Pebble Hunting: Lies, ... (03/22)
10 comments have been left for this article. BP Comment Quick Links
| |
|
|
March 29, 2013 Pebble HuntingThe Most Interesting Major Leaguer You Probably Haven't Heard OfToday is the final weekday of the 2013 offseason. On Monday, there will be real stats, living stats that will change by the next day. On Monday, every stat will have to go through the sample-size filter, and even in the final month of the year we will be looking at some team that’s outperforming its pythagorean record, or some pitcher with good BABIP luck, and caveat everything we say with “but regression is likely and things could change.” We don’t have to do that when we’re talking about 2012, because 2012 is over. No matter how lucky or unlucky, no matter how much regression should have happened, it didn’t happen. The numbers are in the books and they’re not changing. So, on this last weekday of the 2013 offseason, let’s talk about my favorite statistical line of the 2012 season. A line that defies belief, that defies explanation, and that will never, ever, ever change. As of Sunday, that line will no longer be relevant, because we'll care mostly about what's going to happen in 2013. This line will just be a forgotten marvel affecting nobody. But today it’s still relevant, and it’s totally real. *** In 2012, 484 pitchers threw at least 20 innings. The best FIP among them belongs to Craig Kimbrel, and the second-best FIP among them belongs to Aroldis Chapman. Neither is surprising. The third-best belongs to Jake McGee, which is noteworthy but still something most of us were vaguely aware of. The fourth-best belongs to Junichi Tazawa, which still surprises some people, even five months after the last pitch of the season was thrown, but not all people. The fifth-best belongs to Aaron Loup, and maybe one in five of you all knew that. Maybe one in three of you knows what team he plays for. If I just keep doing this for 35 or 40 more players, I will have enough words to submit this piece, and then I can go buy myself a yogurt. But 520 pitchers threw at least 15 innings, and among those, Kimbrel remains the best but Aroldis Chapman drops to third. The new no. 2 becomes Tom Layne, with 25 strikeouts, three walks, and no home runs allowed in 16 ⅔ innings. That’s the sixth-best strikeout rate in baseball, the fourth-best K/BB ratio, etc. He has kept doing such things in spring training, with 15 Ks and just three walks in 10 innings (although also two homers, or two more than he allowed last season). Tom Layne, everybody. Apparently the second-best pitcher in baseball, or so. Also, he had a 6.37 ERA in 78 minor-league innings last year, about half of those innings in Double-A, where, at 27 years old, he was the second-oldest player on his team. In 2011, in 122 innings at Triple-A, he walked more batters (57) than he struck out (56). He was traded mid-season in 2012 for “future considerations.” Kevin Towers: What’ll you give me for Tom Layne? ***
*** As a minor leaguer, Layne has never not been old for his level. As a minor leaguer, he has 5.9 strikeouts per nine, and 3.7 walks per nine, over six seasons. In his first major-league outing, he struck out Brian McCann, Dan Uggla, and Tyler Pastornicky in order.
*** Layne was a starter for the bulk of his minor-league career. When he worked in relief, it wasn’t in a LOOGY role. After the Padres traded for him, and he had five lousy starts in Triple-A, they sent him to Double-A, where he had two more lousy starts. Then they moved him to the bullpen, where he pitched less than an inning per outing—suggesting LOOGY work—and thrived. In 25 ⅔ innings as a reliever, he struck out 29 with a 1.05 ERA. "He's done really well since we put him in the left-handed role," said vice president of player development and international scouting Randy Smith. "His stuff played better." He certainly looks like a LOOGY, dropping down to some extreme arm angles and working primarily off a couple sweeping breaking balls. A San Diego Union-Tribune article after his big-league debut quoted both teams' managers as crediting Layne's success to a "funky" throwing motion and two different arm slots. The truth, though, is that he doesn't throw from two different arm slots so much as four, maybe five. On Aug. 18th, he faced Brandon Belt. Belt had seen him once before that week; presumably he had seen him in Triple-A, too. The at-bat lasted nine pitches, so we can get a good sense of how Layne works against lefties. Pitch 1: Fastball or cutter, 86, release point 6.528 feet off the ground (as recorded by PITCHf/x).
About two-thirds of the time, Layne starts lefties off with a fastball. This is a fastball, but it’s also three or four miles per hour slower than his typical fastball. It might be a cutter, but one of the themes of Layne’s repertoire is how seamlessly the different types of pitches blend into each other. He throws this pitch from a pretty typical three-quarters arm slot, which is his typical release point early in an at-bat. Pitch 2: Fastball, 90, inside for a ball, release point 6.526 feet off the ground. This was a bit of an unusual pitch for Layne in this count. On 0-1 counts against lefties, Layne usually throws a breaking ball and often drops down. But here he gives Belt the same look as the first pitch. Pitch 3: Slider, 77, release point 5.857 off the ground.
