Home Unfiltered Articles Players Baseball Prospectus
Basketball Prospectus home
Click here to log in Click here for forgotten password Click here to subscribe

Click here for Important Basketball Prospectus Premium Information!

<< Previous Article
The Bynum Question (04/13)
Next Article >>
Handing Out The Hardwa... (04/15)

April 14, 2009
The Envelope, Please
Picking My MVP

by Kevin Pelton


This one should be easy. The NBA's best player happens to play on the NBA's best team. Dwyane Wade has carried his Miami Heat, playing a larger role in the offense than any other player in the NBA has. Chris Paul has quietly continued to put up historic numbers from the point-guard position. You'll never convince Kobe Bryant's fans that their beloved superstar is not the league's best player. With due respect to all three guys, however, LeBron James is the MVP, and there's no question about it.

That being the case, I could probably end this column right here. However, you're probably expecting a little more. Fortunately, a deeper look at the numbers only reinforces how impressive James' season has been.

Historic Numbers

James is, by a variety of measures, having the most productive season in modern NBA history. You've probably read that his PER this season is the best of all time, and while I've never bothered to estimate my ratings for players prior to the league tracking blocks, steals and turnovers, James does in fact have the best per-minute win percentage dating back to the 1979-80 season, .841. That means my system estimates that James and four average players would win 84.1 percent of their games if he played all 48 minutes every game.

That's not quite the same as having the most valuable season in modern NBA history, which is the issue the Wins Above Replacement Player stat is designed to settle. By that metric, James has been worth 26.7 WARP to the Cleveland Cavaliers this season, a figure which is very impressive, just shy of the NBA's top single-season mark. The fact that James has played "just" 37.7 minutes per game this season limits his value slightly. Here's the all-time leaderboard for WARP along with each player's per-minute Win% and this year's top five (through Monday):

Player              Year    Win%    WARP

David Robinson     93-94    .835    28.0
Michael Jordan     88-89    .832    27.8
Kevin Garnett      03-04    .824    27.4
Michael Jordan     87-88    .816    27.0
Shaquille O'Neal   99-00    .825    26.9
David Robinson     95-96    .830    25.9
Michael Jordan     89-90    .807    25.6
Hakeem Olajuwon    92-93    .801    25.5
Shaquille O'Neal   93-94    .791    24.9
Kevin Garnett      04-05    .801    24.9

LeBron James       08-09    .839    27.0
Chris Paul         08-09    .833    25.7
Dwyane Wade        08-09    .798    24.3
Dwight Howard      08-09    .772    20.8
Tim Duncan         08-09    .716    15.6
Pau Gasol          08-09    .656    14.9
Kobe Bryant        08-09    .651    14.4

With one game left to play, James remains a little shy of Michael Jordan's best season and MVP campaigns by David Robinson (which, alas, narrowly tops Jordan for best in modern history) and Kevin Garnett. Since Cleveland secured home-court advantage throughout the playoffs by beating Indiana on Monday, James may not play tomorrow when the Cavaliers host Philadelphia in their season finale, so this is probably his final spot.

The bigger surprise here is that Paul's season rates almost as well, on a per-minute basis, as any in modern NBA history. His performance has been overshadowed by the number of great individual seasons in the league, as well as the fact that the Hornets have been hit by injuries and have been something of a disappointment. That hasn't stopped Paul from playing even better than he did a year ago, when he was my pick for MVP over James.

It's also worth noting that even as he battles sore knees, Duncan has had an outstanding season. I don't think he belongs on my MVP ballot, but he definitely deserves a mention for his consistent play.


As I've written about extensively this season, there are two schools of thought on measuring player value--one that starts with individual stats and one that starts with plus-minus numbers that describe team impact. The interesting thing is that at the top of the league there tends to be a lot of overlap between the two methods this year. James comes out tops in the league in net plus-minus, per 82games.com, and is second in the single-season adjusted plus-minus available at BasketballValue.com.

