Michael Jordan Career Plus – Minus Stats


DISCLAIMER – It turns out the +/- figures below taken from the basketball-reference game logs here are actually team margin of victory per game and not on-court / off-court +/- stats. Enjoy:

I’ve compiled season-by-season raw plus / minus stats for Michael Jordan from 1987-1998 (less 1995 when he barely played). Why, you ask? Because I’ve always wanted to know how his +/- numbers stacked up. Observe:

Season Age PER TS% MOV not +/-
1986-87* 23 29.8 0.56 0.9
1987-88* 24 31.7 0.6 3.4
1988-89* 25 31.1 0.61 1.6
1989-90* 26 31.2 0.61 3.3
1990-91* 27 31.6 0.61 9.1
1991-92* 28 27.7 0.58 11
1992-93* 29 29.7 0.56 6.5
1995-96* 32 29.4 0.58 12.2
1996-97* 33 27.8 0.57 10.8
1997-98* 34 25.2 0.53 7.1

The plus minus figures didn’t go back before 1986, so I couldn’t get Mike’s rookie year. I left off year 2 because he missed most of the season with a broken foot. I also left off 1995 because he missed most of the year playing baseball, and I left his Wizards years off because they never happened. They never happened!

We’re left with the 10 years that we really gauge Jordan’s career on.  What I find interesting is that the numbers match up with how I think about the sport.  PER is highest when the team isn’t as good and the +/- numbers reflect that.  As the team got better and Jordan focused less on his own production and more on sharing responsibilities and doing the little things to help the Bulls function in the triangle, the PER dropped, and the +/- spiked.  In fact the +/- backs up Bill Simmons’ claim that Jordan peaked in 1992.  The boxscore metrics see 1992 as the worst of MJ’s post-foot injury and pre-baseball seasons.  The +/- says it was Jordan’s best pre-baseball season.  In fact , I’m sure someone smarter than me could find a way to combine box score metrics with +/- so that the end result is a number that factors team success and individual production together to help eliminate some of the player-rating inaccuracies inherent in each.

Obviously raw +/- is of limited use, but it does demonstrate Mike’s contributions to his team in a broad way.  It shows how dominant the Bulls were with Jordan on the floor with no other context factored.  It doesn’t consider quality of teammates, quality of opponents, five-man units, or any of the other variables that cause so much noise.  On the other hand these numbers actually do relate to winning rather than simply to filling out a box score.

For a modern comparison, LeBron James‘s raw plus / minus actually peaked in 2009 with the Cavs at 10.9.  Last year when the “is LeBron better than anyone else has ever dreamed of trying to be in their wildest fantasies” talk began, LeBron’s raw +/- was 9.5.  So, taken with a grain of salt, Jordan’s peak years of 9.1, 11, 12.2, and 10.8 are in line with what we see from the best player today, which is what we’d hope to see if we thought there was any value in the raw plus / minus numbers, which I’m not sure of yet.  EDIT: Per the commenter below this comparison is faulty.  The Cavs and Heat MOV was actually even lower, but that’s not an indictment of LeBron in any way.  It just shows how great Chicago was.

I think I’m going to have to run the numbers for a few more players to see if they give consistent results.  I’ll keep y’all posted.  That’s a blog joke.  Sorry.


Tags: , , , , , , , ,

6 Responses to “Michael Jordan Career Plus – Minus Stats”

  1. Manav Says:

    Wow. From where were you able to acquire pre-97 +/- data?

  2. James Says:

    Is this +/- per 48 minutes or per 100 possessions? Is this on court +/- or actual plus/minus?

  3. Manav Says:

    Actually just tried to calculate net ON/OFF differential myself, and I realized that your numbers are actually Chicago’s team MOV (or +/-) per game numbers, NOT Jordan’s on court +/-. However, for Lebron you DID cite his on court +/-. If you were to look at the same stat for James, it would be +8.9 in 2009 and +7.9 last season (2013). So you’re comparing two completely different metrics here, one indicates how much the team outscored their opponents by (Jordan figures) and the other indicates how much the team outscored their opponents by with the specific player on the court (Lebron).

    • jpalumbo Says:

      The +/- figures came from the gamelogs here: http://www.basketball-reference.com/players/j/jordami01/gamelog/1987/ So you’re saying I’ve been playing with Chicago’s margin of victory and not MJ’s actual on court / off court +/-? If you’re right, that’s disappointing… And a huge waste of time. Yeah, looks like that’s the case. I spot checked the last three seasons on MJ’s list and they match Chicago’s MOV. Damn it! Thanks for pointing that out!

  4. Manav Says:

    Yes Basketball-Reference only has individual player +/- back to 2001. NBA.com’s stat database recently saw an overhaul, and they now have that data back to the 96-97 season, but that’s as far as we’ve been able to find publicly available thus far.

    For 96-97 we can see that Jordan was a +10.0 in RawPM, and a +7.3 in 97-98. However, this is just raw +/-, and in order to really get a proper picture of the kind of differential these guys were creating, we need regression and/or a statistical +/- model (regression of box-score coefficients on the regression model). There are various sources of that kind of data for 01-13, but the 97-00 data is still being worked into a usable format.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: