Horace Grant          ORL   2.3
Mitch Richmond        SAC   2.3
Nate McMillan         SEA   2.3
Kevin Johnson         PHO   2.2
Gheorghe Muresan      WSB   2.2
Glen Rice             CHH   2.1
Darrell Armstrong     ORL   2.0
Chris Mills           CLE   1.9
Chris Gatling         TOT   1.8
Vin Baker             MIL   1.8
Danny Manning         PHO   1.7


To put those numbers into perspective: From 2001 to 2013 the highest value was achieved by LeBron James in 2010 with 9.2 as well. He has 9.1 in 2009 and 2013. Kevin Garnett finished with 8.5 in 2004. In turn, Jordan 1997 scored in that metric as well as peak-level James, Karl Malone anno 1997 scored as well as peak-level Kevin Garnett. BUT, the amount of players with +5 or +2 is unusually high, usually a season has about 5 to 6 players with +5 or more, and about 35 players with +2 or more.
On the other end of the scale I get Roy Rogers with -5.1 and Brian Shaw with -6.0 being worse than any other player with enough minutes played in the dataset from 2001 to 2013. Before, Trenton Hassell in 2010 finished with -5.0.
So, not quite sure what to make of it ...

[ceiling raiser]

It is great to see someone having taken the time to clear up the mess the NBA.com pbp provide for the most part. Seeing such numbers gives some hope to get the matchupfiles for 1997 to 2000 at one point in the future. Anyway, I took the liberty to merge those numbers with my SPM by calculating the appropiate coefficients via a regression on the team's game by game performances.
Here is the result for the Top50 players of that season:

Code: Select all

Name                  Tm    Rat
Michael Jordan        CHI   9.2
Karl Malone           UTA   8.6
Shaquille O’Neal      LAL   6.6
Grant Hill            DET   6.2
Tim Hardaway          MIA   6.2
Scottie Pippen        CHI   5.7
Patrick Ewing         NYK   5.5
Mookie Blaylock       ATL   5.4
Christian Laettner    ATL   5.3
John Stockton         UTA   5.2
Terry Mills           DET   5.2
Alonzo Mourning       MIA   5.1
Jeff Hornacek         UTA   5.0
Gary Payton           SEA   5.0
Hakeem Olajuwon       HOU   4.8
Charles Barkley       HOU   4.3
Arvydas Sabonis       POR   4.2
Toni Kukoc            CHI   4.2
Chris Webber          WSB   4.1
Clyde Drexler         HOU   4.0
Detlef Schrempf       SEA   3.9
Hersey Hawkins        SEA   3.6
Anfernee Hardaway     ORL   3.5
Vlade Divac           CHH   3.3
John Starks           NYK   3.0
Rasheed Wallace       POR   3.0
Kenny Anderson        POR   2.9
Ron Harper            CHI   2.8
Dikembe Mutombo       ATL   2.8
Tyrone Hill           CLE   2.8
Shawn Kemp            SEA   2.8
Greg Ostertag         UTA   2.7
Kevin Garnett         MIN   2.7
Reggie Miller         IND   2.6
Terrell Brandon       CLE   2.5
Bo Outlaw             LAC   2.5
Joe Dumars            DET   2.4
Byron Scott           LAL   2.3
Kendall Gill          NJN   2.3
Horace Grant          ORL   2.3
Mitch Richmond        SAC   2.3
Nate McMillan         SEA   2.3
Kevin Johnson         PHO   2.2
Gheorghe Muresan      WSB   2.2
Glen Rice             CHH   2.1
Darrell Armstrong     ORL   2.0
Chris Mills           CLE   1.9
Chris Gatling         TOT   1.8
Vin Baker             MIL   1.8
Danny Manning         PHO   1.7


To put those numbers into perspective: From 2001 to 2013 the highest value was achieved by LeBron James in 2010 with 9.2 as well. He has 9.1 in 2009 and 2013. Kevin Garnett finished with 8.5 in 2004. In turn, Jordan 1997 scored in that metric as well as peak-level James, Karl Malone anno 1997 scored as well as peak-level Kevin Garnett. BUT, the amount of players with +5 or +2 is unusually high, usually a season has about 5 to 6 players with +5 or more, and about 35 players with +2 or more.
On the other end of the scale I get Roy Rogers with -5.1 and Brian Shaw with -6.0 being worse than any other player with enough minutes played in the dataset from 2001 to 2013. Before, Trenton Hassell in 2010 finished with -5.0.
So, not quite sure what to make of it ...

