Last season was a lost year for Starlin Castro. While many were optimistic about his ability to take a step forward (myself included), quite the opposite happened. Castro got worse, appreciably, in essentially every way.
To give you some idea of how bad Castro's season was, it was BELOW baseball prospectus' bottom 10% projection for him last season – by a lot. Just extrapolating the numbers, it appeared there was around a 1.5-2% chance that Castro would have "reached" his projection.
There are a few explanations that were given this year as to why Castro struggled so mightily. I'm going to dive into this profile this year and try to figure some out.
The first thing I wanted to look at was the type of pitches that he was seeing this year. For this, I used Brooksbaseball. I've had arguments in the past about the relative merits of pitch types for hitters, but I can at least agree that there are 3 general types of pitches: Hard, Breaking, and Offspeed. Out of these, it jumps out to me that there is essentially no change at all in the profile, save a 3.5% shift from breaking stuff to hard stuff.
Next, I looked at swing percentages. If these numbers were down, then there would certainly be some truth to the idea that he was hesitating. Lo and behold, he was swinging MUCH, MUCH less this year than last. He offered at 1.3% fewer hard pitches, 2.4% fewer breaking pitches, and over 10% fewer offspeed offerings! Even more interestingly, if you break it down by month, you'll see that he essentially swings at the same number of fastballs all year (save an outlier in April), and the relative same with breaking pitches (though those tend up from 45 to 54% as the year goes on), but his offspeed offer percentage is way down in July, and way up in August. Those aren't particularly meaningful (Castro never saw more than 46 offspeed pitches in a month), but it does sort of illustrate an evolving approach as the season wore on.
The main culprit is the Whiffs per Swing, though. Before 2012, Castro averaged around 11.3% whiffs per swing on fastballs. In 2012, that number climbed to 13.6%. In 2013, though, it exploded to 16.2%. The problem wasn't just that Castro was taking to many strikes, he was not hitting the balls he normally would. Missing on 5% more fastballs meant Castro had 41 additional swinging strikes this year on fastballs alone. He also missed on 28.7% of breaking pitches, up from 24.7% the previous year. He did connect on 4% more offspeed stuff, but since the overwhelming majority of pitches that Castro saw were fastballs or breaking balls, it isn't as relevant.
Brooksbaseball has another tab, called plate approach. I'm not exactly sure what it measures, but it also indicates that Castro was slightly more passive this season on hard/breaking pitches, and majorly passive on offspeed stuff (though inline with his non-2012 numbers). Furthermore, Castro's approach got more aggressive in August (not that it helped him any, as his line before and after August 1 is roughly the same).
One bright spot, however, is that Castro DIDN'T swing at more pitches out of the zone. In fact, it appears that his rate of swinging at pitches out of the zone decreased relative to pitches inside of the zone. The problem is wholly contact-based and not pitch selection based. That's a very, very good thing.
If you take a look at Starlin's zone profile, you can a pretty stark look at how Castro's contact has eroded. In 2011, Castro missed only 4 of 113 pitches straight down the middle. In 2012, it was 9 of 95. In 2013, it was 19 of 108. That's a huge increase, and speaks to his troubles. In fact, it seems like he couldn't square up on anything up in the zone last year, even if it was a strike up and in the middle. When Castro did make contact early in his career, he was getting hits. His BABIP was .344 in 2011. In 2013, it fell to .290, and it was felt.
I'm not a professional scout, but given the fairly stark differences in whiffs per swing, and the inability to get squared up on the ball, especially high in the zone, leads me to think that there might be an issue that is beyond the simple mind games that many chalk Castro's struggles up to. I wonder if the bat speed is a little slower than it was in previous seasons.
One last thing which might indicate that Castro was unlucky this year. Last season, 13% of Castro's outs were LD%. Generally, that number is much lower (in the 8.5 to 9.5% range). That would indicate that a higher percentage of Castro's line drives were just not finding holes (that wouldn't be the case if he hit a ton more LD this year than usual, but he didn't hit any more this year than last).
I'm not trying to pretend like I have all the answers here, but I thought it would be interesting to see what the numbers said about Starlin's season. It appeared to me that he has a little pensive at the plate, but more importantly, he just had worse swings. It appears that that was true. Combine a few steps back with his approach and a few bits of bad luck, and you have one tremendously shitty season. I'm not very optimistic he can return to the player he once was, but I wouldn't be shocked if it happened. I would be shocked if he was this bad going forward. I think a luck-neutral floor of .255/.300/.360 is about right, with below-average defense at SS (though, to me, Castro more-or-less passed the look test defensively for me this year).