More thoughts on NBA economist: spending and wins

I linked earlier to this interview of economist Kevin Murphy regarding the NBA lockout.

http://www.nba.com/2011/news/features/steve_aschburner/10/27/lockout-q-and-a-kevin-murphy/index.html  It's in the player’s interest to have as few restrictions as possible on spending so it isn’t surprising that Murphy brings up the weak correlation between spending and wins.  But beyond management skill there may be other reasons why the link between spending and wins is weak.

Wins is a simplistic measure of outcomes.  There is a pass/fail element in the NBA that just looking at wins misses.  The most notable is making and missing the playoffs, and closely related, finishing in the top four of the playoff teams and being in the bottom four.  Because the league is split into two conferences and those conferences aren’t of equal quality the payoff for wins isn’t equal and thus the incentive to spend is also unequal. 

For example, in the East, Indiana made the playoffs with a 37 – 45 record, while Houston (43 – 39), Phoenix (40 – 42) and Utah (39 – 43) missed the playoffs in the West.  It would make sense for Indiana—and by extension other marginal teams in the East—to pay more for wins than a comparable team in the West because the payoff will be greater, namely a spot in the playoffs.

Teams that are badly situated will have to overspend.  Players are likely to want to play for franchises that are likely to win, are located in desirable cities, and have nice weather.  So bad teams or teams located in small and or cold cities are going to have to spend more to attract comparable players.  If you’re offered the same amount to play in Miami or Minneapolis during the winter, where would you go?

Rookie contracts:  There are only 30 teams so it wouldn't take all that many outliers to result in a weak link between spending and wins.  One factor that would result in a high ratio of wins to money spent is landing a high quality player through the draft.  Getting lucky in the draft lottery can result in a team (see Chicago, Oklahoma) having a key player or two who is significantly under paid.  And as Murphy points out, because there are only five players on the court and no limit on how much the top player handles the ball or shoots this can have a large impact on wins to spending.

Multi-year contracts:  The existing labor agreement and market results in no cut contracts for multiple years.  Thus, it shouldn’t surprise that spending won’t tie very closely with winning since franchises aren’t able to adjust quickly to the lack of results.  For example, Cleveland is stuck with players and contracts that might’ve made sense with Lebron James on the team but don’t without him.  Once a player signs their second or third contract for more money they may get hurt or more likely lose interest in the game but the team still has to pay them.

In short, it doesn’t surprise me that spending more doesn’t result in wins in a particular season but that result should be treated skeptically. I would suspect that the correlation would be greater if you looked at a longer time frame, say five or ten years.  It would also help if you allowed for other factors like draft position, climate, and expanded the definition of success by taking into account making or missing the playoffs.