Plus - Minus in the NBA

If you just listen to game commentators and play fantasy basketball, you probably think that Monta Ellis is an exceptional basketball player. 25 points, 5 rebounds, 4 assists, 2 steals...these are hall of fame numbers when projected over a player's career. However, NBA executives and talent evaluators know that Ellis not actually a very good player because his team performed noticeably worse when he was on the floor than when he was on the bench. In fact, Golden State was outscored by their opponents by almost 11 points over the course of the 2009-2010 season while Ellis was on the floor but actually outscored opponents by a 4 point margin when he was on the bench. How do we know this? Let me introduce you to basketball's most useful-and most understood-statistic: Plus-Minus.

Plus-Minus measures a player's net contribution to his team by ignoring real numbers in the typical statistical categories-points, rebounds, assists, steals, and the like-and instead evaluating his team's relative performance when he is on the floor. Put simply, the statistic measures his team's net scoring versus their opponents when the player (or set of players) is on the floor and comes up with a net positive or negative number. In Monta Ellis' case, his team was outscored by 397 points over the season (-397) when he was on the floor. If you believe the statistic, that means that Ellis is actually a net negative contributor to his team, which is a far cry from what his raw numbers would suggest. Having said that, let's take a look at some of the NBA “stars” who had surprisingly negative Plus-Minus ratings:

Jonny Flynn -626 Brook Lopez -583 Al Jefferson -516 Monta Ellis -397 Corey Maggette -286 Charlie Villenueva -285 Andray Blatche -279 Chris Kaman -272 Troy Murphy -250 David Lee -250 Al Harrington -225 Kevin Love -222 Baron Davis -192 Ben Gordon: -190 Darren Collison -177 David West -176 Kevin Martin -175 Andrea Bargnani -170

Thanks SSShupe at flickr.com for Monta Ellis picture.

The first thing you might notice is that the statistic is clearly imperfect because it doesn't measure how strong a team is overall. For example, Brook Lopez has a -583 plus-minus rating, which was in the NBA's top ten worst overall, but Lopez is actually a very good player. The reason his rating was so awful is that the 2009-10 Nets were a historically bad team that was outscored in nearly every game and Lopez played so many minutes over the season that he was bound to have a ridiculously awful rating. The team might have been marginally better with him on the floor but was still atrocious.

Thanks Chamber of Fear at flickr.com for Brook Lopez picture.

That said, even in Lopez's case, the status reveals a few interesting points that no other relevant statistic shows. First, even though Lopez is a rare breed-a twenty one year old 20-10-2 seven footer who can knock down free throws-he lacks speed and length on defense and, without a capable defensive power forward to protect him, gets exposed in the paint defensively. That is why, despite his abilities, Lopez did not add appreciably to his team's 2009-10 performance. In Lopez's case, the stat would be most useful as an indicator of the kind of power forward needed alongside Lopez (a defensive behemoth with length and the opposite of recently acquired Troy Murphy) but, in other cases, the Plus-Minus stat exposes so-called “star” performers as net negatives to their teams who should be jettisoned immediately.

Monta Ellis is a great example of this type of player. Ellis puts up star-level offensive statistics but he never passes the ball out of double teams, takes an incredible amount of shots that he converts at a net inefficient rate, and sees defense as an activity best left to his teammates. As a result, his team was terrible with him on the floor but, unlike in Lopez's case, his team improved drastically when he went to the bench: from -11 to +4 overall. This is where the Plus-Minus statistic is critical: it allows GMs and scouts to distinguish between statistical and real contributions and, on top of showing management which players need to be shipped out, it highlights players who are actually critical contributors to their teams' success.

Not surprisingly, many of the NBA's stars headed the positive side of the Plus-Minus list. LeBron James led all players and Dwayne Wade, Kevin Durant, Kobe Bryant, and Pau Gasol all made the top twenty. However, while the contributions of those types of players on both ends of the floor is obvious to most fans, there are a number of players who are incredibly useful to their team's whose contributions often get overlooked:

Anderson Varejao: +511 Vince Carter: +472 Matt Barnes: +420 Channing Frye +361 Mike Bibby +353 Rashard Lewis +324 Chris Andersen +322 Richard Jefferson +310 Lamar Odom +305 Nene Hilario +301 Nick Collison +284 Andrew Bogut +223 Jermaine O'Neal +188 John Salmons +186 J.J. Redick +181 Ryan Anderson: +166 Boris Diaw +162 Marc Gasol +110

Thanks Keith Allison at flickr.com for the Anderson Varejao's picture.

In many ways, the “Good” statistics derived from the use of plus-minus are even more useful than the “Bad” ones. Did you think Anderson Varejao was worth eight million dollars per year? Without plus-minus, there would be no reason to think so but the statistic shows the true impact of the Brazilian's hustle and defensive abilities on his team's performance. Having a +511 rating is no accident-even as part of a group as competent as the 2009-10 Cleveland Cavaliers-and using Plus-Minus allows GMs to see that rather than overlook his contributions.

Ultimately, as Brook Lopez's extremely negative rating shows, Plus-Minus is nowhere near perfect. That said, it is extremely useful as a comparative performance statistic and fans interested in trying to understand the true contribution of each player to their team will definitely find it useful.

Top of This Page