Who is the most valuable NBA player?

Mike G · Post by **Mike G** » Fri Jan 20, 2012 7:22 pm

It's ridiculously early, of course. But it seems you may vote now and change your vote later.
If anyone wants to vote for someone not in this list of 10 (max allowed), nominate them here. Apparently the options can be edited: I can replace any name that doesn't have any votes.

Last year, as I recall, it was very early when someone announced that Rose had been decided by the media to be mvp. And it proved to be true. So maybe we can get the jump on the early declarations.

J.E. · Post by **J.E.** » Fri Jan 20, 2012 7:35 pm

Should we vote depending on who we "think is going to win" or on "who we think is the best/most influential player". There is a world of difference between the two, for me at least.

EvanZ · Post by **EvanZ** » Fri Jan 20, 2012 7:44 pm

I voted for LeBron, but apparently it's Durant's turn. You can pretty much bank on it, if OKC ends up with the #1 seed in the West.

What if Philly is the best team? If the logic is "The best player on the best team.." (which I've heard many times), does Iguodala or Thad win it? Obviously not. But that would put the lie to that test.

Mike G · Post by **Mike G** » Fri Jan 20, 2012 7:52 pm

How about: literally the player with most value on the court. Normally, this player would also be most valuable on any team.
Some people may value 'sexy', a good story, or some other marketability. But I think we're all after sizing up value on the court, yes?

Evan, was it you who 'announced' here that Rose was going to win last year?
This is the first I've heard that "it's Durant's turn"; but we don't have to vote for him.

Also, see if you can change your vote. I can't seem to.

EDIT: Yes, we can.

EvanZ · Post by **EvanZ** » Fri Jan 20, 2012 7:55 pm

Yeah, it does let me change my vote.

Here's last year's thread:

http://www.apbr.org/metrics/viewtopic.php?f=2&t=6

Here's what I said in the third comment (but it wasn't until Mar. 24):

My vote is for Howard. I think Rose is going to win the award for recognition of his high USG rate and lack of any All-Star teammates this season. I think that's a decent argument.

But this just goes to show how little voters/media care about defense.

Here is my "MVP" page using ezPM WARP total (not that I would use this in isolation, of course):
http://thecity2.com/ezpm-player-ratings/

xkonk · Post by **xkonk** » Fri Jan 20, 2012 8:01 pm

I voted for LeBron. I don't see anyone obviously better than him around the league.

Here are the stats for 6 contenders (the max you can put in at once). http://bkref.com/tiny/74k2G

LeBron is far and away the best in WS48 and presumably will be in WS overall when he plays more minutes; he's already first anyway, just not by a lot.

Mike G · Post by **Mike G** » Fri Jan 20, 2012 8:05 pm

EvanZ wrote: Here's last year's thread:

http://www.apbr.org/metrics/viewtopic.php?f=2&t=6

Thanks. It was greyberger who wrote

Howard should win it. Rose has got it in the bag though.

And this was right after I put up the poll, so it was shocking to see it.

EvanZ · Post by **EvanZ** » Fri Jan 20, 2012 8:12 pm

He wrote that right after my comment. He was more definitive, I guess. By that point in the season, wasn't it fairly obvious that Rose had the MVP momentum behind him? Anyway, the reason I brought up Durant is because I read this:

http://www.nba.com/mvp-ladder/2011-12/i ... ef:nbahpt1 (not sure how Wade ends up on this list...placeholder in case he makes a huge comeback?)

And that follows up the GM poll who favored him, as well.

DSMok1 · Post by **DSMok1** » Fri Jan 20, 2012 9:13 pm

Don't forget CP3. Depending on how well the Clippers play, he's got a good narrative.

J.E. · Post by **J.E.** » Sat Jan 21, 2012 12:14 am

Here's my view. I'm sure it's controversial. Last year I was lobbying for Nowitzki. Seeing how the playoffs and finals turned out I'd say I got that going for me.
My opinion is that a player who is in discussion for MVP should have positive On/Off going for him almost always*. If he has no good On/Off we can deduce the team isn't playing (much) worse without him. If the team is not playing much worse without him, how can such a player be called the "most *valuable* player"?

*I'd say it's OK not to have a stellar On/Off if you're a starter on a team with a very good bench unit. This would require some bench players to be better than their starter counterparts. I'd say Derrick Rose might be an example for this scenario, with Gibson/Asik/Brewer probably being a better 3 man unit than Bogans(Hamilton)/Boozer/Noah

Non prior informed RAPM says the best players so far are Vince Carter, Thaddeus Young, Taj Gibson, Al Harrington, Mike Conley, Andre Miller and Vlad Radmanovic, in that order.
None of these is a starter except for Mike Conley.

