One of the many stories coming into spring training from a Mariner perspective was that Blake Beavan had adjusted his delivery. From everything I had heard, he was standing more upright, using his massive size and frame more. Reportedly, he looked pretty good in his first Spring Training outing, and I didn’t see it, but we do have Pitch F/X data from the outing. The only way to quantify the new delivery using the data, at least as far as I could think of, was by release point data. So I grabbed the chart from the spring training game against the Padres, and then his last outing in 2012. The spring training outing is on the left.
As far as height and consistency, it is about the same, right around 6 feet and inconsistent. However, the data shows that he released the ball closer to his body against the Padres in Spring Training. What does this mean? It is hard to tell at this point. Delivery repetition and throwing strikes is not Beavan’s problem, and he doesn’t have real platoon splits (if you look at DIPs, there is a big BABIP gap, but that is probably not predictive), so releasing the ball closer to his body in order to get lefties out doesn’t seem to be something that Beavan should be concerned with. He also has been pretty healthy, so it is not like this is an attempt to have him repeat his delivery more to stay healthy. It is very likely that this is pretty meaningless. It does appear that there is a difference, but unless it is giving him more movement on his pitches (it doesn’t seem so) or allowing him to throw harder (it is hard to compare big league starts to spring training starts that don’t last very long), it isn’t anything that is real meaningful.
Another spring training story has been the change of Taijuan Walker’s curveball. According to reports, he is moving to more of a spike curve. Walker pitched in the Future’s game in 2012, which was the first time he appeared in Pitch F/X data. His game against the Padres is the 2nd time we have Pitch F/X data, so we will compare the two and see if we can see a difference between the curves in the very small sample size we have.
In the Future’s game, Taijuan’s only other Pitch F/X game, his curve was 77.86 MPH (but the Kansas City park tends to be a little bit fast in MPH). It had a 5.02 inch horizontal break and -5.92 vertical movement. This most compares to Jeff Francis horizontally (Esmil Rogers if you are looking for the closest right-hander) and Franklin Morales (Phillip Humber is the closest right-hander) vertically.
The two curves he threw against the Padres in his Spring Training outing were slightly slower (76.55 MPH on average, but again, once factoring in Kansas City, this is probably just about the same) and had less horizontal break (.81 inches) but more vertical break (-7.11 inches) which is what you would expect in a spike curve. He really went from a 2 plane break curve to a curve that breaks almost exclusively down. Liam Hendricks provides the closest comparison out of qualified starting pitchers in downward movement, while his almost non-existent horizontal movement is closest to Felipe Paulino.
If you look at the 8 qualified pitchers with less than 1 inch of horizontal movement on their curve, they have an average GB % of 45.25 and 28.52 % whiff/swing %. If you look at the 8 curves with the most dramatic horizontal movement, they had an average GB % of 43.5 and average whiff/swing % of 28.9 %. So the horizontal curves were slightly better at getting misses, but were worse at getting grounders. Of course, Taijuan didn’t below to the dramatic horizontal curves, and instead belonged to the ~5 inch of horizontal movement club. The 11 pitchers that have an average horizontal movement of 5.00 to 5.20 inch movement have an average GB % of 54.27 % and whiff/swing % of 24.31 %. Obviously this is a much lower whiff percentage, but it would seem that Taijuan may be leaving a lot of grounders on the table. What about his change in vertical movement? If you look at the 12 pitchers that were from -6.00 inch vertical movement to -5.8 inch vertical movement, where the old Taijuan was, you get an average whiff/swing % of 21.95 and average GB % of 52. So, these were, just like Taijuan’s old horizontal movement, high grounder, low whiff pitches. If looking at people around Taijuan’s vertical movement (-7.00 to -7.2), the 7 pitchers have an average whiff/swing % of 22.32 and average GB % of 54.43 %. So on average, these curves with the better downward vertical movement are just better curves. The horizontal movement sort of suggests that he is trading some grounders for whiffs. Just as an anecdotal note, Taijuan was a very good groundball pitcher in A-ball in 2011, but in 2012 in AA, he was not a very good groundball pitcher at all.
I thought his curve was good before, but if this very small sample size of data is any indication, this new spike curve could be even better. So it seems that the Pitch F/X data is showing us the differences that are we reading about the two pitchers in stories, and does seem to help us quantify, or least start to interpret, their meaning.