APBRmetrics

The discussion of the analysis of basketball through objective evidence, especially basketball statistics.
It is currently Fri Nov 24, 2017 4:31 pm

All times are UTC




Post new topic Reply to topic  [ 60 posts ]  Go to page Previous  1, 2, 3, 4
Author Message
PostPosted: Wed Feb 15, 2012 1:33 am 
Offline

Joined: Thu Apr 14, 2011 11:10 pm
Posts: 4748
Jerry, would you consider doing S3PM and / or RAPM for triplets or quads for this season? The bigger sample sizes might be desirable, though there is a trade off. Still with overall RAPM, S4PM and one or more of these, one have multiple snapshots and maybe some depth of vision.


Top
 Profile  
 
PostPosted: Wed Feb 15, 2012 3:34 pm 
Offline

Joined: Fri Apr 15, 2011 8:28 am
Posts: 807
Crow wrote:
Jerry, would you consider doing S3PM and / or RAPM for triplets or quads for this season? The bigger sample sizes might be desirable, though there is a trade off. Still with overall RAPM, S4PM and one or more of these, one have multiple snapshots and maybe some depth of vision.

RAPM for quads and triplets is definitely not something I'll do in the near future considering that it performs worse in out of sample testing. That's not really something I want to invest time in.

S3PM sounds like it might be interesting, but I can't promise you that I'll do it. It depends on how much time I have and how many other projects might get in the way

_________________
http://stats-for-the-nba.appspot.com/


Top
 Profile  
 
PostPosted: Thu Feb 16, 2012 1:18 am 
Offline

Joined: Thu Apr 14, 2011 11:10 pm
Posts: 4748
Ok, thanks for the reply & consideration of S3PM.


Top
 Profile  
 
PostPosted: Thu Feb 16, 2012 7:15 am 
Offline

Joined: Thu Apr 14, 2011 11:35 pm
Posts: 153
J.E. wrote:
mystic wrote:
We can see that certain players are doing better than others, maybe it can show some chemistry issues for some lineups/players, but I'm hard pressed to see anything more in those numbers

It's certainly not meant for predictions. As already stated, I think it's a nice tool when explaining why player A is most likely better (in terms of influence on team points) than player B to someone who does not know regression or chose to ignore RAPM results. It's pretty straight forward and more intuitive. It's certainly useful when trying to make a point like Nowitzki being better than Marion(or Collison over Perkins) because it's a more straight forward way to compare players, without a complicated algorithm behind it


I think those are good summations of the uses and limitations of the tool. Because it is so dependent on the quality of the player's teammates, I think it's use is limited to comparing players who are on the same team. As J.E. says, I'm thinking it's a way to present results that are plus-minus-ey in nature, without going whole hog for APM or RAPM. But I need to think about it some more, it's an intriguing way of thinking about things.


Top
 Profile  
 
PostPosted: Sat Feb 18, 2012 1:54 pm 
Offline

Joined: Fri Apr 15, 2011 8:28 am
Posts: 807
Marion missed what felt like 10 easy layups and dunks tonight, coupled with a some misses when he was posting up guards (one of the better defensive guards in Holiday, but still). I'm certainly not surprised to see his offense in deep red in comparison with other Mavs; and I think they need to change something about their minute distribution if they want to compete. Maybe start
Kidd-West-Carter-Nowitzki-Haywood or
Kidd-Carter-Odom-Nowitzki-Haywood

_________________
http://stats-for-the-nba.appspot.com/


Top
 Profile  
 
PostPosted: Mon Mar 12, 2012 3:56 pm 
Offline

Joined: Sun Apr 17, 2011 3:33 pm
Posts: 98
Jerry, are you planning on updating the RAPM stats at some point on your website, or have we lost you to an NBA team. :)


Top
 Profile  
 
PostPosted: Tue Mar 13, 2012 3:33 pm 
Offline

Joined: Fri Apr 15, 2011 8:28 am
Posts: 807
I'm on vacation until the 26th. I can only update the 'uninformed' RAPM player ratings from here. Everything else gets updated as soon as I'm back.

_________________
http://stats-for-the-nba.appspot.com/


Top
 Profile  
 
PostPosted: Sat Mar 31, 2012 5:13 pm 
Offline

Joined: Fri Apr 15, 2011 8:28 am
Posts: 807
I made updates to S4PM and (un)informed RAPM. Chris Paul is #1 in both RAPM versions. Garnett seems, again, to be playing quite well despite his age. James Harden seems to be the driving force behind OKC's offense (but his defense leaves much to be desired). The Bobcat's rookies, Biyombo and Walker, are in last and 10th to last place. Stoudemire should probably never play again, this season, if NY wants to win

_________________
http://stats-for-the-nba.appspot.com/


Top
 Profile  
 
PostPosted: Sat Mar 31, 2012 6:42 pm 
Offline

Joined: Sun Apr 17, 2011 3:33 pm
Posts: 98
J.E. wrote:
I made updates to S4PM and (un)informed RAPM. Chris Paul is #1 in both RAPM versions. Garnett seems, again, to be playing quite well despite his age. James Harden seems to be the driving force behind OKC's offense (but his defense leaves much to be desired). The Bobcat's rookies, Biyombo and Walker, are in last and 10th to last place. Stoudemire should probably never play again, this season, if NY wants to win


Cool. Thanks for the update.


