So far this season, Dillon Gee has presented us with pretty mediocre results, sporting a 4.50 ERA through 5 starts (excluding tonight's, 5/9; I will be excluding tonight's from the rest of this post as well). However, underlying this ERA is a very nice 3.69 FIP and an even more outstanding 3.10 xFIP; he's certainly not being helped too much with a .326 BABIP against. Gee is doing three things we all would like from a pitcher: striking out more batters, walking fewer batters, and generating more ground balls. Considering the 16.7% HR/FB against Gee, we would expect his home runs allowed to decrease as well. Gee's peripherals this season all seem to indicate that much better results in terms of runs allowed should be coming.
Through 5 starts and 32.0 IP (adding the 6th start would make a difference in the coming analysis, but I don't have time to work on this after tonight), it appears that Gee may be making strides towards being more than just a serviceable starter. However, there's the all-ubiquitous small sample size warning, so we have to ask- is Gee's performance sustainable?
Gee's peripherals this year are all different from last year's. As stated, his strikeouts are up, walks down, HR/FB up, GB% up, BABIP against up, etc. However, based on hypothesis testing- I'm not really sure how applicable it is to baseball statistics, but I'll proceed regardless- the change in these statistics are all statistically insignificant. Thus, it may simply be by chance that Gee has peripherals that lead to such a nice FIP/xFIP. So, we'd then like to look at whether it could be by chance or not that these changes occurred.
Let's hark back to last year, when Gee had stunning results up until the end of June. He didn't have the gaudiest peripherals, but he seemed to be winning every start and consistently obtaining good results. However, once July and August came around, Gee started to fall apart, walking too many batters and such. Some labeled it as regression; others thought of Gee as wearing out. Considering what happened last year, could the same happen this year, that Gee is outperforming his capabilities and will ultimately regress?
Warning: this is potentially tl;dr material. If you don't enjoy sifting through junk, feel free to scroll to the bottom and where there are multiple asterisks.
Let's try to find a cause for Gee's deteriorating performance last season and see whether or not there's a reason that cause will replicate itself this year. I looked at his release point throughout 2011 to start:
We can clearly see a trend that as 2011 progressed, Gee's release point appeared to grow closer to his body (vertical release point held constant). Something seems to be happening here; perhaps Gee did grow more tired, but then we might expect a release point growing further away from his body as opposed to closer. Nevertheless, his release point changed progressively, meaning that his own pitching processes changed.
I tried to find if this had any impact on the nature of his pitches, although this does not imply a causal relationship. Looking at his pitch locations, an effect does seem to take shape. Here are his fastball locations for the two halves of the season (locations of other pitches did not appear to change; also, I know that I'm splitting this into two distinct time ranges when release point gradually changed, but you can get a sense of the location change anyways):
As we can see, his fastballs in the latter half of the season were often more in the center of the zone/middle of the plate as well as generally being located more poorly, very bad signs for a pitcher.
Now, here are his movement charts for his pitches, and please pay attention to the fastball movement (in green) as the 2011 season progresses.
Again, a worrisome trend appears with Gee's fastball as opposed to his other pitches: his fastball appears to grow more flat (less movement) as the season progresses, leading to potentially worse results. From the data shown, we can see that Gee's pitching did indeed change, and that some of the change in his 2011 results may not have been by pure luck/regression alone.
However, I'm not sure how large of an impact this may have had on Gee's performance. Considering that Gee was previously outperforming his peripherals in the first half of his 2011 season, I would believe that at least some component of his deteriorating performance in the second half was due to chance alone, but I would also believe that he simply pitched worse, possibly by tiring or changing his mechanics, etc.
****Gee's 2012 performance is so far, so nice. However, if the same changes that occurred last season happen again, we could expect Gee to start falling apart somewhat again.
There's one more potentially significant change, however, to what Gee's being doing this year that could make a large difference. Gee has had greater success with his slider/cutter, generating many more swinging strikes on them while throwing them for fewer balls and generating more swings outside the strike zone on them. Different data sources differ on the change of the frequency of the pitches thrown, but Gee might be throwing the cutter much more often (I don't pay super close attention to the games this season :( sorry). Nevertheless, the success of these pitches are statistically significant in their greater effectiveness, yielding a nice overall 11.5 SwStr% for Gee's work this year as opposed to 9.1 last year. The O-Swing% against him has also increased from 32.6% to 38.7%.
Working out the data in Gee's work has been quite confusing, so I apologize if some of this post isn't too clear- feel free to ask any questions in the comments. Also, it's been tough to discern exactly what's been changing for Gee.
****Ultimately, from what I've seen, we could claim that Gee is in line to be a better pitcher than we thought he could be. My conservatism doesn't have me jumping on some #2 starter optimism, but I feel that we're selling Gee short when we view him simply as a back of the rotation starter; rather, I feel like he has some more potential. If he doesn't change his pitching methods (e.g. release point) and tire out or something, I feel that his performance this season, in terms of strong peripherals, could be somewhat sustainable.
p.s. I am really tired right now so I apologize, again, if my post appears to have less clarity as it progresses.