Feierabend’s Learning Curve
When you don’t have time to write 10,000 words, draw a picture. Here’s today’s:
It’s the same pitch-by-pitch graph I did for Morrow and RR-S over the weekend, but this time, it’s comparing Ryan Feierabend’s start yesterday with Jarrod Washburn’s last start in Cleveland. I think the graph helps illustrate one of the main things that Feierabend is going to have to work on: keeping hitters off balance, and this won’t show up in average velocity readings.
Washburn and Feierabend have nearly identical average velocities on most of their pitches – they’re both 85-90 with the fastball, 75-80 with both their breaking ball and change-ups. However, as you can see in the graphs above, Feierabend is much, much more consistent with his velocities – there’s basically nothing in that 80-85 range all game. Washburn, on the other hand, has a much greater variety, with a far more random scattering of pitches than Feierabend’s fast-or-slow approach.
This is basically what people mean when they talk about keeping hitters off balance. It’s not just about mixing your pitches (though that’s important too), but mixing velocities within your pitches. This is what Feierabend needs to get better at.
I can’t believe I’m saying this, but Feierabend needs to learn from Jarrod Washburn. Washburn really represents his upside – a pitch to contact flyball lefty who looks okay in front of good defenses and in a generous home park – so emulating Jarrod isn’t a bad idea for him. With a little bit of work, Feierabend could basically replicate what the Mariners think they’ll get from Washburn next year, and hopefully, if he improves the final month, the team will realize that Wash is simply not worth $10 million to this organization when he’s so easily replaced.
Comments
21 Responses to “Feierabend’s Learning Curve”
Leave a Reply
You must be logged in to post a comment.
Another beautiful post, Dave.
Any idea what Chuck Armstrong sees when he sees these graphs? “Ooh, pretty blue”.
But Ryan’s graph is so pretty to look at.
Now if Ryan does begin to pitch a bit like Washburn do we then have to sign him to a 10mil/season deal?
But Feierabend is consistent! Isn’t consistency something we look for in pitchers, only overshadowed by “grit” and “veteraness”?
Feierabend said in his after-game comments that he was throwing his fastball “only for effect” and was using his breaking stuff to get guys out. How much would he benefit from some tutoring by Moyer?
This is what could be:
Felix
Bedard
Morrow
RRS
Feierabend
Total cost would be something like 10 million right, mostly for Bedard?
Instead we have Silva(12 million), Washburn(10 million), Batista(8 million).
I just don’t get how a business could make such awful business decisions.
Chuck Armstrong doesn’t read blogs. The internet is hard for him.
Have you seen the business world lately?
(Yay! Buttons are back!)
Chuck Armstrong prefers pie charts over graphs, I believe.
Pie charts are tasty.
The “easily replaced” point is the key. Whatever the difference is between Feierabend and Washburn in terms of pitching ability, it’s not $10M. But how much more valuable is Washburn because of his ability to change speeds? A win or two? Not even?
By most (decent) statistical measures, Washburn’s been worse than Feierabend this year. By tRA, Feier wins 5.14 to Wash’s 5.42. By xFIP, it’s 4.48 to 5.04.
Are you sure that Washburn’s clever deployment of an 82 MPH fastball actually works? A quick look at his pitch fx card actually supports that hypothesis, but there’s still some evidence out there saying that Feierabend is a better pitcher *right now*.
How about an Armstrong/Lincoln Learning Curve graph?
Now that would be … interesting, to say the least.
mln…the line would be a flat line.
Those three have nothing more to learn, and thus probably have lost any desire to try to learn.
Marc’s right in a sense – this is just my personal theory, and I can’t show a huge mound of evidence that defends my position. I might be totally wrong.
However, I do think we’ve seen enough pitchers succeed for long periods of time with highly questionable stuff to realize that there’s certainly an element of game theory that can overcome a lack of pure talent. Moyer is the obvious example that is always cited, but he’s definitely not the only one – we’ve seen a lot of pitchers get hitters out simply by changing speeds and throwing strikes.
I think Feierabend would benefit from being less predictable. Right now, a hitter can basically focus on three narrow boxes that differentiate the velocity and movement of a pitch that leaves Feierabend’s hand – adding in a few more boxes to make their decision harder seems unlikely to be a bad thing.
Well, I wonder how many boxes you need, though? We’re talking about boxes in time here, not space, and they’re pretty tiny. Google tells me that Feierabend’s range of speeds is from roughly 100 ft/s to 132 ft/s. Over 60.5 feet, that’s 0.6 to 0.46 seconds, or roughly a difference of .147 seconds. Humans can perceive things much quicker, but in terms of actual reaction time you’re already slicing things pretty fine. Does he just need one more speed grade in the middle? Or does he need to have more of a drunken walk around the velocity chart like Washburn?
I don’t know where to post this, or even if it belongs here [deleted, ot]
Question: how much of hitting is guessing correctly what the pitcher is throwing? If a lot of it is, then I suspect it’s better to have these distinct boxes than to have Washburnian overlap between them. Why? If one guesses wrong on a Washburn gray area pitch, one is going to be closer to it (and if it’s close enough, actually do something worthwhile with it) than if one guesses wrong on a non-gray Feierabend pitch.
I dunno if I’m positing that clearly or not. Hitter is guessing a pitch will be in the 85 box. From Wash being subtle, he gets an 82 (different box, but 3mph different). From Feier, he gets an 80 (different box, but 5mph different).
Does that seem reasonable? Yes, Washburn is mixing more and his spread is all over the place, but the difference doesn’t SEEM like as much (and often isn’t).
Edit: PS, are y’all having problems with your RSS feed? This post never did come up in my feed…
I thought you were going the other way with this one, Dave.
If a guy’s guessing (90 MPH fastball) and he’s wrong, might he still be able to hit the 84 MPH mildly curving curve whereas he’d miss the 78 MPH one?
I thought that differentiation between your pitches was generally an advantage.
Its all about timing.
If a hitter is sitting on a particular pitch, and the pitcher throws that pitch at a little different speed than the batter expects, it becomes a different pitch.
Moyer does this a lot. His change-up can come in at varying speeds, so that even if the hitter is sitting on it, he doesn’t expect it to be that exact speed, and his timing is off.
It reminds me of something I read in the book Men at Work, by George Will.
In it, Orel Hershiser talks about really having 9 or 10 pitches, because he threw his curve (among others) at many different speeds.
Same grip, same arm action, but different speeds based on what he thought would work best at that particular time.
Oh, wouldn’t it have been fun to have had PitchFX when Hershiser was pitching?
Moyer is the obvious example of this phenomenon because he’s the most extreme. As Dave’s fond of pointing out: There’s nobody in the last twenty years or so who’s been quite like him. You’re not supposed to be able to get major leaguers out with an 82 MPH fastball.
Is this really likely? There are eight men fighting for five jobs next year, really six fighting for three (I assume Felix, Bedard and Morrow are pretty much locks). If I’m Feierabend, the man they call the Bus isn’t someone I’d trust to teach me anything.
“A picture is worth a thousand words”
Two things:
Can you add color coding to the charts as to what pitches were strikes, balls, foul balls or hits?
It would be nice to see a Jamie Moyer chart