A Little More On Fastball Swinging Strikes

by Jeff Sullivan on Mar 3, 2009 9:26 PM PST in The More You Know Comment 44 comments

There were 168 starters last year who threw at least 500 fastballs (that got picked up by PITCHf/x). Ranked in descending order of swinging strike rate, Jonathan Sanchez ranks 5th (9.1%), Matt Cain ranks 12th (8.5%), Randy Johnson ranks 13th (8.3%), and Tim Lincecum ranks 18th (7.4%). Even Barry Zito was above the median. Between that rotation and a Lewis/Rowand/Winn outfield, San Francisco is not going to allow runs.

Among the names at the bottom of the list (Dallas Braden, Livan Hernandez, Kyle Lohse, Jeremy Bonderman, and Garrett Olson), Bonderman's name is by far the most interesting. His trick shoulder has taken a lot of life out of his fastball and left him as a one-pitch pitcher. Under contract for $25m over the next two years, the Tigers badly need him to bounce back, lest he become another expensive pile of Willis.

J.P. Howell threw 722 fastballs and missed only 24 bats. He also posted a 2.22 ERA and 9+ K/9 over 89.1 innings despite being a soft-tossing lefty facing predominantly right-handed batters. Having him and Chad Bradford in the same bullpen gives the Rays a unique bit of flavor. Pairing them with Grant Balfour is funny to me.

Felix posted the same swinging strike rate on his fastball as Tom Gorzelanny.

Craig Hansen and Bobby Jenks: blazing fastballs, tons of contact. Remember, kids: velocity isn't everything.

I can't imagine what it feels like to be one of the 20 starters last year who missed fewer bats with their fastballs than Jamie Moyer.

0 recs | Comment 44 comments | Share

Story-email Email Printer Print

Comments

Kyle Lohse

):<

by JI on Mar 3, 2009 9:44 PM PST reply actions 0 recs

41 million dollars!!!

vivaelbeñsheets

by vivaelpujols on Mar 3, 2009 10:28 PM PST up reply actions 0 recs

lalalalalaIcan'thearyoulalalalalala

by JI on Mar 3, 2009 10:29 PM PST up reply actions 0 recs

Tom Gorzelanny? Yikes!

Why would I watch Transformers when I can play with them at my house?

by Taylor H on Mar 3, 2009 9:51 PM PST reply actions 0 recs

I though Garrett Olson's strength was his fastball?

I remember reading his movement on his fastball was fantastic, and that the problem lied in locating his curve and change.

by Ezzra on Mar 3, 2009 10:13 PM PST reply actions 0 recs

He has a reasonable slider and change, but his fastball is an issue

by Jeff Sullivan on Mar 3, 2009 10:15 PM PST up reply actions 0 recs

Moyer has a fastball?

I thought he sported a change-up and a super change-up.

The ageless wonder indeed.

by Wilder. on Mar 3, 2009 10:27 PM PST reply actions 0 recs

I wouldn't call it a fastball.

by Fin on Mar 3, 2009 11:26 PM PST up reply actions 0 recs

It's faster than I can throw.

by msb on Mar 4, 2009 12:12 AM PST up reply actions 0 recs

He sure knows how to establish the change up

I wish Washburn knew how to use his fastball like Moyer

by Kermit. on Mar 4, 2009 2:35 AM PST up reply actions 0 recs

I wish Washburn pitched for Philly like Moyer

by oneseasoncom on Mar 5, 2009 11:09 PM PST up reply actions 0 recs

Think they'd trade straight up?

There are no good individual basketball statistics.
54!

by joof on Mar 7, 2009 9:16 AM PST up reply actions 0 recs

That reminds me of the good old days with Ken Griffey Junior Baseball for N64 where pitchers could throw 46 mph super changes that moved all the way across the strike zone. And then you could have pitchers who threw super fast for 103 mph.

by Mbad on Mar 4, 2009 12:13 AM PST up reply actions 0 recs

Slowball.

by appleshampoo on Mar 4, 2009 2:02 AM PST up reply actions 0 recs

And

I saw him get a few strikeouts with that slowball in Chicago last year. Amazing thing watching hitters flail at it. I guess a few of them might have been the cutter, but still, it’s coming in at 82 MPH!

by appleshampoo on Mar 4, 2009 2:03 AM PST up reply actions 0 recs

But when the hitter is expecting a change at 68 mph

82 gets to you a lot quicker than you’d like.

I love watching Moyer pitch. He spots that change on the corner so well that hitters take it, and then when he finally does throw a hittable strike the batter hasn’t taken enough swings to have the timing down (assuming he ever could, given the seemingly random speeds of Moyer’s pitches).

I like using semi-colons; they make me feel smart.

by Llewdor on Mar 4, 2009 9:27 AM PST up reply actions 0 recs

I love watching Moyer pitch in theory but in practice it is God awful

by Jeff Sullivan on Mar 4, 2009 10:01 AM PST up reply actions 0 recs

I love watching Moyer pitch in practice, because in theory it's basically impossible.

He is the bumble bee of MLB.

by marc w on Mar 4, 2009 10:41 AM PST up reply actions 0 recs

Bemble bees can theoretically fly now

Have you now heard?

by Graham on Mar 4, 2009 2:47 PM PST up reply actions 0 recs

Do you need a nap?

by marc w on Mar 4, 2009 3:13 PM PST up reply actions 0 recs

Always

by Graham on Mar 4, 2009 5:23 PM PST up reply actions 0 recs

I had heard that.