The Mavericks are playing +20 basketball with Carter, -5 without, but he's ~10th in minutes. His impact per possession may be high, but his total impact is limited by playing time. I'd say if you want to at least be in discussion for MVP you need to play mostly against starters and 30MPG+.
The 76ers are playing +20 basketball with Young, +7 without. He's 3rd in total minutes for the 76ers but still below 30MPG, and, obviously, not a starter either.
Chicago is playing at +19 with Gibson, +4 without. He's ~8th in minutes and, you guessed it, not a starter (are you sensing a trend yet?)
Denver is playing at ~+14 with Miller and Harrington, ~-2 without. RAPM can't tell which one is better. Having a teammate with similar impact is probably not a good thing if you want to get MVP votes. Neither player is a starter, but at least they're 4th and 5th in minutes.
The Hawks are playing at +17 with Radmanovic, +2 without. You know the story, not a starter blabla

Last but not least, Memphis is playing +13 basketball with Conley, -17 without (per 82games). He's a starter and playing >30MPG. Hurray! When he's on the court:
-Memphis' eFG% is 8% higher
-FT made and attempted go up for Memphis, down for the opponent
-Memphis TO's go way down, opponent TO's go up

So RAPM with no data/priors from last year would vote Conley

Best players according to prior informed RAPM are Nowitzki, Collison, Ginobili, LeBron, D12, CP3, Millsap, Conley, ThadYoung, Deng, Lowry, Nash, Jeff Foster and Gerald Wallace, in that order.
We can throw out Ginobili due to injury and Jeff Foster due to playing time.

Dallas is playing +5 with Nowitzki, +1 without. That's good but not great and not even close to what he did last year.
Miami is playing +9 with LeBron and +6 without. Also good but also not great.
Collison is not a starter, but for good measure: OKC is playing +12 with him, 0 without
Orlando is playing at +10 with Howard, -9 without. He's leading Orlando in minutes. So there's clear cut MVP candidate #1
LAC is playing +5 with Paul, -10 without, he's 4th in minutes. The entire Clippers starting unit looks good in +/-, but given Paul's performance in prior years we can assume that he has the largest positive impact. there's MVP candidate #2.
Utah is playing +6 with Millsap, -7 without. He's a starter and 2nd in minutes. #3
Deng has a negative On/Off for the same reason that Rose's isn't stellar either: Chicago's strong bench unit. Might want to think about giving Gibson more minutes
Houston is +8 with Lowry, -16 without. He starts and is 3rd in minutes. There's clear cut MVP candidate #4
Phoenix is playing -1 with Nash, -4 without. Good but not great
Portland is playing +8 with Wallace, -10 without. He starts and is 3rd in minutes. Clear cut MVP candidate #5

Conley and ThadYoung appear on both lists, we throw out Young because of not starting. That leaves us with Millsap, CP3, Conley, GWallace, Lowry, D12. These six are very good choices according to RAPM/plus-minus/On-Off. Howard's team is the best of the six, so right now I would vote for him

A word on given poll choices:
LMA has almost the same On/Off/RAPM as GWallace. He might be deserving but again, having a teammate with similar impact is probably not a good thing if you want to get MVP votes. Portland will probably not win enough games for him to be considered.
Anthony is playing great for the Knicks, but the Knicks overall are slightly below average. Can't see anyone voting for him
Bryant has a good On/Off, but, like Portland, the Lakers might not win enough games
Durant, to me, is not even OKC's best player, and has not been for years. He's not a horrible choice though. If he wins, he'll have to thank Collison. I also expect many people voting for him.
LAC is not good enough, and CP3 is probably more deserving than Blake
Rose could be a decent choice, given that Chicago is probably going to win many games. He'll have to thank Deng (and his coach, and the bench unit)
Atlanta is playing +5 basketball with Josh Smith, +15(!) without. Vladimir Radmanovic is good at basketball. I'd say Josh Smith is a horrible MVP choice this year.
Minnesota is playing at +2 with Love, +3 without. I know he'll get votes and it makes me cringe. Obviously his On/Off is bad because he's playing on a elite team and has elite PFs/Cs that back him up..... . He'll have to thank Rubio and Anthony Tolliver. If anyone's really considering to vote for Love, then please go watch some actual games of Minnesota instead of looking at BoxScores. I'm sure if you've done that, there's no way you still think he's one of the best players in the league

Mike G · Post by **Mike G** » Sat Jan 21, 2012 3:13 am

I thought raw on/off numbers had been completely supplanted by APM, which had been further refined to RAPM. If you have to give disclaimers regarding strength of backup players, etc, you may not be saying anything about real impact.

When you discount a player because his team may not win enough games, are you guessing who will win The MVP, rather than who is Most Valuable?

The Wolves are on a pace to win about 35 games. Without Love, I give them 10 to 20. Who would score? Who would rebound?

EvanZ · Post by **EvanZ** » Sat Jan 21, 2012 4:40 am

Only 4 of us have voted and all for LeBron? Is anyone going to dissent?