Top
 Profile  
 
PostPosted: Sat Mar 31, 2012 7:31 pm 
Offline

Joined: Mon Apr 18, 2011 10:09 am
Posts: 470
J.E. wrote:
I made updates to S4PM and (un)informed RAPM.


Any reason why you removed the S4PM for prior seasons?

J.E. wrote:
Garnett seems, again, to be playing quite well despite his age.


Quite amazing his performance level. Last night was a perfect example what the difference between someone like Garnett and someone like Kevin Love is.

J.E. wrote:
Stoudemire should probably never play again, this season, if NY wants to win


So true. I said that as they signed Tyson Chandler, that they should get rid of Stoudemire otherwise that signing is complete nonsense.

Btw, McGee+Nick Young for Nene Hilario, what the Wizards did was really great. Getting rid of two knuckleheads without any positive impact in the game while adding a underrated big. Since aquiring Nene the Wizards played similar to a +1.3 team (SRS), while they played like -8.4 before. No idea what the Clippers and the Nuggets were really thinking when they traded for Young and McGee.

_________________
http://bbmetrics.wordpress.com/


Top
 Profile  
 
PostPosted: Sat Mar 31, 2012 11:20 pm 
Offline

Joined: Thu Apr 14, 2011 10:41 pm
Posts: 911
Location: Hotlanta
My guess is Denver is simply preparing to have cap space going into the off-season. I don't see them re-signing McGee. They're not that dumb. I thought Nene has several good years left, but maybe they know something we don't?

_________________
The City: A Golden State Warriors-Centric NBA Blog
NBA WOWY: Find statistics of your favorite team with any arbitrary combination of players on or off the court.


Top
 Profile  
 
PostPosted: Sat Mar 31, 2012 11:53 pm 
Offline

Joined: Mon Apr 18, 2011 10:09 am
Posts: 470
EvanZ wrote:
My guess is Denver is simply preparing to have cap space going into the off-season.


They have $51.3m tied up next season for Afflalo, Harrington, Chandler, Andersen, Gallinari, Brewer, Koufos, Lawson, Faried, Hamilton plus $1.4m guaranteed for Mozgov and $1.2m as caphold for the #18 pick. These are 11 roster spots, we have to add $0.5m for one more caphold. That makes overall already $51.8m or about $6.2m in capspace (assuming the salary cap will not change, which is not unlikely). And to use that capspace they have to renounce the full MLE, the BAE, the rights to all players like McGee or Fernandez. If they want to keep Mozgov, we can reduce the capspace to $5m. Well, I don't see how they can sign any player close to Nene for that amount of money.

Since the trade the Nuggets are playing like a -4.6 SRS team, they played like a +3.0 SRS team before. That is a crazy difference.

EvanZ wrote:
I don't see them re-signing McGee. They're not that dumb. I thought Nene has several good years left, but maybe they know something we don't?


Well, I would like to know their reasoning behind that. There were rumors about them trying to trade Nene basically right after they signed him. So, it is an interesting development, especially when we keep in mind how great they handled the Anthony trade.

_________________
http://bbmetrics.wordpress.com/


Top
 Profile  
 
PostPosted: Sat Apr 07, 2012 3:28 am 
Offline

Joined: Thu Apr 14, 2011 11:10 pm
Posts: 4748
Wonder what % of teams who win 60% of games have a +3 RAPM player or multiple players. Easy enough to compute such comparisons. Looks like all of the teams over 60% have at least one +3 guy (on prior-informed RAPM) except the Lakers. The 7 teams over 60% have about half of all the +3 guys.


Top
 Profile  
 
PostPosted: Sun Jul 14, 2013 11:26 am 
Offline

Joined: Fri Apr 15, 2011 8:28 am
Posts: 807
What's up everyone

Just wanted to say that "Same 4 Plus Minus" numbers are up again at http://stats-for-the-nba.appspot.com/, from '07-'08 until '12-'13.

