Zito Thoughts, Part 3 of 3
by Ken Arneson
2007-01-12 1:01

We completed Part 2 (Part 1 here) with the question, is Barry Zito really an exception to the rule, a pitcher who can reduce batters’ batting average on balls in play? If so, how does he do it, and what makes him different?

I’m just a fan, not a scout, but I have watched a lot of Barry Zito, so I think I have some idea of how Zito approaches batters.

Zito’s approach is a lot more complex now that he has five pitches instead of three. I think it’s useful to go back to 2000-2003 when Zito had only the fastball, curve and changeup, and study his basic three-pitch approach. To narrow it down further, we’ll focus just on right-handed batters.

So here’s the plan. It all starts with the curveball.

* * *

The Curveball

curveball


Zito doesn’t actually throw his curveball all that often–maybe 15-20% of the time. But Zito’s curveball sets up everything else. Without it, he’s a mediocre pitcher who has to survive with a good changeup, and below-average fastball.

To throw his curve, Zito grips the baseball very loosely in his hand. Occasionally, Zito will end up on blooper reels, becase the loosely-held ball will slip right out of his fingers. But somehow, this loose grip helps him get a tremendous rotation and break on the pitch. It’s almost certainly the best curveball of this generation.

When it comes out of his hand, the ball looks like it is going to end up high and just out of the strike zone to a right-handed batter:

And then it breaks, and ends up down and over the plate:

That’s a very difficult pitch to hit. Major league hitters can hit it, but only if they know or can guess that it’s coming. If they’re guessing something else, and it lands in the strike zone, forget it.

So to prevent the batters from just sitting on the curveball, Zito uses his fastball to keep them honest.

* * *

The Fastball

fastball

Zito’s fastball, a four-seamer, is usually only 86-88mph, although he can touch 90 on occasion. So he can’t overpower anyone with it. However, he uses the threat of the curveball to trick the batter into swinging at a bad fastball.

Remember where the curveball starts out? High and away to a right-handed hitter. Zito will often throw his fastball to the exact same spot where the curveball starts out:

You can’t hit that pitch hard, if you hit it at all. It’s too far off the plate to reach with the good part of the bat. Most hitters who make contact on that pitch hit the ball off the end of their bat, and the result is a weak popup to second base or right field.

Normally, a batter wouldn’t even think about swinging at a pitch in that location, but because the curveball starts out with the same trajectory, Zito can often induce batters to chase there, where other pitchers can’t.

Now, you can’t just keep throwing the same pitch up there, because eventually the hitters adjust and start to lay off. So Zito keeps batters honest by hitting the outside corner with the fastball as often as he throws it outside:

That’s the pitch the batter wants to hit. A good batter will lay off the curve and the high-and-away fastball until he has two strikes, and sit on that 86mph fastball on the outside corner. That’s like hitting in batting practice! No problem, right?

Right, except then there’s…

* * *

The Changeup

changeup

Zito’s curveball is spectacular, and gets all the press, but his changeup is also a great pitch. The release point is exactly the same as his fastball, but it’s 10-13 mph slower, usually around 72mph. So when Zito throws it to the spot where the hitter is waiting for that fastball:

it actually ends up right about here:

What often happens is that the batter will get his hips started too early, and hit the ball using only the strength of his arms. Zito takes all the power out of the swing, so what happens is usually a weakly hit ball. If our RHB tries to pull it, he hits a grounder to the left side; if he tries to go the other way, it’s usually a popup or a lazy fly towards center or right.

* * *

The Jammer

Now we come to the final, and key, piece to Zito’s approach. So far, all the pitches we’ve looked at have been on the outside part of the plate to a right-handed batter. Zito gets the batters looking and thinking and leaning outside corner, outside corner, outside corner…and then he comes in with his fastball, on the inside corner here:

It’s the same height as the fastball the right-handed batter wants to hit on the outside corner, so he’s tempted to swing at it. However, the batter has a really hard time getting the barrel on the bat on that inside pitch. He’ll often hit it with the thin, handle part of the bat, and end up hitting a weak popup to third base or left field.

Zito probably gets more outs with this pitch than any of the others.

* * *

And that’s Plan A, in a nutshell.

If Zito does, in fact, defy conventional DIPS theory with this, I think it’s the combination of those two fastballs–the one up and out of the zone, and the one on the inside corner–where it happens. He’s getting people to hit the ball not just on the upper half of the bat for popups, but on the end of the bat (up and away) or on the handle (inside corner), as well. The ball hits the weakest part of the bat, and makes for some very easy-to-catch balls in play. All from an 86-mph fastball, fooling batters thanks to one wicked curve.