Thanks. Just wondering, does the bolded process give you off/def splits? If so, I'd be very interested in those.

[mysticbb]

Thanks. :) Just wondering, does the bolded process give you off/def splits? If so, I'd be very interested in those.

It could, but given the nature of the boxscore itself and the fact that the SPM metric is entirely based on the boxscore, I wouldn't put anything into such a split. And to be blunt: For me personally it doesn't matter whether a player is doing his thing on offense or defense, the overall scoring margin matters to me. Thus, I stopped worrying about a better method to split up those values between offense and defense, even though the older data on the webblog might suggest otherwise.

[colts18]

On the other end of the scale I get Roy Rogers with -5.1 and Brian Shaw with -6.0 being worse than any other player with enough minutes played in the dataset from 2001 to 2013. Before, Trenton Hassell in 2010 finished with -5.0.
So, not quite sure what to make of it ...

If there are more guys at the higher end of the spectrum and more at the lower,I would assume the standard deviation for the 1997 RAPM would be higher.Possibly suggesting that there needs to be more regression to 0.

Do you have any data that shows combining your SPM and RAPM is better? Is it more predictive?

[mysticbb]

If there are more guys at the higher end of the spectrum and more at the lower,I would assume the standard deviation for the 1997 RAPM would be higher.Possibly suggesting that there needs to be more regression to 0.

I actually normalise the values based on standard deviation. Therefore, that isn't the reason. I just have fewer guys in the middle of the pack, but the range is about the same.

Do you have any data that shows combining your SPM and RAPM is better? Is it more predictive?

Indeed, that is the whole reason I'm doing it. The merged metric exceeds the individual ratings in terms of predictions, while explaining the results in the same fashion as SPM alone (RAPM alone is worse at explaining).

[sp6r=underrated]

Are you going to be posting the merged data on your website and are you considering adding the 2013 campaign?

[Warspite]

You guys just have clue how great Terry Mills was that yr with the short 3pt line. Dumars was shooting 55% from 3 at one point in the season.

[Doctor MJ]

I guess people evaluated talent better in the 90's because those rankings seem super clean.

Odd to me that you say this while all the other negative comments here are based on singling out high marks from minor players just like always happens with +/- data.

[sp6r=underrated]

which is a totally fair thing to point out.

A subset of plus/minus fanatics on this board really would like to banish the box score from analysis and rely solely on plus/minus. This is demonstrated by rampant hostility to xramp, the number of threads in which people solely cite RAMP as a justification for players rankings.

In that sense the subset of hardcore plus/minus fanactics remind me of advocates of real business cycle with their focus on structural matters and disregard for fiscal/monetary policy to smooth business cycle. Accordingly pointing out that guys like Terry Mills are gods in plus/minus stats is necessary to defend box score from the plus/minus fanatics.

[Doctor MJ]

Man, it makes me so angry to realise now that Karl Malone stole Terry Mills MVP.

I will never understand how people take these numbers as literal assessments of players. At best they're one other indicator, which provides a broad "gist", and which is sometimes wrong.

I'd say your main issue is in how you see people who use RAPM...

Since we can't know when it's wrong and when it is right,

and this has much to do with it. I don't have confidence in how to deal with the noise in all +/- scenarios, and when I don't, I don't give that data much weight, but there are times when the big picture becomes pretty clear.

it's a horrible tool to justify similarly scored players being better or worse than each other based on their APM. Yet KG fans continue to trump his APM

Because there's a lot of data on this. You can't do reasonable analysis by looking at one piece of data with a lot of noise in it and using it to throw out stuff with a lot more signal.

(not his NPI APM of course, because it's not as good, so somehow invalid) as the test of his being better than players with comparable APM's (like Duncan or Shaq). Silly. He was much worse than both those guys.

I like how you mention NPI just to try to discredit people based on how they don't use, and then proceed to assert something not at all supported by NPI. You would be a better analyst if you actually tried to figure out how to use data for your own understanding instead of trying to use the data simply to spite others.

ftr, in general the issue with the Garnett supporter stridency has never been an insistence that he should be seen as better than Duncan or Shaq, but rather that every narrative relating to Garnett being somehow a failure based on what happened in Minny is based on clearly disproven assumptions.