Over the last few days, I have been discussing some of the higher profile upcoming arbitration cases. I rely partly on my arbitration model developed exclusively for MLB Trade Rumors, but will also break out some interesting comparables and determine where the model might be wrong. 2019 projections are available right here.
Nolan Arenado already earned $17.75 million in 2018, and put together another storied season. He led the league with 38 home runs, and hit .297 while knocking in 110 runs. He won yet another Gold Glove, yet another Silver Slugger, made the All Star Game, and again finished in the Top 5 of MVP Voting. That makes for his sixth, fourth, fourth, and third consecutive season in each of those accomplishments.
Entering his fourth year of arbitration, it stands to reason that Arenado will earn more than any player ever has in arbitration. After all, Josh Donaldson currently holds that record at $23 million just last year. Since the arbitration system generally awards raises based on platform year performance (except for in the player’s first year of arbitration eligibility), Donaldson’s level of pay would only reflect a $5.25 million raise, which is frankly too small for Arenado.
Indeed, the model predicts a big move up to $26.1MM for Arenado, which would represent a raise of over $8MM from his 2018 salary. Notably, too, there is some uncertainty present, since he earned $17.75 million last year only as part of a two year deal. Players do sometimes get treated differently in the event of returning to arbitration following multi-year deals. In the industry, this is often termed a “re-slot.” Probably the most recent notable example of such was Bryce Harper. In 2017, he received an $8.625 million raise from $5 to $13.625 million after a two-year deal. In that case, Harper had won the MVP Award in the first year of the two covered by the earlier agreement, and may have been treated as someone who would be re-slotted. In this case, if we were to estimate what Arenado would have earned in arbitration in 2018 based on his 2017 performance, we might have guessed $20.7 million rather than $17.75 million.
So, if anything, we would probably expect Arenado to argue for more than $26.1 million based on a re-slotting. Already in uncharted waters in terms of salary, that may or may not be a hard sell. After all, at that point, we would be looking at a “Kimbrel Rule” situation. That “rule” is named after Craig Kimbrel, and basically describes the maximum amount that we should expect a player to break the record raise for his service class. This rule exists because Craig Kimbrel’s first year of arbitration eligibility was so unique that my model had him earning significantly more than any closer had ever earned. We therefore capped his projection. The Kimbrel rule as applied to Arenado would give him a $26.75 million maximum, not much different than his $26.1 million projection.
If we look for comparables, it’s hard to find any for Arenado regardless of the method used to evaluate him. Donaldson would actually be a floor in my estimation. He hit .270 with 33 HR and 78 RBI in 2017, but in just 496 PA. With Arenado stepping up to the plate a whopping 673 times, and hitting .297 with 38 HR and 110 RBI, he will surely not command a smaller raise than Donaldson’s $6 million.
One potential comparable for Arenado could be teammate Charlie Blackmon, who produced .331/37/104 the year before last, and got a $6.7 million raise. We might expect that to be a rather close match for Arenado, although Blackmon’s ultimate raise was called into question by other agents at the time. Still, if we think that Arenado would come in at about a $6.7 million raise, that would land him at a $24.45 million salary — less than his $26.1 million projection. On the other hand, if Arenado’s Gold Gloves or higher service time factor in, we may still be north of that and closer to his projection.
Arenado will be an interesting, record-breaking case this year. With several complicating factors impacting his case, there’s a fairly large spread of dollars theoretically open for discussion. Ultimately, I suspect that the model is pretty close, but I could see being off by a couple million in either direction.