* * *

When Things Go Wrong

Some days, when Zito has good control of all three pitches, Plan A works like a charm. On those days, like in Game 1 of the 2006, Zito can outpitch the Johan Santanas of the league. Other days, however, some adjustments need to be made.

Zito probably goes to the mound with all three key pitches working maybe a dozen times a year. The other 20 or so times, he’ll have trouble throwing strikes to varying degrees with one or two pitches.

His worst innings/days are when he’s having trouble with the release point on his fastball. His fastball isn’t good enough to throw down the middle of the plate and get away with it. When Zito misses on that inside fastball, and the ball leaks out over the plate, it tends to travel a very long distance. So Zito tends to err on the side of caution. He’d much rather walk a guy than throw him an 86mph batting-practice fastball down the middle of the plate. Sometimes, he’ll find his release point with the next guy, or he’ll use his changeup to get the next guy out, and be fine. Sometimes, though, he pays for the walk, and the result is a bad inning.

A more common problem, however, is that he can’t find the release point on his curveball. Since the curveball sets up a whole sequence of pitches, he has much more trouble getting batters out if they know he can’t throw the curveball for a strike. They can ignore the high fastball, sit on a fastball or a changeup, and have a much better chance at hitting the ball hard.

This became such a problem in 2003 that Zito would often begin a game without throwing a single curveball until about the fourth inning. If he didn’t show that he couldn’t throw the curveball for a strike, the threat still existed that he could. This strategy worked for awhile, but it seemed to catch up to him in 2004. Batters made him establish that he could throw the curveball for a strike before they chased that outside fastball.

On his very worst days, which happens two or three times a year, (unfortunately one such day was in the 2006 ALCS against Detroit), Zito can’t find either his fastball OR his curveball, (or maybe the umpire can’t) leaving him with nothing but a 72mph changeup to get people out with. People praised Detroit’s for taking some sort of brilliantly intelligent approach against Zito, but I found that argument silly. The Tigers would have looked just as foolish as Minnesota did the week before if Zito could throw his curveball for a strike that day.

When Zito gets clobbered, it’s says more about Zito that day than it does his opponent. Zito has allowed seven or more runs in a game eleven times in his career. Know which team has given him such a clobbering the most?

Tampa Bay, three times.

* * *

The New Pitches

You can pretty much draw a chart depicting how many runs Zito will give up by how many pitches he has working on a given night:

Pitches Innings Runs
3+ 7-9 0-2
2 6 3-4
1.5 5 5
1 4 6

It was after the last Devil Ray buttkicking that Zito implemented his radical change in approach, changing from a three-pitch pitcher to a five-pitch pitcher. He added a two-seam fastball, and a slider.

I honestly can’t see the difference between Zito’s two-seamer and his four-seamer, but since 2005, he seems to be able to get more ground balls when runners are on base than he used to. I think he uses it when he needs to throw a strike, and the hitter may be looking for that four-seamer. It has just enough of a difference in movement to get the batter to hit on top of the ball, especially on that inside fastball. The result has been an awful lot of ground balls to Eric Chavez.

The slider is mostly a weapon for Zito to use against left-handed batters, but Zito will also throw it in there against righties to try to steal a strike every now and then. It’s a pitch that starts out here:

and ends up here:

Again, this is another pitch that, if hit, often gets hit right at Eric Chavez, and now, presumably, Pedro Feliz. Fortunately for Zito and the Giants, Feliz seems to have some pretty good range at third.

* * *

These extra pitches give Zito more options when he gets in a jam. Too often in 2003 or 2004, Zito would get into a game/count situation where he only had one or two choices of which pitch to throw. The batters would foul off a bunch of tough pitches, and he’d have to come right in there again with the same pitch. Eventually, the batter would wear him out, and either hit him, or get a costly walk.

Now, Zito can toss in the slider or two-seamer when he’s in a jam like that, and give the batter something else to think about. He has more ways of getting batter out these days, and he’s a better pitcher for it.

But these extra options come at a price. Zito wanted to implement the slider way back in 2003, but then-pitching coach Rick Peterson didn’t let him, because when he throws it, his arm angle is slightly lower than with his fastball, curve, and changeup. With two separate release points to worry about, finding those points can be a problem.

In the two years, since he added the slider, he’s had more walks and hit batsmen than any previous season. He’s had a particularly hard time in the first few weeks of the seasons, with April ERAs over 6.00 each of the last two years. He does seem to find his groove after awhile, however.

But it is something to be concerned about, if you’re a Giants fan. I see no evidence that Zito’s stuff has declined, but there is evidence that his control is faltering, and that batters have adjusted to him more and more over the years.

* * *

On the other hand, Zito is switching leagues. He’ll be facing a whole new set of batters who haven’t seen Plan A before. He may not have to use the slider and two-seamer all that often, and perhaps he can go back to that three-pitch sequence, and lock in that release point earlier in the year.

