In the process of writing another article I found myself attempting to project salaries two seasons in advance and found that my results were very crude and rudimentary. Namely, players who were entering their first eligible year of arbitration or a successive year of arbitration. I’ve read up a lot on the topic of value and surplus value and found a basic and inaccurate market value of $8m per 1 WAR. On average, of that value arbitration players earn 25% of that for their first year, 40% of that for their second year & 60% of that for their third year (I attempted using 15% for super-two arbitration eligible players). So I ended up taking this rudimentary approach and seeing how accurate it was for this year’s eligible players. As expected, it was very well off. The reason being is that the arbitration process does not use WAR in it’s evaluation, and the study that brought us the numbers above were taken on average over the span of 8-10 years when $8m did not equal 1 WAR anyways. Instead, the evaluation uses more quantity over quality type numbers that can correlate to a dollar amount easier than a ratio or average. Stats like batting average, ERA, FIP and SO/9 are still taken into account, just not nearly as much as stats like innings pitched, plate appearances, runs, RBI, HR, XBH, SO, Wins and Saves.
I don’t want to deliver guesstimates in my content, I want to deliver richer content and provide a true projection. So without any background or previous experience in calculus or statistics, I decided to develop my own arbitration projection model. I was not intending to make this a yearly task as I only wanted to produce content relative to the Rangers. However, the model turned out to be a lot more accurate than I thought it would. Take that with a grain of salt, as only 1 year’s worth of data has gone through it. I won’t get my hopes up. There is a lot to improve, change, tweak and add to it in order to improve the projections. I was hoping to get my initial projections out before last Friday but the model took me longer to build than I thought (about a month), and I’ve been busy since the new year. Onto the projections: