In my newest article for The Hardball Times, I analyze college pitcher abuse through five different lenses: per-game workloads, rest before/after outings, Pitch Smart guidelines, staff entropy, and Tommy John surgeries. In every section, it’s clear that reckless pitcher management is a running theme in NCAA baseball. You can see more here.
I published two articles on ground balls last week at The Hardball Times. The first used boosted trees to assess the impact that pitch and contextual factors have on predicting whether grounders will be hit. The second took a closer look at how pitching inside hardly scares batters into hitting more grounders, even though the Pirates are strong believers and have been a great GB team in the past three full seasons.
The second part of my cold-weather series at THT went up today, and so here are the companion charts of each curve’s smoothed 95% confidence bands. Check out yesterday’s blog post for a few notes on how to evaluate each graph.
Region of Origin
This is the appendix to my article today at The Hardball Times. Below are the smoothed 95% confidence bands of every individual curve shown in the charts you see at THT. These charts give you an idea of the considerable uncertainty found at five-degree intervals of temperature. As we would expect, these confidence intervals tighten up between ~60 and ~85 degrees, reflecting the fact that most games are played within this temperature range. Every five-degree bin is most uncertain at the very cold and very warm temperatures.
I aimed to keep these charts visually similar to those at THT, but there are some differences. Mainly, you need to watch for different y-axis scales, brought on by samples that are small for some subsets. One example is the chart below, as there are only so many plate appearances taken by catchers.
Times Through the Order (TTO)
Javier Baez is back in the majors to try to meet his own expectations as well as those of fans and prospect junkies. Last year, he struggled mightily, flashing his huge power potential but posting the worst strikeout rate in history. So since winter ball and spring training, Baez had been working to revamp his approach and mechanics. Changes are easy to see; compare this GIF of a home run last year off a righty pitcher’s slider…
…with this GIF of Baez’s only 2015 home run, which also came off a RHP’s slider:
The increased control of the 2015 version is palpable, right? A big leg kick that went back, around, and forward is gone; now, Baez’s toe just barely lifts off the ground as he shifts from a closed stance to a square stance right before the pitch enters the hitting zone. He’s more balanced now, attacking pitches with less torque but still plenty of it. He previously had two phases of his takeback:
But that’s been reduced to one motion, and a quieter one at that:
The bat used to be nearly parallel to the ground with the pitch having completed most of its journey to home plate. From that point in the swing, Baez had just 8 frames in the GIF until he made contact for his 2014 homer. That positioning would require the bat to complete a long journey in a rapid timeframe to enter the hitting zone.
In the 2015 swing, the bat is tilted slightly inward towards the pitcher at an earlier point. So the bat has more time (11 frames away from contact in the image below) to travel a shorter distance.
These seem like a good improvements, but the 2015 swing isn’t representative of all the cuts I’ve seen in ~20 Baez plate appearances this year. Think of the 2014 and 2015 home run motions as endpoints on a range of swings that Baez often falls between. Sometimes he sticks with the simplified mechanics throughout his plate appearances, and other times he’s used them as a two-strike approach. The big leg kick still makes appearances as well, as he used it and less-exaggerated hand movement in Pittsburgh on Tuesday.
And earlier in the month, he reverted and took a former teammate’s advice as literally as possible:
One could argue that Baez should get a pass on that last swing, since it came on a 99 mph Cholula from Aroldis Chapman. But I argue the opposite—this swing reflects particularly poorly on Baez. In a crucial matchup against baseball’s best closer, Baez took a non-competitive swing when he had the count (3-1) in his favor. The 22-year-old’s erratic and overly-aggressive approach was on full display in the Chapman PA. Clearly, there are more adjustments to be made.
Where does this leave Baez? The statistical improvement was there for him this year, but this multi-swing approach would seem difficult to maintain moving forward. Few MLB hitters (let alone successful ones) have hitting mechanics that can change so drastically from swing to swing. Baez is young enough to figure it out, but he’s very much a work in progress.
My newest piece for The Hardball Times went up earlier this week. I assess whether the rockpile at Angel Stadium impairs batters’ plate discipline when it acts as the backdrop for pitches.
Kevin Siegrist pitched a scoreless 8th inning in last night’s Mets-Cardinals game, in spite of the fact that he was tipping his pitches. Look at the GIF below, which takes freeze-frames of Siegrist’s release on a fastball and a changeup (the first two pitches he threw to Wilmer Flores). The fastball comes out of Siegrist’s hand at the higher arm slot, while the changeup is thrown at the lower angle.
These slot differences persisted throughout the game and the first half of this season. The next plot breaks up all of Siegrist’s release points by pitch type, using Brooks Baseball’s bias-corrected data. Typically, changeups have come out of his hand nearly a half-foot lower and a half-foot farther rightward than fastballs.
None of the usual pitch-tipping hoopla has surrounded Siegrist, and that’s because no one is looking to explain any struggles. After a rocky 2014 season, the lefty has been excellent in 2015. His 1.5 RA9-WAR ties him with Aroldis Chapman for 10th-best among MLB relievers. At a micro level, he’s getting plenty of swings and misses, posting strong whiff rates of 25.1% and 31.3% on the fastball and changeup, respectively. So even though he tips, and even though his fastball-heavy arsenal is predictable before he even steps on a mound, he’s had no issues.
Why haven’t batters exploited this information to a greater extent? Several possible reasons jump out. For one, a difference in release point doesn’t leave batters with much time to process this intel—at least, not like when a pitcher shows a pattern in his glove positioning or head tilt before beginning his throwing motion. Release point differences are even tougher to gather from Siegrist, since the 6’5” pitcher uses his long levers to hide the ball and create deception. Batters, then, lose split seconds in which they could have picked up those different arm slots.
Then there’s the obvious fact that his pitches are tough to hit. Siegrist has a big-time fastball, even before the above factors help it to play up. At an average of 93.8 MPH, he throws harder than all but a small group of southpaws.
Pitch tipping doesn’t always portend the automatic doom that gets hyped by announcers and the media. As with anything, the pros can outweigh the cons and mask lurking issues—Siegrist’s large margin of error seems like the key here. But the lefty’s pitch tipping is worth keeping an eye on, because as he evolves (and declines), these tells might engender major struggles and force him to make another round of mechanical changes.