If the Brewers are going to avoid losing 100 games this season, Wily Peralta either needs to be sent down to AAA, which is being discussed, or fix whatever the hell has happened to him. Not only does the first option mean he’ll likely have to work on his pitching issues at the lovely, but extremely hitter-friendly, Colorado Springs, but any trade value that he once had would all but evaporate. So, in an effort to push the later solution (and to ensure some catharsis) let’s do it. Let’s fix Wily Peralta. He may never have the shine of his highly-touted prospect days ( #1 Brewers prospect and #56 overall in 2011), but I’ll be damned if I just sit back and watch as he washes out as one of the worst pitchers in the majors (plus I don't want to work on my battery project for a little bit).
Wait, how bad has he been? As of May 17, 2016, the year of our Lord, Wily Peralta has the major’s worst ERA (7.30) and 3rd worst FIP (5.58) for qualified pitchers. His BABIP is really high, a major league leading .392, and normally you’d expect that to come down to major league average (.296), but he has allowed the 10th highest Hard Hit Rate (36.9%) and just tons of overall contact. It’s crazy: he can’t get anyone to swing at his pitches outside the zone (24.2% O-swing, 5th worst in MLB) and can’t get anyone to not make contact to pitches in the zone (96% Z-contact, worst in MLB). That poop cocktail has led to a awful K/9 (5.53, 6th worst), by far lowest of his career. Even when he does get ahead in the count 0-2, hitters still have an OBP of .355 (the average OBP for hitters after getting to a 0-2 count is .203). All this to say, yes, he’s been one of the worst pitchers in the majors.
To fix this problem, we must first understand the problem. The previous facts merely confirm Peralta’s been awful, but it doesn’t tell us why he’s been awful. So let’s dig deeper: Wily throws 4 pitches as defined by pitch f/x: fastball (FB), 2-seam fastball or sinker (FT), change-up (CU), and slider (SL), none of which have been effective. The value of each of those pitches can be quantified in a metric known as pitch value / 100 (normalized to times thrown), where positive values are good and negative values are the opposite of good.
As seen above, in 2014, Wily’s best season in the majors (which was still only ok, 3.53 ERA, 7 k/9, 4.11 FIP), his fastball and change-up weren’t great by this metric, .79 and -2.12 respectively, but this year they’ve plummeted to -2.62 and -7.76. For reference, this corresponds to the 8th worst fastball and 3rd worst change up in the majors. What’s sad is the CU was good last year, even though he used it only ~7% of the time. What’s changed? How did all his pitches go from not that good to absurdly awful? Well, in general, three things affect a given pitch’s effectiveness: speed, movement, location, and sequence. [There may be other things, like deception and release point, but for any specific pitcher, these things should not change much]. I don’t know how to think about pitch sequencing, let alone know where to acquire data on it, so we’ll have to ignore that aspect for now. Instead we’ll focus on the other ones, starting with speed.
Release velocities for Wily’s 4 pitches are plotted for the past 4 years above. [Note that release velocities are slightly different than measured pitch speeds on referenced websites] His fastball release velocity (both FT and FB) has decreased the past 2 years (black), from 96.5 mph in 2014 to 95.1 mph this year. That’s not a good sign, and neither is the increase in velocity of his change-up (85 mph in 2014 to 86.16 mph in 2016). People, I’m not entirely sure whom, say that effective off-speed pitches should be 10-15 mph below the fastball velocity, so an increase in change-up velocity coupled with decrease in fastball velocity makes both pitches less effective. His other pitch, the slider, seems to have also gained velocity, which is actually a good thing, if the movement is maintained.
Speaking of movement, let’s take a look: Below are the vertical and horizontal movements of each of his pitches since 2013. Note that negative values on the horizontal graph indicate movement towards a right-handed hitter. And just a quick aside on what makes a ball move: The spin from a pitch forces the seams of the baseball to interact with the air and create asymmetry in the pressure surrounding the ball. The ball is then pushed towards the area of lower pressure. The air viscosity and pressure affect the magnitude of this effect, but so does the spin rate and velocity of the ball. The faster the ball travels, the higher the spin rate must be in order to achieve the same amount of movement.
