Nylon Calculus: Big men are getting under-valued in free agency

This week, I ran my not too great free agency model against this year’s free agent class. The model predicts the percent of cap space the average annual value (AAV) of their contract using age, position, and statistical profile. For a number of years Win Shares in the prior year has been the best predictor of AAV, with adjustment for age, usage and games started.

In prior years I have tested position differences, and found little, if anything, that wasn’t captured in production data. That appears to have changed this offseason. Centers and power forwards are getting paid less this off season, systematically under-performing the model projection compared to other positions.

On average, centers are getting paid almost $3 million less than the model projects, while power forwards are getting about $1.7 million less than projected. Meanwhile, small forwards, point guards and shooting guards have, so far, gotten paid more than the model estimated.

More simply, centers and power forwards are getting paid less than their production as measured by Win Shares, or Value Over Replacement Player (VORP), a stat derived from BPM. Those stats are shown below (all dollar figures in millions, this is the NBA after all):

In the above table the last three columns tell the free agency tale so far. Centers and power forwards are under-performing the model and getting less for their box score production. I also checked on ESPN’s RPM stat and found the same result, free agents centers have signed at $1.5 million per RPM win, while point guards and wings have signed at around $3 million per RPM win.

Next: Was 2018 the best rookie class since the merger?

In any market perceived value more or less equals value, at least in the short term. Right now, league front offices are not valuing centers. Whether that’s catching up to Warriors era reality or an overreaction to it, time will tell.