We've heard a lot this season about Cleveland's improved supporting cast, and rightly so. Yet the plus-minus numbers show the Cavaliers to be an entirely different team with James on the bench. While the sample size is small--and probably includes a lot of garbage time--Cleveland has been outscored by 7.3 points per 100 possessions in those situations. By contrast, with James on the floor, the Cavaliers outscore opponents by 13.8 points per 100 possessions. That puts James just ahead of the other MVP contenders--and a surprise entrant in the top five--in terms of impact:

Player     Tm   Offense  Defense  Total

James     CLE    +13.8    - 7.3   +21.1
Paul      NOH    +15.9    - 3.1   +19.0
Odom      LAL    + 8.8    - 8.2   +17.0
Wade      MIA    +11.1    - 3.8   +14.9
Bryant    LAL    +13.4    + 0.9   +12.5

Adjusting for teammate and opponent quality slightly changes the leaderboard, vaulting Wade to the top of the list:

Player     Tm   Adj +/-

Wade      MIA    +23.3
James     CLE    +17.7
Paul      NOH    +15.3
Odom      LAL    +15.2
Kidd      DAL    +14.5

As a reminder, the inherent noisiness of adjusted plus-minus makes it difficult to use it to rank or compare players over the course of a single season. The standard errors are such that there's about a 15 percent chance James' "true" adjusted plus-minus rating is in fact better than Wade's, despite the apparent gap between them. It is interesting that three of the top candidates for MVP come out atop the single-year leaderboard. According to adjusted plus-minus, Odom has been more of a difference-maker this season than Bryant, who slips to 26th in the league.


It is no secret that James has emerged as one of the league's top triple-double threats and the most likely candidate to do the unthinkable--repeat Oscar Robertson's triple-double season. Analysis by Basketball-Reference.com's Neil Paine suggested James could do just that if the Cavaliers played at the same breakneck pace of the 1960s NBA.

Several years ago, John Hollinger invented a junk stat called the Versatility Index, which could also be called the Triple-Double Index. His formula multiplies the three "triple crown" per-game stats--points, rebounds and assists--and took the cube root.

Player     Tm     PPG     RPG    APG    VI
James     CLE    28.3     7.6    7.3   11.6
Paul      NOH    23.0     5.5   11.0   11.2
Wade      MIA    30.2     5.0    7.5   10.4
Duncan    SAS    19.4    10.5    3.5    8.9
Bryant    LAL    27.0     5.3    4.9    8.9

Predictably, James leads the league in Versatility Index. The same three players stand out from the rest of the league, with the surprise being that Duncan ranks just ahead of Bryant in fourth.

Hollinger's formula doesn't even take into account James' defense. He is one of 15 players in the league averaging at least a steal and a block per game. Man defense is the most improved part of James' game this season, which is why he earned a spot on my Every Play Counts All-Defensive First Team. Paul and Wade aren't far behind, but nobody in the NBA can help his team in more ways than James.

The Ballot

If I was casting a vote for MVP, it would look like this:

1. James

2. Wade. I could go either way between Paul and Wade for second place. Wade has one of the highest usage rates of all time, yet has remained an efficient scorer, and is a good enough defender to make my All-Defensive Second Team. His adjusted plus-minus makes a strong case to confirm the subjective notion that Wade is, in some sense, more indispensable to his team than any other player in the NBA.

Ultimately, the difference-maker to me is that Paul's individual stats appear to be slightly padded. Despite playing more minutes on the road (39.1 mpg) than at home (37.9), Paul sees his assist average jump from 10.3 in road games to 11.8 at New Orleans Arena. You might think that discrepancy could be explained by the fact that teams shoot better at home, but counter to intuition, the Hornets have made slightly more field goals per game on the road. By Occam's Razor, the explanation is that Paul's assist numbers are helped out by friendly scoring at home, which limits some of his advantage over Wade in terms of box-score stats.

