by Daniel Gettinger 

Before I begin, I want to take a quick opportunity to introduce myself.  I am Daniel Gettinger, and have been blogging over at There are Better Deals in August for the past few months.  I am very happy to announce that I will now be a part of Friar Forecast, and would like to thank Myron for the opportunity to join him, Mike, and Ben/Padman in contributing to the site.

Back in November, the Padres withdrew their $4 million contract offer to Trevor Hoffman.  Hoffman, who did not respond to the offer in a timely manner, clearly believed he was worth more than $4 million.  The Padres, who refused to increase their offer, thought $4 million was more than generous, and was the maximum they were willing to spend to retain their closer.  This begs the question: what can Hoffman expect to earn as a free agent? 

To try and answer this question, I have built a simple model that predicts the average annual salary of a free agent reliever.  The model is not intended to predict performance, or return on investment, but merely what a reliever can expect to receive as a free agent.

The model is constructed using data from the 2007/2008 free agent class.  Because teams (try to) pay players for future production, not past production, the model uses predicted performance rather than actual past performance.  Marcel forecasts were used as a proxy for predicted performance.

I spent a while messing around with various combinations of independent variables, but the best equation (as is often the case) ended up being one of the simplest.  Predicted innings pitched, predicted runs allowed, and a dummy variable noting whether a player is considered a “closer” were able to explain 83% of the variation in average annual salary received.  All of the variables were highly significant.  The final equation was:

Average Annual Salary=-3.2+0.3462*mIP-0.4772*mR+6.2627*CLOSER(dummy)

To test whether the model continued to accurately predict free agent paydays, I tested its predictions against what 2009 free agent relievers have signed for.  Due to this year’s rough economic climate, I expected to see the model overestimate reliever salaries.  This was in fact the case, with only Kyle Farnsworth exceeding his predicted salary.  On average, relievers received approximately $500K less than the model predicted they would receive.  Once the change in economic climate is corrected for, this simple model was able to predict average annual salary relatively accurately.

Notable relievers who have signed this year include Francisco Rodriguez, Kerry Wood, Damaso Marte, and Jeremey Affeldt.  The model predicted the relievers to earn an average annual salary of $13.47 million, $10.32 million, $3.92 million, and $3.70 million respectively.  Their actual salaries were: $12.33 million, $10.25 million, $4 million, and $4 million.  As a super quick estimator, the model seems to work pretty well.

Now, the whole point of this exercise was to estimate what Trevor Hoffman can expect to receive as a free agent.  This is where things get interesting.  As an established “closer,” the model predicts Hoffman would command a $9.45 million salary.  The Padres however probably recognize that both “closer experience” and the relative importance of protecting 9^th inning leads are overrated.  It is quite conceivable that the Padres care not to pay for abstract qualities such as whether somebody is a closer, but instead mostly for measurable run prevention.  When Hoffman is not classified as a “closer,” but instead as a typical reliever, his predicted salary is $3.2 million. 

Remember, the Padres offered Hoffman $4 million.  The difference between their offer, and his predicted salary (discounting the closer premium) could be attributed to a number of factors including: Hoffman’s large fan support in San Diego, a slight boost for his proven reliability in the “closer role,” and forecast error.

Regardless, the model provides some interesting insight into the divide between Hoffman and the Padres.  The Padres have no desire to pay the large “closer premium,” and instead wish to pay Hoffman a salary in line with his predicted performance.  Meanwhile, Hoffman sees no reason he should not get paid like other closers.  After all, he has earned more saves than any other player in major league history.  Whether the Padres and Hoffman are able to close this gap remains to be seen.