What’s Wrong With Jason Hammel?
Last season Jason Hammel showed glimpses of being the top of the rotation starter the Orioles have needed for years. The reality was that Hammel pitched very well, but wasn’t a top of the rotation kind of guy. He was however, a very solid performer whose season was limited due to injuries. When on the field, he was solid, pitching to a 3.43 ERA in 118 innings.
Hammel’s peripherals last season might have been even better than that ERA would suggest, posting a 8.62 K/9 and 3.20 BB/9 which led to a FIP that suggested he could be even better than the 3.43 ERA he actually posted.
This season though, is a different story. Hammel’s ERA of 4.10 isn’t terrible, and truthfully isn’t significantly worse that the mid-3 ERA he posted last year. His FIP though is 4.99, which hints at some troubling trends in some of his peripheral stats. In just 41.2 innings this year, Hammel has a 5.62 K/9 and 3.46 BB/9, with both stats going in the wrong direction. It’s also worth noting that Hammel’s HR/9 has nearly doubled since last season, rising from 0.69 HR/9 in 2012 to 1.30 HR/9 this season. This is the result of Hammel giving up 13% more fly balls this year as his HR/FB% rose only slightly.
This increase in fly balls, and subsequent drop in ground balls is, in my opinion, the crux of Hammel’s issues this season. In 2012 he posted a terrific 1.89 GB/FB ratio, which would have placed him 15th in MLB last season had he pitched enough innings to qualify. This season Hammels has posted a much worse 1.07 GB/FB ratio which would have placed him 69th last year.
Certainly, getting ground balls isn’t everything, and many pitchers are successful giving up more fly balls. Hammel however, has been successful because of his ability to limit fly balls and subsequently home runs, especially while pitching in Camden Yards last season.
So what is causing this dramatic change in Hammel’s ability to get ground balls? We’ll take a look at BrooksBaseball’s pitchF/x data to get a grasp on that. One note for this analysis is that BrooksBaseball classifies what many of us would call Hammel’s two-seamer as a sinker. BrooksBaseball profiles the pitches by movement and using comparable pitchers across the league, so it should give you an idea of how effective Hammel’s two-seamer is when it can be classified as a sinker.
Starting with the good news, Hammel has actually increased his sinker usage thus far in 2013, from 31% in 2012 to 36% this year. The bad news here is that his velocity is down on all 5 of his pitches, with a 0.93 MPH drop on his fastball and a 1.18 MPH drop on his sinker. The charts for both seasons can be seen below, starting with 2012:
The outcomes of his pitches are also interesting, in that they tell a compelling story about Hammel’s ineffectiveness thus far this season. The red boxes highlight Hammel’s inability to get swinging strikes on his pitches this season. That includes drops of 7.60% on his fastball, 6.35% on his sinker, and 14.60% on his slider. Those three pitches also happen to account for roughly 85% of the pitches he will throw in any given game. Hammel also saw a significant drop in whiffs on his curveball, but gains on his changeup from last season to this one.
The blue boxes here highlight the change in Hammel’s ability to get ground balls with the three pitches he uses most. The GB/BIP column is especially depressing when looking at the “Fourseam” and “Sinker” rows. The significant gain in ground balls for his slider is promising, but it is offset by the dropoff on the other two pitches. Specifically he has seen drops of 14.96% and 18.80% on his fastball and sinker respectively. The GB/FB column shows that he’s seeing dramatic increases in fly balls which we discussed briefly earlier in this post.
You might be thinking, “Clearly a lack of movement is causing him to get less whiffs, and fewer groundballs”. You’d be wrong. The charts below show very clearly that from last season to this one, his Vertical and Horizontal movement have stayed largely the same:
So what exactly IS going on here? The answer to that is less clear. A review of pitchf/x charts from Texas Leaguers show that Hammel is essentially following the same location patterns when throwing his fastball and sinker. It’s not as if Hammel is all of a sudden throwing the ball up in the zone all the time and getting hit hard. In fact, he’s finding the strikezone a little less than he did last year, so opposing hitters aren’t getting more pitches down the middle to hit harder. The two charts below show Hammmel’s locations on his Sinker/2-seamer, starting with 2012:
Ok you say, maybe the difference is from his splits. That is, Hammel is struggling against lefties and they are artificially deflation his numbers. Well, that doesn’t seem to be the case either. The graphs below from Fangraphs show how Hammel’s performance against righties and lefties has changed over the years. You’ll notice in the below graph that the averages for hitters from both sides are up, but it seems that they are up similar levels this year:
So if that isn’t it, maybe hitters from one side are just hitting more deep fly balls and home runs against Hammel. The graph of slugging percentage below suggests that, much like batting average, there seems to be little explanation to why Hammel is struggling. The slugging percentages for hitters on both sides are up big, but once again they seem to be rising in concert with one another:
The answer to the question posed in the title to this post is clearly a very complex one. There doesn’t seem to be a simple answer other than “He’s giving up more fly balls”. That of course makes you wonder why he is struggling to keep the ball on the ground, which we don’t have a clear and definitive answer to. Perhaps Hammel has just been unlucky when it comes to home runs. Typically we talk about pitchers being unlucky when their BABIP is extraordinarily high, but in this case it could be that he’s giving up home runs at a rate that is unlikely to be sustained throughout a full season.
Perhaps pitch sequencing is an issue for Hammel, something that will require more than 40 innings or so to determine. At this point in the season it seems that the only significant sequencing difference Hammel has made is to throw fewer sinkers to lefties on 2 strikes, and more sliders. This however is likely unreliable because of the tiny sample size thus far.
Ultimately, let’s add this to the list of things to watch for throughout the next few months. It’s entirely possible that this is just early season noise, and Hammel will settle back in to 2012 form soon. It’s also possible that there is something that HAS changed, and we just haven’t found the key to it yet.