In 2013, Tyler Chatwood had a great season for the Colorado Rockies. But it was a great season with question marks. Over 111 innings pitched, Chatwood posted a 3.15 ERA, which is desirable for a starting pitcher and rare for a Rockies starting pitcher. But doubt could be found in the numbers beside his ERA. In particular, Chatwood’s low strikeout rate (5.3 per nine innings) and high walk rate (3.3 per nine) suggested that he was going to have a hard time repeating the performance. Too many balls in play, even if the majority of them were grounders, would mean more base runners, and more base runners would mean that Chatwood’s luck would eventually run out.
That hasn’t happened so far in 2016. Chatwood has put up a 2.15 ERA in his first 37 innings in 2015. On the surface, the 26-year-old has improved. Not only that, but those numbers that sit beside his ERA have also improved, which should temper skepticism. Not only that, but Chatwood has exhibited a modified approach over his first six starts of 2016.
The clearest evidence of Chatwood’s improvement is in his strikeout and walk numbers. He’s struck out 6.5 batters per nine innings so far in 2016, which is an additional strikeout per nine innings compared to 2013. Perhaps more importantly, he’s limiting walks. His 1.9 walks per nine is about 1.5 fewer free passes per nine innings compared to 2013. In other words, he’s mitigating the possibility that batters will reach base with more strikeouts, and he's keeping them off base himself with fewer walks.
These changes make sense as explanation of improved outcomes. Indeed, they’re almost self-evident. But there are other changes regarding Chatwood’s process that might be even more illuminating. In 2013, Chatwood inhabited the ideal of the Rockies pitcher. He allowed a lot of contact, but most of the balls in play were on the ground—like Aaron Cook. Chatwood’s 58.5 percent groundball rate in 2013 was third in all of baseball among pitchers with at least 100 innings pitched. The 51.5 percent groundball rate Chatwood has posted so far in 2016 still classify him as a groundball pitcher, but that figure isn’t even in the top 20 in baseball right now.
Comparing 2013 and 2016, Chatwood’s groundball rate has declined about seven percentage points, and those points have to be attributed to other types of balls in play, either line drives or fly balls. Chatwood’s line drive rate has declined as well though. In 2013, 20.6 percent of the balls in play against Chatwood were line drives, but in 2016, that figure is 18.8 percent. That makes about two more percentage points that have to go somewhere else, and there is just one more place for them to go: fly balls.
In 2013, Chatwood had a fly ball rate of 20.9 percent. In 2016, that rate has gone up nearly nine percentage points to 29.5 percent. This is a good thing, and it might be evidence that Chatwood is not only for real, but that he is an improved, and different, pitcher than he was in 2013.
Intuition might suggest that a Rockies pitcher should not allow fly balls due to Coors Field’s thin air. However, there is no evidence that groundball rate is positively correlated with success at Coors Field. Indeed, the Rockies as a whole have regularly posted some of the best groundball rates and lowest fly ball rates in all of baseball, without translating that into on-field success. While fly balls are dangerous because they turn into home runs, they are also the type of ball in play most likely to turn into an out. They are not necessarily undesirable outcomes, even for Rockies pitchers.
Chatwood’s elevated fly ball rate does not appear to be an accident, either. It looks like it’s a product of his pitch selection. When almost 60 percent of the balls in play hit against Chatwood were groundballs in 2013, he threw his sinker 41.6 percent of the time. That makes sense—sinkers down in the zone yield grounders. Chatwood threw his four-seam fastball 30.7 percent of the time. He relied on his slider and curveball equally as breaking pitches, as Chatwood threw them 11.9 and 12.1 percent of the time, respectively.
The profile is different in 2016. Chatwood is throwing his sinker 36.1 percent of the time, which is five percentage points less than he did in 2013. Conversely, he’s throwing his four-seam fastball more, 34.3 percent of the time, almost as much as his sinker. Four-seam fastballs are more likely to generate fly balls. Additionally, the slider has become Chatwood’s choice breaking pitch. He’s thrown in 20.5 percent of the time in 2016, compared to just 4.6 percent for his curveball.
Where Chatwood is placing these pitches is also telling. Specifically, he’s living on the edges of the strike zone, and he isn’t living in one particular edge for long. In the first zone profile below, the only bright red square is in the bottom right of the graph below. That’s where his slider, which he almost exclusively throws to righties, goes. Otherwise, Chatwood isn’t giving hitters a lot over the plate to work with. And hitters are swinging at the pitches outside of the zone, which suggests that his approach is keeping them, at least, off balance. Couple this information with his low walk rate, and it looks like Chatwood is a pitcher in command of the baseball.
The 2016 version of Chatwood exhibits different pitch usage than the 2013 version. There are more four-seam fastballs and sliders, and fewer sinkers. The result has been more fly balls, and better results. The 2016 Chatwood appears to be the better pitcher. He has still thrown just 37 innings, so there is a lot of time for either his results or process to change. Additionally, as the weather warms, more of the fly balls Chatwood is allowing might end up over the fences. But as of now, Rockies fans should believe in Chatwood until he gives reason not to. He seems to have answered some of the lingering doubts from 2013, and now we’ll see if Tyler Chatwood can eliminate them entirely.