Real Plus-Minus Stats Released [ESPN]

[ChokeFasncists]

Real Plus-Minus
http://espn.go.com/nba/statistics/rpm/_/sort/RPM

Starters:
Jeremy Lin -0.67
Kobe Bryant -1.34
Jordan Hill -1.39
Wesley Johnson -2.71
Carlos Boozer -2.78

Bench:
Ed Davis 0.56
Ronnie Price -0.10
Robert Sacre -0.65
Wayne Ellington -0.92
Nick Young -1.88


Wins Above Replacement: (WAR)
The estimated number of team wins attributable to each player, based on RPM

Ed Davis 0.83
Jeremy Lin 0.65
Ronnie Price 0.58
Kobe Bryant 0.48
Jordan Hill 0.39
Robert Sacre 0.27
Wayne Ellington 0.21
Nick Young 0.08
Carlos Boozer -0.14
Wesley Johnson -0.15

[Sofa King]

I expected this to be worse

[TyCobb]

Nice to have the two starters I wanted removed from the lineup (Johnson and Boozer) to be the worst statistically as well.

[Kilroy]

The WAR number for Kobe is ridiculous... Shows the inherent limitation in stats. It also shows that no matter what happens this team isn't winning too many games.
Ed Davis is the only Laker to crack the top 100... And according to this he's more productive than Dwight.

Which reminds me... I really need to apologize to floppymouse..

[Little General]

Can someone explain what these numbers say?

[Slava]

I'm surprised that 8 of our 11 players have a positive WAR and we still suck balls.

[crazyeights]

These threads feel like we're the faculty and Choke is presenting his year long post-graduate thesis on the Lakers.

[crazyeights]

Here's an interesting article by Kevin Pelton about how he calculates WARP:

http://sonicscentral.com/warp.html

What is WARP?
WARP stands for Wins Above Replacement Player. The term and concept are borrowed from sabermetrics and, specifically, Baseball Prospectus. Conceptually, the WARP system seeks to evaluate players in the context of a team made up of them and four completely average players. The performance of this team is then compared to that of a team made up of four average players and one replacement-level player. The method draws heavily on the work of Dean Oliver.
What are the benefits of this method?
For one, this rating system is very flexible. Players can be rated on a per-minute basis (using the theoretical "winning percentage" of the team with four average players), in terms of their offense and defense and in terms of their overall value (WARP itself). Using replacement level shows the value of players that can play heavy minutes and avoid injury while continuing to perform above replacement level. Using wins gives a measure of value that is easy to understand and constant over time. Lastly, by eschewing the traditional linear weights method so common in basketball analysis, I believe WARP does a better job of incorporating defensive value.

What are the limitations?
Like all rating systems based on box-score data, WARP cannot account for contributions that are not tracked in the box score, most notably on defense. It does no better than linear weights methods at evaluating players like Bruce Bowen. Also, it requires a number of assumptions - the value of assists, the trade-off between usage and efficiency, and replacement level.

Also interestingly enough, DRPM is missing from the OP. Too bad as we've been focused on our historically awful defense.


121. Robert Sacre +0.46
140. Ed Davis +0.19
155. Julius Randle +0.07*

179. Ronnie Price -0.16
210. Ryan Kelly -0.40
213. Wayne Ellington -0.42
296. Nick Young -1.05
313. Xavier Henry -1.12
330. Jeremy Lin -1.22
341. Jordan Hill -1.29
365. Wesley Johnson -1.54
369. Jordan Clarkson -1.58
383. Carlos Boozer -1.79
417. Kobe Bryant -3.04**

*Julius played one game
**Kobe was 417 out of 420 players listed.


Interesting observations:

Tony Parker was #400
Melo #414
Zach LaVine was #420 (dead last)

[Slava]

I won't be surprised Kobe is so low on DRPM, just like Melo & Parker, they just log so many minutes that all kinds of stink whether good or bad attaches to them on these metrics.

[Kilroy]

But that's the problem with stats like this... They're called one thing, but in reality they're something else.

To the uninitiated, these stats seem like some magical answer to all questions they've ever been asking... It's instant credibility and instant BBIQ...

But it's all an illusion. In order to really make any real decision about what the numbers tell you, you actually need to know MORE about the game rather than less. Which is why there are so few teams using these metrics effectively.

[tenten]

But that's the problem with stats like this... They're called one thing, but in reality they're something else.

On the surface, if you knew nothing about the game, you'd assume, based on these stats that we'd win more games if we benched Kobe.

But you have to dig deeper before you realize that's not what the stat is really saying. So to the uninitiated, these stats seem like some magical answer to all questions they've ever been asking... It's instant credibility and instant BBIQ...

But it's all an illusion. In order to really make any real decision about what the numbers tell you, you actually need to know MORE about the game rather than less. Which is why there are so few teams using these metrics effectively.

Actually, based on the WAR, it means that we will win more games by playing the top 5 guys the most:

Ed Davis 0.83
Jeremy Lin 0.65
Ronnie Price 0.58
Kobe Bryant 0.48
Jordan Hill 0.39

I don't think I've ever seen that line up in any game.

