Why Rays’ Kevin Jepsen Received $3.065 Million in Arbitration

The deadline for teams to exchange figures with their arbitration-eligible players can be a contentious time. Luckily for the Tampa Bay Rays, all their hard work proved to be enough this offseason as they avoided hearings with all eight of their eligibles. What were the factors that led them to the salaries to which they agreed with their players?

Earlier today, we put Alex Cobb‘s case up for discussion, and now we will talk about Kevin Jepsen, who received $3.065 million in arbitration.

Jepsen, who just went through the arbitration process for the second time, is coming off a breakout season in 2014. The 30-year-old right-hander delivered a 2.63 ERA, a 10.4 K/9, a 3.2 BB/9, and a 0.6 HR/9 in 74 appearances and 65 innings pitched. He had achieved success before, most notably with a 3.02 ERA in 2012, but he had never looked as dominant as he was last year. Here is how his year compared to several similar pitchers.

To be super clear, all of these statistics came from pitchers entering at least their second go-around in arbitration and are their statistics from the previous year only.

The most interesting thing that jumps out is how ERA and especially ERA+ do not have as much significance as you would expect them to. Instead, the most telling statistic by far is holds, and then we need to make sense of everything else. The hold is such a random statistic–a reliever receives one when he appears in a save situation prior to the ninth inning and “holds” his team’s lead–but for whatever reason, it is important here.

If we did a linear regression using holds to predict a pitcher’s arbitration raise, we would get an expected arbitration raise $1.42 million for Jepsen’s 22 holds. However, he actually received $1.5625 million, so there are more things we need to talk about here.

One major factor working in Jepsen’s favor was how similar his case was to Luke Gregerson‘s from last year. While he fell 3 holds short of Gregerson’s total, he made up for it by beating him in ERA, games, K/9, and groundball rate. However, his FIP was actually higher than Gregerson’s because his walk and homer rates were higher, and that was enough for him for him to end up with a lower number when combined with the holds.

Ever more puzzling, though, is how he ended up getting more than Robertson. Not only did Robertson record 8 more holds, but he also beat him in strikeouts and walks on his way to a much better FIP (2.49 to 2.78). How were Jepsen’s edges in games and groundball rate enough to counteract that?

One thing that is quite counterintuitive is that games are a much better predictor of raises than innings pitched. The model using only holds explains 60.6% of the variation in raises–adding in games as a second predictor increases that to 85.2%, but adding in IP only gets it to 61.4%. For some bizarre reason, games matter a lot while innings pitched don’t at all.

That really doesn’t make any sense, but Jepsen’s statistics put him in a particularly good spot to take advantage of that quirk. He threw 74 games but only 65 innings, and luckily for him, it is the 74 appearances that count a lot more. His edge versus Robertson in games is quite significant (9), and that was enough to negate most of his deficit in FIP.

Speaking of oddities in the system, Robertson was also penalized for another reason: his seven losses. While losses are far from the most telling statistic for a reliever, it means something in this crazy process that he lost seven games while Jepsen lost only two. Good luck finding any legitimate way to value Jepsen’s 2014 more than Robertson’s 2012, but in salary arbitration, Jepsen had enough to beat Robertson.