Here’s Layne dropping down for the first time; it’s not nearly as far as he can/will drop down, but it does give the batter an uncomfortable angle and his slider a slower, sweeping look. Belt hits a liner toward left field, foul, but he breaks his bat and Layne breaks down his swing:
Pitch 4: Fastball, 91, release point 6.181 feet off the ground. It’s the hardest pitch Layne has thrown. His arm slot is notably lower than it was for the first two fastballs, and notably higher than his third-pitch slider. Four pitches into this plate appearance, you could argue that Belt has seen pitches from three different arm angles. Pitch 5: Slider, 84, foul, release point 6.694 feet off the ground. The movement on this pitch doesn’t resemble the previous, sweeping breaking ball at all. It’s closer to an over-the-top curve, but hard. It’s almost a fourth distinct arm slot, but close enough to the first two that we'll keep the count at three. Pitch 6: Slider, 79, release point 6.86 feet off the ground.
This is a bit of a hybrid of Pitch 3 and Pitch 5, and it comes from a new over-the-top arm slot. Against lefties, Layne threw 18 2-2 pitches this year, and 16 of them could reasonably be called sliders. But the range of these pitches, and the range of release points on them, is huge:
Pitch 7: Curve, 74, release point 5.56 feet off the ground.
And now he's sidearming it. A fifth different arm slot. PITCHf/x calls this a cutter, even though it’s the slowest pitch he has thrown all day. Pitch 8: Slider, 83, foul, release point 6.87 feet off the ground. This pitch had no sweep and hung up in the zone. Belt missed it. Pitch 9: Slider, 75, release point 5.70 feet off the ground.
Very similar pitch to the seventh, with a low arm angle, but it’s not quite as extreme—more very-low-three-quarters than sidearm. It's also a little harder. It's basically a better version of Pitch 3. To put this in perspective: the Padres' starter during this game, Eric Stults, threw 85 pitches, and 79 of them were within a seven-inch vertical range of release points. Layne would throw from a 16-inch range of release points in just eight pitches, and if they fit generally into two release points
there is still plenty of variation within those clusters. Say we set the lowest arm slot as the ground floor; he threw one pitch from two inches higher than that, another from four inches higher, another from seven inches higher, two that were from 12 inches higher, one that was from 14 inches higher, and one that was from 16 inches higher. (There is actually one pitch missing from the chart above; it's the sidearm pitch, our ground floor.) *** When Layne entered his first game with the Padres, his hometown announcers described him as a sinker/slider guy. PITCHf/x wants to credit him with a curveball. There are references out there to a cutter. For instance, after his first appearance: "Then Layne got pinch-hitter Tyler Pastornicky to chase a 77-mph cutter to end the inning." A 77-mph cutter is hard to explain from a pitcher who throws a 90-mph fastball and an 82-mph slider and a curveball that, if you believe PITCHf/x classifications, reaches into the low 80s, too. But a curve and a slider are often just two points on the same spectrum; and a slider and a cutter are often just two points on the same spectrum; and so it seems reasonable that a curve and a cutter might also be two points on the same spectrum, and that all "three" of Layne's pitches are the same pitch, performed a bit differently. If you lump all of his non-fastballs into one category, you'll see a strong correlation between arm slot and everything else. A lot of the differences that PITCHf/x is picking up in spin or velocity seem to originate with the different release points. In lieu of a bunch of pitches, Layne seems to basically have two: a fastball, and a slider, and he does a whole lot with that slider by throwing it out of his ear and down off his hip and everywhere in between. *** Or maybe he's just inconsistent. *** A month later, he saw Brandon Belt again. This time the plate appearance lasted eight pitches. Layne never dropped down; every pitch was around three-quarters or higher. Only one pitch was slower than 82 mph. I'm not sure what to draw from this, but it happened. My hypothesis is that the arm angles aren't even about deception, that the deception is really a nice side benefit. Rather, the arm angles are about making the pitch do what he wants the pitch to do, that his arm slot is his gas pedal and his steering wheel in one, and that he chooses an arm slot the same way that he and his catcher choose which pitch to throw. Belt popped out. *** On Monday, when the stats come alive again, it would be impossible and irresponsible for me to note, based on just 27 plate appearances, that a left-handed specialist has been very good against righties as well as lefties. That's a worthless number of plate appearances, predictive of nothing. But we're not talking about predictions when it comes to 2012. We only have to talk about what actually happened, and what actually happened was that Tommy Layne also shut down righties, in 27 plate appearances. They hit .240/.296/.320 against him, but more than a third of them struck out and only two of them walked, and none of them homered. Against right-handed batters he had the best FIP in baseball, better than Kimbrel even. I don't even know where to start with that one, so I won't. *** So the season starts anew for Tommy Layne on Monday. At that point, we will have to wrestle again with whether Tommy Layne is good or not, and what the Padres should do with him in their bullpen, and whether he's being overused or underused and what his future is. But for one last weekend we can just relax and say, definitely, without any question or consequence, that Tommy Layne was very, very good.
Sam Miller is an author of Baseball Prospectus. Follow @SamMillerBB
|
" My hypothesis is that the arm angles aren't even about deception, that the deception is really a nice side benefit. Rather, the arm angles are about making the pitch do what he wants the pitch to do, that his arm slot is his gas pedal and his steering wheel in one, and that he chooses an arm slot the same way that he and his catcher choose which pitch to throw."
Bingo! Any pitcher who tells you that varied arm slots is about visual deception doesn't know what he's talking about. The deception comes into play only insomuch as the hitter has no idea how the pitch will move.
I agree to an extent. I always thought a big part of a changing arm angle was to get the hitter to think and not react.