Carlos Zambrano and the cut fastball
by Harry PavlidisFebruary 19, 2010
Every once in a while a pitcher actually talks in detail about his stuff. Thursday morning, Cubs.com beat writer Carrie Muskat shared this juicy detail via her Twitter account:
Not only has #cubs Carlos Zambrano lost 15 pounds, but he also is dropping his cutter. Says that's the one pitch he couldn't control wellWell, that just about knocked me over. My good friends at Another Cubs Blog suggested, via Twitter, that Zambrano may be going back to his old (pre-2006) ways—relying more heavily on his sinking fastball and putting a little extra on his slider more often. Which made me feel good, since I had just speculated along those lines in my own tweet (seriously, Twitter is great for baseball fans, if you haven't noticed).
Now that I've caught you up on the latest from the world of social media, let's talk about Zambrano's stuff. I'll explain why this news was a surprise to me.
Zambrano's seven pitches
Seven pitches, that's a lot. A fastball (four-seam), sinker (two-seam), splitter, slider, cutter and a few slow overhand curveballs and even a straight change here or there. Take away one, and discount the rare birds, Zambrano is already calling himself a four-pitch pitcher. More from Muskat, this time on her blog:
The 28-year-old pitcher says his repertoire will now include a sinker, slider, split and "high fastball."That fourth pitch is almost certainly a four-seam fastball, which Zambrano likes to get up in, and out of, the zone. And, yes, he's still just 28, but 2010 will be his age 29 season.
Going back to 2007, here are the seven pitches thrown by Zambrano, from most to least thrown.
| Pitch | MPH | Number |
| Sinker | 91.8 | 2116 |
| Fastball | 92.2 | 1858 |
| Cutter | 91.1 | 1352 |
| Splitter | 85.1 | 850 |
| Slider | 82.4 | 755 |
| Curveball | 67.9 | 39 |
| Change-up | 82.5 | 16 |
Dropping the cutter is no small deal. I've enjoyed his ability to work a fastball three ways, but I guess I can still cling to his sinker.
Using some spin deflection charts (deviation from path of a ball only under the influence of gravity, catcher's view, in inches) lets me illustrate those pitches. It's also Olympics season, so I'll use rings instead of colored dots when appropriate.
First, the average pitch (2007-2009) for Zambrano. The color key is used again in the other charts, so make a mental note. The average pitch speed is shown within the bubble.
Now the rings of Zambrano for 2009 only.
Yes, that's a lot of overlap. Let's fill those in with the actual data points to really blur the lines.
Before you conclude that the cutter is not correctly split from the fastball, these pitches are classified game-by-game (sometimes by inning or at-bat). Here are those same rings (unfilled and filled) for Zambrano's Aug. 25, 2009, start.
But, as the data points show, there are places on the margins that could be adjusted.
What's not shown is the 2 mph difference in speed as you cross the border from fastball to cutter. Still small enough to be a data error and/or reclassified.
Run values and this Cub fan's paranoia
Now for the surprise, or at least my surprise at the news. Based on my pitch classifications and my calculations of linear weights and pitch-by-pitch run values, Zambrano's cutter is arguably his best pitch. It's clearly the best of his fastballs. The sinking and "high" versions are easily outperformed by the cut fastball.
rv100 is based on the actual outcome (ball, strike, single, out, etc.) while rv100E uses balls and strikes along with batted-ball type—as in line drive, fly ball, ground ball and pop-up (as designated by MLBAM's Gameday stringers). Each batted-ball type is given a value based on the league average for that type—a weighted average of run values for each outcome (a little single, a little out, a little home run, etc.). It overcontrols for variance in parks and fielders, but offers a comparison point (or regression target) for rv100. For both metrics, negative values indicate runs saved compared to an average pitch.
| 2007 | 2008 | 2009 | Total | |||||
| Pitch | rv100 | rv100E | rv100 | rv100E | rv100 | rv100E | rv100 | rv100E |
| Sinker | 0.51 | 0.06 | -0.66 | -0.46 | -0.36 | 0.03 | -0.36 | -0.20 |
| Fastball | -0.34 | 0.30 | -0.19 | 0.73 | 0.14 | -0.09 | -0.08 | 0.27 |
| Cutter | -1.21 | -0.91 | -1.38 | -0.75 | -1.14 | -0.70 | -1.26 | -0.78 |
| Splitter | -2.87 | -0.11 | -0.58 | 0.61 | -0.99 | -0.07 | -1.19 | 0.17 |
| Slider | -1.93 | -0.71 | -0.24 | -0.75 | -0.44 | -0.77 | -0.65 | -0.75 |
Unless throwing a cutter is somehow too demanding physically—I do not know if the pitch would bother his shoulder or forearm, which is also a pain point for Zambrano—I really don't understand why he's not throwing it anymore. He did offer a clue about control, which provides the direction for the next step in this study. Meanwhile, someone who knows more about pitching injuries and bio-mechanics can opine on the purely speculative theory on arm strain. Or maybe a beat writer can ask a few questions.
References and Resources
PITCHf/x data from MLBAM and Sportvision. Pitch classifications by the author.
Harry Pavlidis admits he has a baseball problem. He is a member of Complete Game Consulting and has his own blog, Cubs f/x. Feedback, questions and comments are appreciated - Email harrypav@gmail.com and Twitter @harrypav
Wonderful research Harry. I find it interesting that Zambrano’s RV100 on the cutter is always lower (better) than his RV100E. If his problem with it is control you would think it would be the other way around, that he would be being hit better than the league average on “mistake” pitches.