But that will only last so long. Eventually, batters will learn his sequences, they’ll learn how to figure out what’s working and what isn’t in a particular game, and adjust their approach against Zito. And then Zito will have to adjust back. Can Zito keep those two release points under control for the next seven years?

* * *

I love watching Barry Zito. When that curveball is working, and those batters are flailing at that fastball up and away, and turning themselves in to pretzels trying to hit the jammer, there aren’t many pitchers more fun to watch. And I’m sure he’ll have plenty of those days over the next seven years. But there are too many warning signs, both from the stats and from the eyes, that Zito’s control may be wavering, that there might be too many days where Plan A is not in effect, to make risking that much money on him a wise choice.

I find it a perfectly plausible scenario that the Giants will get their $126 million worth from Barry Zito. But if you’ll notice, there were a lot of “however” and “maybe” and “perhaps”-type words in the last few paragraphs. The Giants could be getting a perennial Cy Young candidate, but they could just as easily be spending superstar money for an average player. If I’m spending $126 million, I want a lot more certainty, and a lot fewer maybes.

Comments: 21
1.   Ravenscar
2007-01-12 07:17

1.  Ken -

Let me be the first to say this post is worthy of the unanimous vote that Ripkin didn't get.

It makes me not know if I'm happy or glad that the Mets didn't get him.

2.   standuptriple
2007-01-12 07:54

2.  Solid as usual KA. Can't find anything I can disagree with.

3.   petereco
2007-01-12 08:12

3.  The only thing I can disagree with here is you said Zito's fastball is a "Four Seamer". The supporting photo you provided is clearly a "Two Seam" fastball. This isn't a correction, as he may throw both a 2-seam + a 4-seam based on the amount of movement he'd like on the pitch(most pitchers do). I don't know about Zito. You know more than I do, as the remainder of your evaluation clearly demonstrates.

4.   Ken Arneson
2007-01-12 08:38

4.  4 Good point...I didn't really look at the seams there, because it's pretty blurry, but looking closely at it, yeah, he's throwing a two-seamer there.

5.   Ken Arneson
2007-01-12 08:41

5.  4. Oops, I meant 3. And, as I say at the end of the article, he does throw both a two-seamer and a four-seamer now. Early in his career, he only threw the four-seamer.

6.   dianagramr
2007-01-12 09:10

6.  Brilliant analysis Ken ...

Make sure to include this in your portfolio of work when you go hunting for your next work gig.

7.   Ken Arneson
2007-01-12 09:20

7.  6 Don't say that! For one, #3 is my boss. For two, the current gig is no doubt going to be so successful that I'll never ever need another gig. Right, boss?

8.   misterjohnny
2007-01-12 09:46

8.  One question, does Zito have more problems finding his release point on the road than at home? Possibly due to different mound conditions?

9.   Daniel Zappala
2007-01-12 10:20

9.  Wow, amazing article. Thanks for teaching me a lot about pitching in general and Zito in particular.

There have been a lot of arguments that pitchers can't control BABIP, but you're arguing for it, and I can see the logic of it. There is bound to be a lot of variation from year to year in BABIP because it depends so much on so many variables -- how many pitches a pitcher has, and whether the batter he is facing at that moment is good enough to exploit the combination of pitches that isn't working, plus the combination of exactly what release point he gets and what swing the batter has. I would argue a pitcher can have control BABIP to a large extent, but we just have a hard time measuring it.

10.   thelarmis
2007-01-12 10:24

10.  Fantastic job, Ken! Very insightful.

I'm an AL East guy, but I'm assuming the Giants-A's meet for home & home in interleague?

11.   Tangotiger
2007-01-12 11:18

11.  Ken, I linked to your threads here:
http://www.insidethebook.com/ee/index.php/site/comments/solving_barry_zito/

And I also compared Zito's BABIP, by hand, against the league. Very startling.

12.   dianagramr
2007-01-12 11:40

12.  7

Oops ... my bad! :-)

Dear Ken's boss,
Please give Ken a big raise ...

Signed,
A loyal reader

13.   Ken Arneson
2007-01-12 12:17

13.  11 That's very interesting that Zito only beats the DIPS effect against RHB, but not against LHB.

If I had to guess why, here's my guess

1) Zito can't set up the LHB with that fastball that goes to the same location as where the curve ball starts out. That's because it would be a fastball right at the batter's head. He'll never swing at it. So the curveball doesn't really set anything up against LHBs. Instead, Zito has to work the other way around: use his fastball and change to set up the curveball as his out pitch. It's a much more conventional pitcher-batter approach, and the numbers probably reflect that.

14.   Ken Arneson
2007-01-12 12:17

14.  10 Yes, the Giants-A's play each other 6x/year.