First thing to note is that Peralta’s slider is really more of a hard curve, since it has mostly vertical action with some horizontal movement towards lefty batters. Also note that the vertical movement on all of his non-breaking pitches trend as you may expect relative to their velocity: both FB and sinker have gained vertical movement because they have lost velocity, thus have more time to drop due to the downward pull of gravity. The CH has lost some of that drop in 2016, since it’s being thrown harder. [Side note: While not quantitative, it’s an interesting observation that the uncertainty in the change up movement seems larger than any other pitch. While harder to see, there is a corresponding large uncertainty in the CU velocity and horizontal movement. Inconsistency of this kind could lead to worse performance of the CU, but I don’t have the time to relate these things quantitatively right now.] The slider also seems to track with velocity, accounting for the increase in movement in 2015, then restoring to where it is currently.
Looking at the horizontal movement, nothing really stands out that would indicate bad performance. His SL moves slightly towards a LHB, while everything else moves towards righties. The increase in horizontal movement for the FB and may be due to the lowering of velocity for these pitches, but the CU’s increase cannot be since it has increased in velocity over the years. So what has caused this increase in movement and why isn’t it helping?
To help answer the first part of that question, let’s look at Wily’s release point in both directions for all his pitches:
Clearly, looking at reduced vertical position and the extended horizontal position, he has been dropping his arm slot slightly, either intentionally or not, which may answer why his stuff has trended to more horizontal movement the past few years. It may also explain the loss of velocity, but that may just be what happens to people when they age. What’s more concerning is the difference in release point for the different pitches. Ideally, all pitches should come from the exact same spot, so the hitter is not clued in on as to what the pitch is. In 2014, his best year, the vertical release point for all his pitches are clustered nicely in the same < .8 inch window. Comparing that to 2016, there is an unseemly difference in vertical position of nearly 1.5 inches between the CH and the other pitches. This may explain how his CU became one of the worst pitches in baseball. More than that, it may help eliminate the hitter’s worry that a CU is being thrown when a pitch is coming from the higher position making his FB and FT less effective as well.
Could this change in release point also affect the final location of pitches? Let’s take a look:
Above are heat maps of called balls for all pitches in 2014 and 2016 from the catcher’s perspective. In his best year, Wily missed most often down, and sometimes away to RHBs. This year, he’s missing much more horizontally. With his release position changing, leading to more horizontal movement toward RHB, he could be over compensating.
But it’s not just where he’s missing, it’s by how much. Below are zoomed in heat maps of called balls on his FB. He’s missing that outside edge on called balls by about 3 inches more than he did in 2014. This has made it easy for hitters to lay off, has the swinging % on FB outside the zone has gone from 33% in 2014 to a measly 18% this year.
Final locations for all sliders and sinkers thrown by Peralta in 2014 and 2016 are below. Several things stand out: first, the slider location has been bi-modal, with some pitches ending up in the corner like they did in 2014, and some that catch a lot of the plate. My guess is that the miss in location in the zone has been the leading cause of its 97.7% contact rate within the zone, up from 79.6% in 2014. Second, the sinker is also getting way too much of the plate, causing a 22% fly ball rate in 2016, up from 15% in 2014. So yeah, he’s either missing way too much or getting too much of the plate, a really bad combination, unless your pitching in the home run derby.
Ok, what have we learned? What’s been going on with Wily? And can we help him correct whatever it is that is broken with him? Well, in order, here’s what we got:
1) Velocity drop with fastballs coupled with velocity increase with the change up has hurt the effectiveness of both pitches.
2) Wily’s been lowering his arm slot for a few years, leading to increased movement on his pitches.
3) There is a large difference in where he releases his change-up compared to his other pitches, which is not ideal.
4) Wily’s missing the zone more dramatically this year with his fastball, leading to low swing rates outside of the zone.
5) He’s also getting too much of the plate with his other pitches, leading to a high rates and quality of contact.
So what should he do about these things?
1) Not much can be done about the dipping velocity, unless it is cause by the changing arm angle.
2,3) Wily needs to work on throwing the change-up from the same position as his other pitches. Not only will that help with deception, but it may bring the change-up’s speed back down.
4,5) The lack of control is probably due to the change in release positions. He has not been able to compensate for the slow drift of his arm angle down, leading to a loss of control. He needs to either realize that this is happening and return his release point back up or eventually find the feel for this new release point and increased movement.
If it true is his release point that is leading to most of these issues, then he is fixable and can still regain some of his value to both the Brewers and any team that may want starting pitching this summer. If not, he may be yet another pitcher who falls off the face of the earth.