You'd have to go one of the rating-pages like http://stats-for-the-nba.appspot.com/ratings/2010.html. Clicking on a player name brings you to that player's S4PM for that specific season. I know it's not easy to navigate (if you want all the S4PM pages for one player you have to visit all the rating pages). I apologize

For '13 LeBron James's S4PM looks quite pretty http://stats-for-the-nba.appspot.com/s4 ... _2013.html (no comparison worse than +10 NET).
Amir Johnson's page is all green, too http://stats-for-the-nba.appspot.com/s4 ... _2013.html and
Steph Curry has no comparison on offense where he's worse than +7 http://stats-for-the-nba.appspot.com/s4 ... _2013.html

Enjoy

_________________
http://stats-for-the-nba.appspot.com/


Top
 Profile  
 
PostPosted: Wed Apr 16, 2014 10:09 pm 
Offline

Joined: Fri Apr 15, 2011 8:28 am
Posts: 807
I updated all the S4PM player pages for this season seeing how the season is almost over and things aren't going to change much with just one game remaining. Remember that it's not supposed to replace any player metric, instead it's more of a head-to-head comparison tool for players that have played with similar lineups, and another way of viewing the +/- lineup data
Also note that PlayerA can be compared to PlayerB if both played with CDEF, but not if A played with GDEF and B played with CDEF, even if we're certain C and G are equally good

Sample size matters here. It's probably more impressive if Player A is +7 over Player B in 2000 possessions, than being +15 in 600 possessions - because of the noise associated with +/- data. The cutoff is 400 possessions which equals a little more than two full games

Strength of opponent is accounted for

Things that may be worth highlighting:
- Ray Allen compares badly to the Heat's younger guards, Chalmers and Cole (healthy sample size on both), mostly because of his defense
- Mario Chalmers looks pretty strong across the board. He may be a little underappreciated right now. There's one player I'll keep an eye on in the playoffs
- Michael Beasley is all negative. Only non-negative comparison on defense is LeBron... (779 poss.)
- Dwyane Wade looks rather weak for a star. I'd guess due to health
- LeBron only better than Ray Allen on defense. Better than everyone on offense

- Kenyon Martin is +29 over Amaré Stoudemire in 860 comparable possessions. That may be a new record. He looks pretty good on defense generally
- Bargnani's best(!) comparison is Stoudemire at -8. Worst is with Shumpert at -33 in 615 poss.
- Iman Shumpert's only negative comparison is Carmelo. Looks strong otherwise with lots of +20s, although some of it is against weak competition (Stoudemire, Bargnani, Hardaway)
- J.R. Smith mostly positive, only worse than Shumpert and Anthony
- Carmelo Anthony almost all positive (except for Artest). +30 over Stoudemire, +28 over Chandler (both small sample size)

- Being on the Bucks makes it easier to have good comparisons. Nevertheless, Khris Middleton may be worth mentioning as his S4PM is better than everyone's but Nate Wolters' and Giannis' (not so great sample size with many of the comparisons, but the fact that almost all of them are positive still counts for something)
- Nate Wolters has only positive S4PMs, thanks to what appears to be solid defense
- Luke Ridnour negative across the board. Seeing how he isn't exactly compared to stars he might/should probably be out of the league soon

- Tyreke Evans is all NET negative. He has one comparison on offense that's >+1 and one on defense.. (12 players in the sample)

- Nick Collison is another player with almost all positive S4PM. +20 over Perkins, +11 over Ibaka, +17 over Adams and even +11 over Kevin Durant (smaller sample size though)
- Reggie Jackson is all positive except for the comparison with Durant. Slightly better than Westbrook (+2) in a healthy 2528 possessions
- Kevin Durant at least +8 against everyone on offense (except for Collison, where he's -5)

- In this metric, Iguodala absolutely destroys (at least +8) everyone on GSW except for Draymond Green. +20 over Barnes in 2650 possessions.

- Matthew Dellavedova +11 over Kyrie Irving in 1300 possessions
- Waiters +7 over Irving in 1300 possessions, due to defense

- Kevin Martin is all negative.. even on offense

- Mike Miller with defensive comparisons of: -6, -16, -14, -13, -21

- Trevor Booker on defense: -22, -14, -11, -11, -10, -14

- Jae Crowder on defense: +1, +11, +9, +7, +6, +21, +14

- Kyle O'Quinn is all positive (thanks to his offense) in 5 comparisons with decent sample size

- Nick Young is all positive on offense. Defense is mixed

- Beverley +7 over Lin in 3000 possessions
- Asik only -1 with Howard in 2000 possessions
- James Harden all positive on offense. Most in the +10 range

- Hibbert -5 in comparison with Mahinmi in 1843 possessions. -6 of it from defense. That should put a dent in people's 'Hibbert for DPOY'-argument
- Evan Turner -11 in comparison with Danny Granger (small sample size)

- Kemba Walker all >+9

- Kenneth Faried only better than Wilson Chandler on defense. -11 compared to Mozgov, -7 with Hickson, -5 with Darrell Arthur

- Chris Paul all positive. +7 over Collison in 2332 possessions. +8 over Crawford, +7 over Redick

- Tiago Splitter all positive on defense, all negative on offense
- Patty Mills +10 over Tony Parker is surprising (1500 poss.)

- LaMarcus Aldridge all positve, several (low sample size ones) in the +20 range

- Rajon Rondo all negative, very much so on defense. Not good on a team as bad as Boston was this season

_________________
http://stats-for-the-nba.appspot.com/


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 60 posts ]  Go to page Previous  1, 2, 3, 4

All times are UTC


Who is online

Users browsing this forum: Bing [Bot], Google [Bot] and 8 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group