Contract Year Effect? Free Agent Effect?

Sports fans often talk about the "contract year effect" – the idea that a player plays really hard in the final year of a contract, signs the big contract, then reverts to poor play (shades of Mark Blount anyone?).  As sports fans we also wonder about free agents: why didn’t the original team retain their services? Are they a bad seed? Not a team player? Do they have bad habits no other team knows about? 

Sports fans often talk about the "contract year effect" – the idea that a player plays really hard in the final year of a contract, signs the big contract, then reverts to poor play (shades of Mark Blount anyone?).  As sports fans we also wonder about free agents: why didn’t the original team retain their services? Are they a bad seed? Not a team player? Do they have bad habits no other team knows about?  We Celtics fans perhaps wonder about this in regards to James Posey: why didn’t Miami want to keep him?  Luckily, early signs point to him being a stand up guy .  Of course, it is not a one way street – it doesn’t just depend on whether or not a team wants to resign a free agent; a free agent also has to decide where to play, and may take a change of scenery or a chance at a title over cold hard cash. 

It turns out there is an economics article that looks at these issues in Major League Baseball.  In 1984 Kenneth Lehn published an article in Economic Inquiry titled "Information Asymmetries in Baseball’s Free Agent Market" (Economic Inquiry, 22(1): 37-44).  Lehn suggests that teams have better information about their own free agents than other teams do, and so will resign the players they think are better and let the other ones go to a new team.  Lehn has some good empirical evidence to back this up.  He looks at the amount of time players spend on the disabled list before and after signing a new contract.  Below is a recreation* of a table from Lehn’s paper (table 1, p. 40) that shows the average annual number of days spent on the disabled list:

*note the numbers are the same as in Lehn’s paper, but I have changed some of the wording

There are two main things to notice.  First, looking left to right on the table, we see that all players spend more time on the DL post contract year (this difference is statistically significant).  This seems to suggest some support for the "contract year" phenomenon: players play well right before signing a big contract and play worse thereafter.  Second, focusing on the last column and looking top to bottom, we see that, post contract year, players who resign with the same team spend less time on the DL than players who sign with a new team (this difference is statistically significant).  This seems to suggest some support for the "free agent effect": players who sign with new teams may do so for a change of scenery, but they may also just not be that good, and their current team knows that better than any other team.  

There are a few caveats to the study.  First, the empirical setting is specific to a 1976 change in how MLB teams handled free agents, and Lehn provides evidence that the effects diminish over time.  Second, the statistical significance of the results is largely driven by the performance of MLB pitchers.  Pitching is such a specific position and it is unclear how (if at all) these results would generalize to other professional sports like basketball.  Nevertheless, the results are interesting and provocative.

How does all this relate to the Celtics?  Without intending any disrespect to specific players, allow me to raise the following points: First, if Tony Allen is not resigned, maybe that is a good thing (maybe Danny and Brian Doo know something we don’t).  Second, there may be a good reason why Atlanta did not resign Batista; what might that reason be and is it any cause for concern for the Celtics organization?