Fun With Payroll Spreads | The Process Report

Fun With Payroll Spreads

(Inspired by Dan Fox after accidentally stumbling upon his piece of the same ilk in the Prospectus archives)

A few days ago, my latest Prospectus article attempted to evaluate the Indians’ past few seasons by comparing their payroll to the league average and using Tom Tango’s methodology to pull their expected playoff chances. The Rays’ payroll is right around $40 million depending on the complexion of the 25-man roster –i.e. whether Felipe Lopez makes the team—, a figure which will be well below average. Assuming the league mean will remain nearly the same (about $90 million), then the Rays’ playoff hopes should be virtually non-existent. PECOTA (and virtually every other projection system) suggests this is not the case to varying degrees.

Nevertheless, the payroll will remain a go-to issue with the narrative being molded to fit the team’s performance. If they perform as they should –ostensibly well— then the 2011 Rays are the efficiency kings once again; if not …well, you can guess. A reduced payroll brings an interesting thought to mind: Is the team spending its money more evenly throughout the roster?

In order to find out, I took the team’s salaries from each of the past four seasons, as well as their projected 2011 salaries and grabbed the average and standard deviation of each. From there, I figured out the coefficient of variation –which is to say that I divided the standard deviation by the average—the higher this number is, the more top-heavy the roster is with a lower number meaning a more even distribution of dollars.

For the sake of brevity, I did not include the standard deviations here, but if you just must have it you can by simply multiplying the two numbers together. Here are the results:

2007
Average: $893,463
Coefficient of Variation: 1.03

2008
Average: $1,460,687
Coefficient of Variation: 1.08

2009
Average: $2,183,208
Coefficient of Variation: 1.08

2010
Average: $2,663,832
Coefficient of Variation: 1.11

2011
Average: $1,563,140
Coefficient of Variation: 1.01

Despite the third highest average salary, this squad will have the tightest distribution of salaries. Such a development probably isn’t too surprising once you compare the top salary to some other arbitrary placement, like say, the top salary to the fifth most. Johnny Damon will make a bit over $5 million this season, while Kyle Farnsworth comes in just shy of $3 million. Last season, Carlos Pena brought home a little more than $10 million while fifth place finisher Jason Bartlett finished $6 million behind. And so on. One thing to note about these numbers is that I only took the salaries of the projected 25-man roster. The team will undoubtedly use at least 26-28 players this season, meaning the exact numbers will change depending on the salaries those additions bring. Still, I wanted to take this one step further and apply it to the bullpen:

Let’s try this with one phase of the team: the bullpen.

2007
Average: $455,033
Coefficient of Variation: 0.29

2008
Average: $1,375,220
Coefficient of Variation: 0.92

2009
Average: $1,913,374
Coefficient of Variation: 0.77

2010
Average: $2,164,950
Coefficient of Variation: 1.06

2011
Average: $1,014,214
Coefficient of Variation: 0.74

Of all the modifiers and superlatives slapped on the 2007 pen, who would have thought highest equality would have been one? Obviously, the Major League minimum kept the number from sinking further into the abyss, but wow, having a bunch of fellas making about the same does not lead to better performance. On the flip side, the 2010 pen is about equal to the 2009 pen in spite of a lower average pay.

What does all of this mean? Not much, although, Fox did find an inverse relationship between winning percentage and the coefficient of variation, thus leaving the 2011 Rays in a better position than their predecessors in at least one regard.



Leave a Reply

#layout { padding-left:20px; }