Over the last three seasons, DJ LeMahieu’s 9.9 WAR ranks fifth among all free agents. Just ahead of him is George Springer (11.3), Trevor Bauer (11.6), and J.T. Realmuto (12.3), who comes in second place. But topping all current free agents over the past three seasons is Marcus Semien, with 12.6 WAR. Semien isn’t some older free agent, either; at 30, he’s younger than Springer and LeMahieu and within a handful of months of Bauer and Realmuto. If we looked at just the past two seasons, the production gap between Semien and his peers is even greater. But for a whole host of reasons, Semien doesn’t look like he’ll come close to earning as much as those peers in free agency, and that could make him the biggest bargain.
When examining the various free agent lists, it becomes clear that my fourth-place rank when it comes to Semien is the high-water mark. ESPN’s Keith Law and Kiley McDaniel have him sixth. MLB.com’s Anthony Castrovince placed him seventh, CBS Sports’ R.J. Anderson put him 11th, and MLB Trade Rumors slotted him all the way down at 15th. I either need to justify my ranking with a little more explanation or admit that I might have had Semien too high. This article will probably be one of those things.
Let’s start with the negatives. Semien wasn’t a particularly good player until 2018 and has had exactly one above-average hitting season in his career (’19), and his production in ’20 wasn’t very good. Add in questions about his defense at shortstop and a very crowded market at the position over the next few seasons, and we have a recipe for someone whose value certainly seems down. Semien’s market was helped slightly by the lack of a qualifying offer from Oakland, but compared to the other factors working against him, it’s a small gain. It isn’t really production or age that is keeping Semien from landing a big contract, though; it’s uncertainty. The error bars surrounding his performance would seem to be significantly higher than many other players due to defensive questions and the truncated 2020 campaign.
On defense, Semien’s tools have never been outstanding, and his misplays early in his career made him one of the worst defensive shortstops in baseball. He worked to get better, and cutting down on those misplays saw him go from awful to passable, with continued improvement turning him into a good shortstop, though there’s some concern that some of that improvement was more thanks to the presence of Matt Chapman and Matt Olson to help with range and scooping throws, respectively. Here’s what Semien’s numbers look like via various defensive metrics over the past few years:
So what do we make of these? If we believe that it takes about three years worth of defensive numbers to understand a player’s true talent, it’s going to be hard to glean much information from 2020 alone, with under 40% of a full season available. It’s like looking at a player’s wRC+ after three or four weeks and making judgments about hitting talent. It’s also worth taking a quick look at Semien’s Statcast numbers from this year and the past few seasons. Outs Above Average uses expected success rate on plays to Semien as well as actual success rate. On the latter, he is in the top quarter of all shortstops with 88%, turning a lot of balls into outs, which partly explains why he is so highly regarded by UZR and DRS. In terms of expected success rate, Semien’s 89% is second among all shortstop since 2017, thus giving him a below-average OAA figure. Statcast simply expects Semien to get to a lot more balls to the third base side.
Semien’s normal third base partner, Chapman, plays deep and is fantastic at getting to balls in the hole towards shortstop. In a somewhat odd combination related to Semien, Chapman’s expected success rate at third base is just 86%, one of the lowest at the position. I’m not sure exactly what to make of this situation, though the A’s engaged in far more shifting while Chapman was out during part of 2020, with Semien playing more to the shortstop/third base hole as well as on the first base side of the diamond, which was a bit different from ’18 and ’19, when his defensive metrics were very good. Semien is well known for having good makeup and work ethic, but sometimes progress takes time, and excess shifting and playing more in the hole might have proved difficult this past year.
Exactly how good Semien is on defense is up for debate, and that creates uncertainty on his future value and role. For example, Steamer and ZiPS have relatively close offensive projections for Semien next season, but Steamer puts him four runs below average on defense, leading to a 2.9 projected WAR; ZiPS has him at 4.2 WAR with him three runs above average defensively and boasting a slightly higher offensive projection. That’s the difference between having the eighth-best projected WAR for a free agent and potentially the best one.
The uncertainty on defense carries over to offense as well, given his career prior to 2019 and his batting line in ’20. As mentioned in our rankings, Semien’s 2020 line suffered because of a two-week slump at the beginning of the season, and because his playoff numbers aren’t included in his final line. An extra 30 plate appearances after 600 or so isn’t going to make a huge difference on a final line, generally; for example, if an average hitter went nuts in the playoffs and posted a 200 wRC+, it might bump his overall wRC+ up by five points or so. For Semien, his seven playoff games would have upped his wRC+ by 15 points to 106. If instead of a 91 wRC+ and 1.2-WAR season (3.2 WAR pace), Semien had a 106 wRC+ and 1.8 WAR (4.8 WAR pace), we might be looking at the context of his free agency in an entirely different manner.
Similarly, a slow start had an oversized effect on the overall numbers. After 14 games, Semien had a 24 wRC+ and -0.2 WAR. Over his next 46 games, he put up a 133 wRC+ and 2.0 WAR — nearly identical to his great 2019 performance. His Statcast numbers back that up, with a .347 xwOBA after his first two weeks. In a normal season, maybe Semien has a rough couple weeks but gets hot in the middle of June and, with 100 games to go, is looking at a 4–5 WAR season heading to free agency instead of reaching it as the three-win player the market might treat him as.
To try to find some context for Semien, I looked back at similar shortstops over their age-28 and 29 seasons since 1980; the five closest were Derek Jeter, Barry Larkin, Jimmy Rollins, Brandon Crawford, and José Reyes. All of them were average to significantly better than average in their age-30 and age-31 seasons (an average of 6.6 WAR for both seasons). Crawford and Reyes quickly fell off after that, but Jeter, Rollins, and Larkin kept playing well, and the group as a whole averaged three wins per season per player from ages 32–34. Larkin and Jeter had far better careers prior to turning 28 than Semien did, so the group isn’t an exact match, but regardless, good shortstops with production like his continued to produce.
Semien is not without his flaws, but he might be looking at a $100 million contract if he had been able to play a full 2020 campaign. There’s considerable uncertainty with his defense, his lone great year, and the shortened season, but signing Semien for around $50 million could be a huge bargain. The market is saturated with shortstops this year and next, but Semien shouldn’t fall between the cracks.