The Best and Worst of Marcus Semien by Owen Watson May 13, 2015 A cursory examination of the WAR leaderboards for shortstops in 2015 shows a few surprising developments. At the very top, we have the nice blend of defensive and offensive aptitude in Zack Cozart, and the mainly defensive-minded duo of Brandon Crawford (who has surprised at the plate this year) and Andrelton Simmons. Then, where we might expect a name like Ian Desmond or Troy Tulowitzki, we instead have Marcus Semien. It’s only May, so we’re still dealing with the usual parts of the game that suffer from small sample size issues when it comes to player comparison and valuation, like BABIP and defensive metrics. Still, there’s something about Semien being toward the top of the boards that warrants our attention: not only is the A’s shortstop there because of his offense, he’s there despite his defense. On Sunday, we saw both the best and worst of him. Let’s start with defense. During the bottom of the second, Logan Morrison hit a routine one-hopper up the middle to Semien, who was shifted along with the rest of the infield. The outcome was one Athletics fans have wearily gotten used to: That was the 24-year-old’s 9th error of the season, briefly tying him with Ian Desmond and Danny Santana for the league lead, that is until he moved into sole possession of first by making another on Monday (he made one last night too). If you don’t like errors, Semien sits at -1 runs by DRS, good for 20th in the majors. Finally, UZR/150 casts an even harsher light on him, ranking him dead last among all qualified major league shortstops. Defensive stats are easily skewed this early in the year, so we should regress, regress, then regress some more, taking care to sprinkle them with grains of salt along the way. Let’s look at both UZR/150 and DRS for shortstops this year graphically, so we have all the information (to be clear, we shouldn’t use this as a strict comparison between the metrics, more for reference): There’s really no doubt that Semien has been one of the worst defensive shortstops in the league: we can see it on a lot of routine plays and we can see it in the data, even if we assume a lot of noise. Eno wrote about Semien switching from a utility role last year to shortstop full time in 2015, as well as the general outcomes historically when such a move is made: we don’t have a ton of examples, though players who make that transition usually aren’t great defenders to begin with. Semien fits that mold. There are two main issues with Semien defensively in 2015; he’s showing poor range in the UZR data we’ve gotten back, and he’s making a lot of mistakes on easy fielding plays (he’s only completed 94% of plays classified as routine). His throws have never been terrible, and given that the UZR data is noisy at this point, focusing on those fielding errors might be the best course of action. So, the question arises: was Semien always prone to these sorts of obvious fielding mistakes, or could it simply be a bad spell? Let’s go back through some old examples. His time at third in 2014 probably gives us the best comparison, as he went through some rough spots: in 279.1 innings there, he made just 89.4% of routine plays. Here’s one from late April of last year: Tough hop, but certainly a play that should be made. Here’s another two weeks later, on May 2nd: Nothing to say about this one, really — just plain missed it. How about one more, the very next day (May 3rd): Everyone makes errors outside of one or two elite guys. And, to be fair, everything looks a lot worse when you stack GIFs of errors back to back to back. But Semien seems more prone to making glaring errors on routine plays than most fielders: you expect to see one or two of these during a full season from an everyday player, but Semien had these three in the span of two weeks last year. That brings me to the next point – the distribution of his errors. Here’s a list of when he made his errors last year (note that Semien was in AAA for almost all of June/July/August): April 7th April 10th April 22nd April 30th May 2nd May 3rd May 24th May 26th September 2nd September 9th (3) September 19th September 23rd If there’s such a thing as errors becoming contagious, Semien embodies it. They generally clustered together in tight packs, like late April/early May, when things went off the rails a little bit. This at least gives hope that his current fielding problems come and go, which is better than him just being an awful fielder every game of the season. Simply being bad at defense is way harder to fix than stretches of games where his confidence in the field might be temporarily low. One issue with trying to judge Semien’s overall defensive capabilities is that our sample sizes just aren’t large enough to really see how he’ll end up as a shortstop (if he is allowed to remain there). We know he has shown a propensity to make a lot of mistakes on easy plays no matter where he is on the infield, and whether that is an entrenched part of his game or just an issue that can be resolved still remains to be seen. Given the overall malaise of the A’s defense, it’s something they’ve been trying to work on before games. What we can be confident in is that there will likely be further growing pains as Semien continues to learn the position. His range has never been poor by UZR except for a brief stint at second base during 2014, so this year’s current negative marks could prove to be a month of examples that bucked the overall trend once we have a couple years worth of data to work with. Then there’s the other part of his game. If we look at the shortstop leaderboards, we’ll see Semien just behind Zack Cozart in wRC+, at 145. He’s sporting an ISO of .204, and, even though he’s still not walking as much as anyone would like (his walk rate has actually decreased from 8.2% last year to 6.1% this year), his slugging is currently making up for it. Simply put, he’s crushing the ball this year. Let’s go back to Sunday: in the third inning, he did this to a middle-middle first-pitch fastball from Felix Hernandez: Then, in the sixth inning, he did this to less middle, more inside fastball against the King: No one has hit two home runs off of Hernandez in the same game since 2012 (Gordon Beckham, come on down), and, even though the second barely got out, it’s was an impressive offensive day. That accomplishment should at the very least make us take note of some positive trends in Semien’s approach and execution at the plate. Let’s first start by taking a look at some select plate discipline stats between last year and this year to get an idea of his areas of improvement (Semien only played 21 games in 2013 as a September call-up, so I’m going to leave that year out): Season O-Swing% Z-Contact% Contact% SwStr% 2014 27.7% 86.2% 78.3% 8.8% 2015 23.5% 91.4% 81.7% 7.7% Lots of things stand out. Good things. His O-Swing% is down, further below league average than it was, showing he could have an improving sense of the strike zone. His Z-Contact% is way up and well above league average, so he’s hitting pitches that you want to be hitting. And finally, his Swinging Strike% is also down, confirming both of those previous figures. Before we go further, what’s slightly strange about Semien is that his swing rate has never matched up well with his high strikeout rate — in other words, he strikes out looking a lot. He struck out looking 30% of the time last year; league average was 24.6%. His rate this year is just under league average, even though his swing rate has gone down slightly, which could also point to a better understanding of the zone. That being said, his O-Swing and contact improvements have also translated into his rate stats. Semien is striking out 6% less overall than he was last year, and his hard hit rate has jumped to almost 32%, putting him 6th among qualified shortstops. The power output has also increased on the back of a 13.3% HR/FB rate, which would seem due for regression if he wasn’t now hitting his average fly ball seven feet further than he was last year (271 vs. 278). Finally, he’s running. His six steals in 35 games are already double what he produced in 64 games last year, and he hasn’t been caught since 2013. That’s the reason he currently leads all shortstops in our Ultimate Base Running metric, and it would add another level to his skill set. It’s still mid-May. We haven’t seen a full year of Semien, so we don’t even have a fully-formed idea of who he truly is. This could just be a really hot streak, and once his BABIP regresses and he cools off, he might go back to striking out 25% of the time with a little bit of pop. But he’s only 24, showing plate discipline and contact rate strides, and his minor league numbers point toward a more disciplined approach than we saw in 2014. Being relatively new to the majors, there are always going to be growing pains: we see that from Semien in the field. Whether he sticks at shortstop long term is a big question mark at this point. However, we’re also seeing what looks to be a developing offensive asset at a position that has very few of those, and that’s worth taking note of.