The Importance of Being Lamar: Why mismatches are so great.

I love Lamar Odom.  And not just because he has a really high tolerance for sugar.  I love him because he (and other players who create mismatches) are so underappreciated.  I'm going to spend the rest of this post in a hypothetical scenario with entirely made up numbers to show you why I feel that way.

Let's pretend that defense (separating out "perimeter defense" and "post defense") can be expressed as numbers from 1-100, with 50 being the average defender in the NBA.  Let's also pretend that height has a defined effect on defense, such that if you are more than two inches shorter than the player you're guarding, you lose 5 points of defense per inch (why 5 points?  I don't know).  Let's further fabricate a speed rating for every player, also from 1-100 (with 50 being league average), and say that being one point slower than your opponent makes you one point less effective at guarding him.

Now, let's fictionalize an NBA team with these characteristics:



POS Per Post Ht Speed
G 60 30 6'1" 90
G 60 40 6'5" 70
F 55 55 6'7" 60
F 40 60 6'10"45
C 30 60 7'0" 35

Let's say Lamar Odom is on a similarly constructed team of his own and playing "power forward" (I use quotes partially because I tend to dislike the classically defined positions in basketball, but also because Lamar Odom in particular, as you'll see, falls outside of these definitions).  Defensively, his versatility allows him to adequately cover other power forwards, small forwards, and even some guards and centers, while even rebounding well for his position (his defensive rebound rate was eighth in the league, surrounded mostly by centers in the rankings).  But, he also provides huge value offensively.  Which box score stat shows this?  None of them.  He doesn't score a bunch.  He gets some assists, but not a ton.  His offensive rebound rate, while still nice, isn't amongst the league leaders.  He does do the following -

Guarding a similar team to themselves, those five defenders above have effective defense ratings of 60, 60, 55, 60, 60 (I'm assuming the three smaller offensive players play primarily on the perimeter and the two larger players play in the post).  Now, put Lamar Odom on the offense (and we'll assume the following teammates are the smallest threats offensively: the small guard and the small forward - Derek Fisher and Ron Artest?).  We'll call Lamar's speed "45", just to not give him any additional advantage beyond his mismatchability (I just coined a term).  Since Lamar primarily plays offense on the perimeter, if the defense played straight up, without cross-matching, they'd have effective ratings of 60, 60, 55, 40, 60.  All other things being equal, the defense will probably cross-match to more effectively play defense, because you'd rather hide your bad defensive match-up on someone who isn't an offensive threat.  One possible cross-match scenario would be this:

The guard match-ups remain the same, but the smaller forward covers Lamar, while the larger forward covers the offense's smaller forward.  This gives us 60, 60, 37, 55, 60 now, but at least the 37 is on a non-threat.  So what happens here?  Lamar is defended fairly well, so he ends up producing somewhat normal offensive numbers, but now a teammate has the mismatch (created by Lamar's presence) so that he's able to more effectively produce.

So, if this effect of Lamar being on the floor doesn't show up in his box score stats (points, assists, rebounds, etc), where does it show up?  How about as one cause for the fact that he's currently eighth in adjusted plus-minus over the last two seasons, while posting pretty mediocre box score stats (his PER of 16.6 and 15.9 over the last two seasons are barely higher than the league average of 15.0).

So, did I just make up numbers to illustrate my point?  Sure.  But, you can change the numbers and it will still work - creating mismatches doesn't necessarily benefit the player responsible in his classic stat line (in fact, in some scenarios it could end up hurting him, if he ends up with a better defender covering him), but it certainly benefits the team.   (Rashard Lewis is another player who has a similar effect - pedestrian box score stats, but tenth in the league in adjusted plus-minus over the last two seasons, and unlike Lamar, you'd have a harder time explaining the discrepancy by saying he's a very good defender).