Mike G · Post by **Mike G** » Sat Jan 21, 2012 5:03 am

I had Kobe a strong 2nd a few days ago. Now there's less separation from 2 to 5 (Kobe, Dwight, Love, Durant) than between LeBron and Kobe.
Comparing his rates to others we've seen or heard of -

Code: Select all

year   diff   per-36 rates     Sco   Reb   Ast    PF   Stl   TO   Blk
2012   .00   Lebron James     33.2   8.5   8.0   1.9   2.0   3.8   .9
1989   .21   Michael Jordan   33.5   8.0   7.2   2.9   2.7   3.4   .8
1987   .34   Larry Bird       28.4   8.9   6.6   2.4   1.7   3.1   .9
2006   .46   Dwyane Wade      29.5   6.0   6.9   2.7   1.8   3.4   .7
2003   .57   Kobe Bryant      28.9   6.7   5.5   2.5   2.0   3.3   .8

2003   .59   Tracy Mcgrady    32.5   6.3   5.4   2.0   1.6   2.5   .8
1964   .76   Oscar Robertson  26.5   6.5   9.4   3.1   2.7   4.1   .9
1999   .79   Grant Hill       24.6   8.0   6.8   2.3   1.6   3.7   .5
2008   .80   Manu Ginobili    30.4   6.1   5.0   2.7   1.7   3.1   .5
1992   .84   Clyde Drexler    25.8   6.7   6.4   3.1   1.9   3.2   .9

These are mvp years, best years of Hall of Famers, etc. Only Jordan was this impressive -- they're equal as scorers; Bird the only slightly better rebounder; Oscar the only bigger passer.

Mike G · Post by **Mike G** » Sat Jan 21, 2012 5:16 am

Kevin Durant

Code: Select all

diff   per-36 rates     yr     Sco   Reb   Ast    PF   Stl    TO   Blk
.00   Kevin Durant     2012   29.6   7.2   3.8   1.8   1.2   3.7   1.2
.21   Dwyane Wade      2011   29.0   7.1   5.2   2.6   1.5   3.1   1.1
.26  Dominique Wilkins 1987   28.4   6.2   3.1   1.9   1.5   2.7    .6
.30   Paul Pierce      2003   27.0   6.9   4.6   2.8   1.7   3.5    .8

.30   Lebron James     2008   30.5   7.9   5.8   2.0   1.6   3.0   1.0
.32   Tracy McGrady    2001   25.7   7.2   4.5   2.0   1.4   2.4   1.5
.35   George Gervin    1981   27.7   5.5   3.1   2.8   1.2   3.3    .7
.37   Chris Mullin     1990   25.6   5.9   3.6   1.9   1.6   3.1    .6

.37   Grant Hill       2000   26.4   7.0   5.1   2.6   1.4   3.2    .6
.39   Julius Erving    1980   28.0   7.3   4.4   2.7   2.2   3.7   1.8
.39   Rick Barry       1967   27.6   6.0   2.8   3.1   1.3   3.0    .9
.39   Kobe Bryant      2003   28.9   6.7   5.5   2.5   2.0   3.3    .8

Durant's assist rates have jumped about 40% this year.

J.E. · Post by **J.E.** » Sat Jan 21, 2012 11:54 am

Mike G wrote:I thought raw on/off numbers had been completely supplanted by APM, which had been further refined to RAPM. If you have to give disclaimers regarding strength of backup players, etc, you may not be saying anything about real impact.

When you discount a player because his team may not win enough games, are you guessing who will win The MVP, rather than who is Most Valuable?

The Wolves are on a pace to win about 35 games. Without Love, I give them 10 to 20. Who would score? Who would rebound?

Raw On/Off tells a similar same story as RAPM in the most extreme cases, regarding player ranking (but not rating, where RAPM is more cautious). I think it is useful here because the very definition of MVP should be (close to) 'a player that is very hard to replace', as in 'team performce drops off considerably without him'. Many of the finals MVPs of the last 10 years fullfilled this criterion (in the regular season and in the playoffs).
I would not use raw On/Off to judge most players but at the very top and bottom it is able give us valuable information. I also think it gives us good information on MVP candidates because they, pretty much by definition, 'should' be found at the top. If they're not at the top, then they're probably not a viable candidate, especially when they also play for a mediocre team and have mediocre backups.

As for Love, the Wolves don't have too much trouble scoring when he's not playing , eFG even improves when he's not playing (47% -> 51%). He obviously improves their rebounding, but they also have a better opponent eFG when he's not playing (46% -> 44%), opponents are getting to the line less and Minnesota creates a whole bunch of more turnovers (13/p48m -> 20/p48m).
His pick and roll defense is still pretty horrible for someone that might get MVP votes and he's also bad at contesting shots

APBRmetrics