I love small sample size observations. As long as you are not counting on them being predictive and you do your best to understand why the change has occurred. In this case I was intrigued by Hector Neris’ early start to the season where he has not allowed a run, while walking only 2 and striking out 15. That is obviously not sustainable because Neris even by optimistic observation, is not one of the best relievers in the majors. However, the numbers are good enough that it is worth taking a second look.
My first start in these journeys is to look for a change in velocity. Neris’ velocity is unchanged, or at least not statistically changed. So we move on to pitch usage, and there is a big difference there.
Data from Brooks Baseball
Now Neris’ changeup and splitter are virtually the same pitch. The splitter is far and away Neris’ best pitch with a whiff percentage of 29.2% in 2015 and a whiff percentage of 38.2% to start the 2016 season. Without talking about the improvement in the performance of the splitter, we can make an assumption that throwing more splitters could improve Neris’ results if it was a good enough pitch to keep hitters off balance, but more on that later.
Before talking about the interaction between pitches, we need to go back to isolating them, so let’s look at the pitch movement. Since Neris is clearly deemphasizing the slider, we are going to ignore it for now. We see that in terms of vertical movement, Neris’ fastball is dropping about half an inch less on average and his splitter is dropping about three quarters of an inch more on average. That adds some more separation between the pitches, but not a massive amount. When we go to horizontal movement and there is both the fastball and splitter showing about 2-3 inches more of armside movement, with the splitter showing more than the fastball. Movement is good, but not necessarily the end all of any improvement. More interesting to me is why we are getting this movement.
My first stop is the release point, and here is the horizontal and vertical release points for Neris over the past three seasons.
Hector Neris Vertical Release Point – Brooks Baseball
That shows Neris’s release point dropping some, and moving towards the third base side. My immediate thought was that he had changed his spot on the pitching rubber, but that is not the case. Instead there is a slightly different change. Here is Neris last August and last night.
It is a really slight change, as you would expect for a slight drop and slight move toward third base. He has arm slightly more extended to his side, it is not really a move to side arm from over the top, but it is giving a different angle on the pitch and he is getting some extra movement. But that really is not the big thing here. To look at that we need to acknowledge that everything is connected. His splitter is good because it looks like a fastball and then comes in 10 inches below a fastball. So we need the two pitches to come out of the same spot, and for that kind of analysis I like to look at release point by game because over a season the averaging can muddle the picture. His horizontal release point between the two pitches has always trended together, but we get a different result if we look at vertical.
What we see in 2015 was a pretty significant gap in release point between his fastball and splitter. This can telegraph to a hitter which pitch is coming. In 2016, he is not perfect but the two are much close and are trending together. This allows him to not only use the movement of his pitches, but the inherent deception of two fastball like pitches coming from the same place. Here is another way of visualizing it from Fangraphs’ PITCHf/x game charts. First 2015 and then 2016.
I don’t know how much the numbers will hold up for Neris because this is not really predictive and is more descriptive. However, his ability to align his fastball and splitter from the same release point, coupled with an increase in splitter usage makes him a better pitcher then he has been in the past.