15.   Ken Arneson
2007-01-12 12:38

15.  8 No, I don't think it's the park; I think it's pretty much a random thing which days he has good control, and which ones he doesn't.

The first couple of innings are always pretty interesting to watch in a Zito start, as both he and the batters try to figure out what tools Zito has to work with that night. Sometimes, he'll come out and struggle with the release point for an inning or two, and then cruise the rest of the way when he finds it.

One thing I'd worry about with the Giants and their awful bullpen: Zito tends to hit the wall pretty quickly when he gets tired. The Giants may be tempted far too often to leave him in one or two batters too long, and his ERA will suffer as a result.

16.   Orly Yarly NoWai
2007-01-12 18:15

16.  9 According to Baseball Prospectus's Baseball Between the Numbers, the idea of BABIP being random was first introduced by Voros McCracken, but research by Keith Woolner, Clay Davenport and Tom Tippett (from Diamond Baseball Mind) found some inconsistencies.

First, "some specialized pitchers- knuckleballers like Time Wakefield, control artists like Ferguson Jenkins, and soft-tossing lefties like Jamie Moyer- do demonstrate a consistent ability to limit hits on balls in play."

Second, major league pitchers allow lower BABIP than minor league pitchers.

Third, "inducing infield popups has proved to be a repeatable skill among major league pitchers."

Fourth, pitchers are consistent w/r/t their GB/FB ratios, and the two categories vary greatly in how often they fall for hits.

All from p. 92.

Zito DOES have a consistently high IFF%, which would help limit his BABIP.

17.   joejoejoe
2007-01-12 18:44

17.  Wow. What a great post. I'm sure you are more familiar with Zito than just about any other MLB pitcher but I'd love to see this kind of analysis (with the fine and rich graphics) of other pitchers. I learned a lot and I've been a baseball fan all my life.

18.   Bill Crain
2007-01-12 19:01

18.  On Tangotiger's site, The Chronicler wrote:
It would be great to have the pitch location data to verify this connection. But it's nice to see subjective and objective analyses complement each other

Even pitch location data really ain't gonna finish it, is it? We're gonna need at least slope of the curve at point of impact (or front of the plate or something). We pretty much have velocity. Is there somebody collecting these data? Thinking of others?

I mean this is about more than just establishing relative values for players and strategies. This is searching for intent along with ability. Can the stats express the value of adjustment? In the context of a game situation? Of a season? Of a career?

Cool, it's the System of the World.
Yeah, OK, maybe just The Next Big Thing.

19.   Professor Oakland
2007-01-12 23:50

19.  I bow. Impressive piece, Ken.

20.   Ryan Armbrust
2007-01-13 20:00

20.  Great work, Ken. That's as good a breakdown of Zito as I've seen. It's one of the most well thought out breakdowns of any pitcher, actually.

Oh, and that page of mine you linked to about the umpires? I noticed that I had some formatting errors on there that escaped me when I switched from blogger to wordpress. They're fixed now. Thanks for the link, though.

21.   Vishal
2007-01-14 04:45

21.  fantastic! thank you, ken. this is like a whole new understanding of how zito works. i've often wondered about the chess-game aspect of the pitcher vs. batter dynamic, and i get a little bit about things like "fastball counts" and "working backwards" and how pitchers choose what pitch to throw and where, and how that lines up with the expectations and anticipations of the hitters, but this aspect of the game i think goes largely unwritten about.

Comments on this post are closed.
This is Ken Arneson's blog about baseball, brains, art, science, technology, philosophy, poetry, politics and whatever else Ken Arneson feels like writing about
Original Sites
Recent Posts
Contact Ken
Mastodon

LinkedIn

Email: Replace the first of the two dots in this web site's domain name with an @.
Google Search
Web
Toaster
Ken Arneson
Archives
2021
01   

2020
10   09   08   07   06   05   
04   

2019
11   

2017
08   07   

2016
06   01   

2015
12   11   03   02   

2014
12   11   10   09   08   04   
03   01   

2013
12   10   08   07   06   05   
04   01   

2012
12   11   10   09   04   

2011
12   11   10   09   08   07   
04   02   01   

2010
10   09   06   01   

2009
12   02   01   

2008
12   11   10   09   08   07   
06   05   04   03   02   01   

2007
12   11   10   09   08   07   
06   05   04   03   02   01   

2006
12   11   10   09   08   07   
06   05   04   03   02   01   

2005
12   11   10   09   08   07   
06   05   04   03   02   01   

2004
12   11   10   09   08   07   
06   05   04   03   02   01   

2003
12   11   10   09   08   07   
06   05   04   03   02   01   

2002
12   10   09   08   07   05   
04   03   02   01   

1995
05   04   02