The First Sixteenth of the Hosmer Deal Is Complete

The 2017-2018 offseason was not one of the more exciting winters in memory, to put it mildly. A large part of that, no doubt, was the result of a relatively undistinguished free-agent class and the absence of some larger clubs from the market, teams saving their ammunition for the likely more exciting 2018-2019 period. Add into that the hard-to-gauge effects of more unanimity among front offices in how to evaluate veteran players and the whispered rumors of the collusion poltergeist, and it was a formula for not a lot happening. And not a lot happened.

For about three weeks around the holidays, the only news in town was the rumbling surrounding Eric Hosmer’s new home. Now, in most offseasons, Eric Hosmer wouldn’t be one of the marquee free agents, having been a rather up-and-down first baseman with some high points, but also some low ones, enough so that he entered the 2018 campaign having never strung together consecutive years of one or more wins. The 2017 season was one of the highlights, however, with Hosmer avoiding those half-long slumps that doomed 2014 and 2016 to sub-mediocrity. It was a legitimately excellent season, Hosmer hitting .318/.385/.498, to the tune of a 135 wRC+, and reaching that four-win mark that serves as an informal threshold for an All-Star season.

In the end, the Royals attempted to retain Hosmer, though the truth of whether he was actually offered $147 million, as the rumors went, will probably be lost in history unless Scott Boras writes a tell-all book after his retirement. San Diego, a team in the middle of their own rebuild, signed Hosmer to an eight-year, $144 million contract, with an opt-out clause exercisable by Hosmer, allowing him to forgo the last three years and $39 million for free agency. To get an estimate, here are the full ZiPS projections for the Hosmer contract at the time.

One thing to note is that there is a bit of a discrepancy between the zWAR (ZiPS WAR) and FanGraphs WAR figures, as we haven’t always used the exact same park and league factors for future seasons and have utilized a slightly different methodology. For next year’s ZiPS, I hope to report both zWAR and fWAR to reduce this occasional confusion. But for right now, I’m still figuring out how to not break FanGraphs.

If 2017 were a projection system in itself, there’d be absolutely nothing wrong with the contract. It wouldn’t just be breakeven, but probably a good one. But if 2016 were a projection system unto itself, the results would be poor — and, entering 2018, Hosmer hadn’t show any ability to predictably churn out four-win seasons, so the projections overall were less sanguine. Nor was ZiPS an outlier here: Steamer’s year-one projection for Eric Hosmer was only seven points of OPS more optimistic, which is likely in the margin of methodological error.

Outside of any disagreement about player evaluation, the structure of the contract was also problematic, at least from the point of view of the team signing Hosmer. An opt-out is essentially an option, which typically favors the party that gets to make the decision to exercise, which is why players have requested team options to sweeten the deal for the player exactly zero times ever. As long as Hosmer is making reasonable decisions, “getting out” of the deal three years later doesn’t do anything to help the Padres. Generally speaking, the contract would be five years when it would advantage the Padres for it to be eight and be eight years when the Padres would prefer five. Did anyone really believe that Alex Rodriguez’s opt-out favored the Yankees?

Evaluating the cost of these options is a bit tricky. In the ZiPS scenario, I instructed ZiPS, in each simulation, to set the probability of Hosmer opting out as the probability (at the time) that he could get a better deal for the last three years of the contract. I then rescaled it from 20% to 80%, since there’s an element of guesstimation here, with the data on exercises of opt-outs in baseball being very limited at this point. In the end, a five-year, $105 million contract with three player options for three years and $39 million that had to be exercised before the 2023 season had the same surplus value as a straight eight-year, $166 million contract had.

So, how did it work? The very first returns from Dixville Notch were promising. As of end-of-day May 6th, Hosmer having hit three homers in his previous five games, his seasonal line stood at a more-than-healthy .297/.403/.534. In the games since then, however, a hair under two-thirds of the games this season, he’s hit only 235/285/339, with only four home runs, enough to drop his seasonal line to just a skosh above replacement level, at 0.1 WAR.

Much ink has already been spilled on Hosmer’s ground-ball tendencies, which have been a regular feature over the course of his career. It hasn’t prevented him from having good seasons, after all, and he managed to hit 25 home runs in 2017 and 24 in 2016 despite all the grounders. In 2018, however, this part of his profile has become even more pronounced. Since 2011, Hosmer ranks 24th in GB/FB ratio among the 450 players with at least 1000 plate appearances. In 2018, he ranks second, at 3.56, second only to Jon Jay. With an additional half-season of replacement-level play in the books, Hosmer’s projections take a dive, especially in the near-future years. Simply put, it’s data that suggests that 2017 isn’t going to be his baseline level of play.

The exaggeration of Hosmer’s already pronounced ground-ball tendencies is also apparent in the more advanced Statcast data. In 2015, his average launch angle was 5.4 degrees; in the years since then, that figure has gone to 3.6, 3.8, and (this year) -0.7 degrees. That last number places him 393rd out of the 396 players who have seen at least 300 pitches this season. His launch-angle charts have always reflected a ground-ball-heavy player, as my colleague Travis Sawchik discussed extensively back in March. Probably the largest change in how Hosmer has hit is reflected in how he’s dealing with offspeed pitches in 2018. The difference from 2017 isn’t hard to spy:

There are a lot of changeups that Hosmer was hitting well last year that he’s simply driving into the ground this year, mostly for outs. The 2018 version of Eric Hosmer has been Willie Mays Hayes without the speed. (I will not explain that reference, you should know that one!) If Hayes’s motto was “hit like Mays, run like Hayes,” then Hosmer’s right now may be “hit like Dee Gordon, run like Commissioner Gordon.”

Is it time for the Padres to panic about Hosmer? That’s definitely a bit too far for me, but there has to be serious cause for concern with this funk in which Hosmer finds himself. It may be unreasonable for him to become some deity of launch angles or anything, but I have to believe that, at this point, some change in his approach and swing could have an effect; despite his longstanding reputation as line-drive hitter, he’s only been about league-average at hitting liners over the course of his career. I still think there’s significant upside left in Hosmer, because he’s had some excellent seasons even with this approach at the plate. If he can find the right hitting guru with whom to gel, it wouldn’t shock me if he became a 30- to 35-home-run hitter and gave the Padres something closer to what Adrian Gonzalez gave them years ago.

Whether Hosmer bounces back or not, it’s in San Diego’s interest to explore all options in turning the Hosmer ship around. With a $144 million commitment if Hosmer continues to struggle, the Padres are in the Eric Hosmer business and, as most businesses can tell you, they either grow or they die.