Finding am empirical way to evaluate NBA team chemistry may seem an impossible task, but Numeranda may be onto something with its latest post (via Basketball Reference, whose own post on the subject brought this one to my attention) using Adjusted Plus-Minus statistics to evaluate the "chemistry" of a given five-man unit. Here' the method:
With these caveats out of the way, here was my thinking: we have at our fingertips (thanks to Aaron [Barzilai, who runs basketballvalue.com]) two measures of APM, an individual measure and a five-man unit measure; is there some way to connect the one with the other? My first instinct would be to simply add the individual APMs together and see how they compare to the five-man unit APM. If player impact works in a simple, additive fashion, the two measures should be roughly equal. Is that the case?
Essentially, this method evaluates the extent to which a given unit over- or under-achieved relative to the adjusted plus-minus figures of the players in it. What does this definition say about the Orlando Magic?
For starters, that the Magic's starting group last year of Jameer Nelson, Vince Carter, Matt Barnes, Rashard Lewis, and Dwight Howard slightly over-achieved, with a 1.06 chemistry rating. More interesting, to me, is the great extent to which replacing Barnes with Mickael Pietrus causes that lineup to under-achieve, with a staggering -9.79 rating. That five-man group should play nearly 10 points higher than it did, which suggests at least one of the following statements is true:
Barnes is a great fit with the Magic's four All-Stars
Pietrus is a poor fit with the Magic's four All-Stars
With Barnes moving on--just not to the Toronto Raptors, as initially suspected--the Magic will call on either Pietrus or new addition Quentin Richardson to start at small forward. Numeranda's data indicate that Richardson is the better choice. Add them to Eddy's list of reasons why that job should be Richardson's to lose.