3. Paul

4. Dwight Howard. Howard's name has yet to come up much in this column, nor has he gotten much love in the MVP discussion. WARP suggests that in terms of value, Howard has been much closer to the top three candidates than to the rest of the pack. In terms of box-score stats, he has provided more value at the defensive end than any player in the league, which is part of why I made Howard my Defensive Player of the Year.

Where Howard suffers in this breakdown is in terms of plus-minus stats. His net plus-minus (+8.9) is solid, but not spectacular. Howard's adjusted plus-minus is almost exactly average (+0.04). I suspect a lot of this is randomness. The standard error of Howard's adjusted plus-minus (6.1) is on the high side. If you break down the numbers, it appears what hurts Howard's rating is the fact that Orlando has played very well with neither Howard nor backup center Marcin Gortat on the court, outscoring teams by 6.9 points per 48 minutes in these situations, usually with Tony Battie in the middle and presumably mostly from early in the season when Gortat had yet to claim the backup job.

Alas, we're talking about 341 minutes of basketball, and it's hard to draw meaningful conclusions from what amounts to seven games. In practice, this somehow plays out such that all three Magic centers are considered neutral or worse, while other players (mostly Rashard Lewis and Jameer Nelson) get credit for the team's strong play.

This is one of those situations where I just don't feel adjusted plus-minus is a meaningful indicator. I'm much more inclined to trust the individual stats, which show Howard to belong right in the MVP debate.

5. Bryant. The Bryant question is the thorniest one of the MVP conversation. While the movement to pick him as the winner seems limited primarily to fans at the STAPLES Center chanting during free throws, a lot of people would have Bryant in the top three, even second to James. None of the numbers bear out this position, and in fact both individual and plus-minus stats show other Lakers to be slightly more valuable to the team.

Bryant's supporters will rightfully argue that he doesn't need to score like James or Wade for the Lakers to win this season, and L.A. has still been as good as any team in the league. I sympathize, but I don't necessarily agree. Ultimately, the award is about how much value a player has in a given season for a given team. In that context, Bryant ranks below the top contenders. While it is not the best strategy for his team, Bryant's value is at its peak when he is using a heavy load of possessions, as was the case three years ago when he averaged 35.4 points per game.

We often talk about the tradeoff between usage and efficiency as an overall concept, but each player has a different relationship. Specialists like Steve Novak or Joel Przybilla would see their scoring efficiency plummet if they were asked to create more of their own offense, while go-to guys can ramp up their possessions with little or no detrimental effect. Such is the case with Bryant, except we're seeing the opposite effect--reducing his possession usage doesn't dramatically improve his efficiency. So it is that despite going from using 38.7 percent of his team's possessions in 2005-06 to 32.3 percent this year, Bryant's True Shooting Percentage is virtually the same (.561 compared to .559).

Fans can argue that Bryant has been one of the league's two or three best players this season. Arguing that he's been one of the two or three most valuable is much more challenging, and I'm not seeing how he belongs ahead of the other top four candidates.

Kevin Pelton is an author of Basketball Prospectus. You can contact Kevin by clicking here or click here to see Kevin's other articles.

0 comments have been left for this article.

<< Previous Article
The Bynum Question (04/13)
Next Article >>
Handing Out The Hardwa... (04/15)

State of Basketball Prospectus: A Brief Anno...
Tuesday Truths: March-at-Last Edition
Easy Bubble Solver: The Triumphant Return
Premium Article Bubbles of their Own Making: Villanova, Temp...
Tuesday Truths: Crunch Time Edition

2009-04-17 - Playoff Prospectus: Handing Out Comps
2009-04-16 - Bad News for Boston: Garnett Out
2009-04-15 - Handing Out The Hardware: One Man's Ballot
2009-04-14 - The Envelope, Please: Picking My MVP
2009-04-13 - The Bynum Question: Comeback?
2009-04-10 - Every Play Counts: All-Defensive Teams
2009-04-09 - Every Play Counts: The Cavaliers' Defense

Basketball Prospectus Home  |  Terms of Service  |  Privacy Policy  |  Contact Us
Copyright © 1996-2017 Prospectus Entertainment Ventures, LLC.