[Kilroy]

Your right... I swapped the DRPM numbers from Crazy's post and the WAR numbers from Chokes post and was reading the Pelton article while I posted and read it as -0.48.

That said, my greater point still stands. Lin and Price replace each other most of the time... Kobe's replacement only plays a little over 10min a game but it's either Young or Ellington and they both also have positive WAR.

So the stat says "Wins Above Replacement," but it's clear by the numbers, that's not really what the stat is telling you.

Do you really think playing that lineup the most during the game would result in more wins? And just to be clear, we're talking about less than 1 win on average.

It takes a lot more than just knowing the numbers to know what to do with them. Just like all stats.

[tenten]

Your right... I swapped the DRPM numbers from Crazy's post and the WAR numbers from Chokes post and was reading the Pelton article while I posted and read it as -0.48.

That said, my greater point still stands. Lin and Price replace each other most of the time... Kobe's replacement only plays less than 10min a game but it's either Young or Ellington and they both also have positive WAR.

So the stat says "Wins Above Replacement," but it's clear by the numbers, that's not really what the stat is telling you.

Do you really think playing that lineup the most during the game would result in more wins? And just to be clear, we're talking about less than 1 win on average.

It takes a lot more than just knowing the numbers to know what to do with them. Just like all stats.

I think so, that's probably the most balance defense/offense lineup we can put out. Anyways we will never know as BS will never play that lineup because of either he's old school or want to tank.

[dockingsched]

Your right... I swapped the DRPM numbers from Crazy's post and the WAR numbers from Chokes post and was reading the Pelton article while I posted and read it as -0.48.

That said, my greater point still stands. Lin and Price replace each other most of the time... Kobe's replacement only plays a little over 10min a game but it's either Young or Ellington and they both also have positive WAR.

So the stat says "Wins Above Replacement," but it's clear by the numbers, that's not really what the stat is telling you.

Do you really think playing that lineup the most during the game would result in more wins? And just to be clear, we're talking about less than 1 win on average.

It takes a lot more than just knowing the numbers to know what to do with them. Just like all stats.

WAR doesn't literally mean wins above the player that replaces you. replacement player refers to an abstract player with generic skills that a team could hypothetically acquire at minimum cost. The way you worded it made it seems like you weren't making that distinction.

[Kilroy]

No, I wasn't... Tthat was part of my greater point, that Advanced stats are not as simple as they seem.
I hoped that it would be clear that the superficial comparison between replacement players didn't make sense from the numbers...

[TyCobb]

Just wait till basketball comes out with 10 different variations of WAR like baseball if it hasn't already.

[leeprettyp]

I'll stick to the eye test

[Danny Darko]

Pardon me... but


I have so many problems wanting to quantize winners this way. The endless iterations lead to manipulation of the data that surrenders all meaning. Also, the number can be a chicken and the egg thing and a guy on a team with amazing leadership, chemistry, and drive vs the same guy on a lousy team can be vastly different. I think it's more useful to focus on A- how does a guy work within the team constructs. B- how does a guy matchup most nights with the guy across from him.

[ChokeFasncists]

These threads feel like we're the faculty and Choke is presenting his year long post-graduate thesis on the Lakers.

~lol~ Ya, I do find that it's a little inert here, professors tend to be stuffy sometimes? And try to inject a little liveliness.

Of course, this is something I like/tend to do and I must say I feel like I learn quite a lot by posting these threads.

Also interestingly enough, DRPM is missing from the OP.

Well, then I'd have to post the ORPM, four sets, too much work!

[Sedale Threatt]

The trick is not to treat any one number as gospel, but a piece of a much larger picture.

It always seems to me that people who complain the most about any particular measure are mainly upset it doesn't reflect very well on their favorite player. Like the whole Kobe-LeBron debate from a few years back -- that torch was passed in 08 or 09, and a bunch of these measures reflected that. But people held on to what they want to hold on to.

Or that the numbers would reflect poorly on Kobe this year. He's probably as bad as he's ever been on defense, and he's taking a isht ton of shots at 39 percent. That's a pretty miserable combination that would obviously impact any team's performance.

Personally, I like the on/off numbers at NBA.com the most, and PER is usually a decent guideline. (Any measure that had the Wilts, Kareems, Jordans and LeBrons of the world on top when the eye test reflected their dominance has to be doing something right.) Stuff like WARP and RAPM tend to be pretty obscure and I agree that it's never really clear what exactly those numbers are telling you.

It's just hard to distill a free-flowing sport like basketball to one single value. Pretty much anybody comes up with these formulas will be the first one to admit they aren't the end-all, be-all. But I think there's enough value among the advanced stuff that that dismissing it entirely is rather misguided.

Personally, I'm glad to have them as box score stats aren't nearly nuanced enough and relying strictly on the eye test just provides way too much room for people to stick with their biases. It's a step in the right direction, at least.