They basically use suction. Their wings are like a vacuum pump. It’s pretty cool.

I like using semi-colons; they make me feel smart.

by Llewdor on Mar 5, 2009 2:28 PM PST up reply actions 0 recs

It is sad

that Dontrelle Willis has basically become synonymous for shit.

vivaelbeñsheets

by vivaelpujols on Mar 3, 2009 10:30 PM PST reply actions 0 recs

I think its funny.

Since, hey, we were right.

by Matthew on Mar 3, 2009 10:45 PM PST up reply actions 0 recs

Agreed

by JI on Mar 3, 2009 10:59 PM PST up reply actions 0 recs

I can't believe he sucks this badly

I mean, you can’t entirely luck your way to 22 wins can you?

by seattlebruin on Mar 3, 2009 11:50 PM PST up reply actions 0 recs

At one point he wasn't bad

Then he lost his velocity and control over time

by OlSalty on Mar 3, 2009 11:55 PM PST up reply actions 0 recs

Willis is a poster boy for overworked young arms

Pitchers that young shouldn’t throw that much. Don’t think about Felix. Do not think about Felix.

by philosofool on Mar 4, 2009 5:23 PM PST up reply actions 0 recs

You mean it would have been a bad idea to trade Adam Jones for Dontrelle Willis?

There are no good individual basketball statistics.
54!

by joof on Mar 4, 2009 12:49 AM PST up reply actions 0 recs

But he's changed his delivery!

There’s a note on Rotowire:

News: Willis has tweaked his mechanics — abandoning his trademark high leg kick — in hopes of having better command of the strike zone this season, the Detroit News reports.

Analysis: The new mechanics have Willis attacking the strike zone, hoping to take advantage of the Tigers offseason upgrades on the defensive side of the ball. “I accomplished what I wanted to do, throwing everything in the zone, making them hit everything,” said Willis after his throwing session on Saturday.

by Teej on Mar 4, 2009 2:02 PM PST up reply actions 0 recs

Sounds like the second coming of Carlos Silva

except a lefty

vivaelbeñsheets

by vivaelpujols on Mar 4, 2009 3:56 PM PST up reply actions 0 recs

Except for the fact that Silva is 1,000x better.

by acblue on Mar 4, 2009 3:59 PM PST up reply actions 0 recs

Well yeah

But that isn’t saying much about Silva

vivaelbeñsheets

by vivaelpujols on Mar 5, 2009 9:59 PM PST up reply actions 0 recs

He's the white Mike Cameron

I fucking hate you Mariners

by kentroyals5 on Mar 4, 2009 7:51 AM PST up reply actions 0 recs

Bat missing pitchers vs. bat missing pitches

Okay, I know that the relationship between being a pitcher who misses bats and a pitcher who gets a lot of strike outs is well established. However, has the relationship between being a pitch that misses bats as opposed to simply getting strikes been shown to indicate anything about a pitch.

It would seem to me that the best measure to put on pitches is simply strike/ball/contact-fair/contact-foul. I’m not sure why it would matter for a pitch whether it’s a swinging strike or a called strike.

by philosofool on Mar 4, 2009 5:35 PM PST reply actions 0 recs

I just looked at the post on morrow's fastball

and from the list of guys that miss lots of bats with their fastball, it looks like the relationship between bat missing heaters and good pitchers is strongly suggested by some data.

Still, I think the use of Pitchf/x has a long way to go before we understand the relationship between pitches and pitchers. Good thing people are working on it.

by philosofool on Mar 4, 2009 5:49 PM PST up reply actions 0 recs

Because I believe swinging strikes are a better measure of pitch effectiveness

Swinging strikes require the direct participation of two parties, whereas called strikes require the direct participation of one. Obviously a lot of pitches are called strikes because they fool the batter (I imagine a lot of pitchers have repeatable CU StC%‘s), but there’s noise in there that isn’t there for swinging strikes.

by Jeff Sullivan on Mar 4, 2009 8:38 PM PST up reply actions 0 recs

Yeah I'm a little uncomfortable just ignoring called strikes because they appear to not be useful

Seems to me there should be a lot of skill in getting called strikes.

Guys who don’t throw in the strike zone often should have lower called strike %.
Guys who can locate breaking balls should get tons of called strikes.
Guys who lack quality pitches should have less called strikes.

I just have a hard time saying that these are all insignificant.

My theory on the correlation between swinging strikes and strikeouts:

To get a strikeout you need to get not just one but three strikes. With two strikes the batter buckles down and doesn’t let hittable pitches past him. The only pitches that do get past him are swinging strikes because nobody wants to strikeout with the bat on their shoulder. The only time I can see this routinely breaking down is with a 3-2 count when a pitcher throws a well placed breaking ball that falls in for an unexpected strike. For a pitcher to get a strikeout he needs to do two things that are related. First, he needs to get two strikes on the batter. For these pitches, a called strike and a swinging strike are probably equally valuable. For the third strike, the pitcher probably needs a swinging strike to be credited with a strikeout. If a pitcher can not locate his pitches enough or fool the batter then he will probably not accumulate enough called strikes to get close to a strikeout. The only way this pitcher will be able to last in the majors is to get a lot of swinging strikes.

I guess I just feel like there has to be useful info in the called strikes info. Maybe its noisy but that doesn’t really mean we should just ignore it. I haven’t fully thought about this to myself yet but there must be some way to constructively use it.

by Edgar for Pres on Mar 5, 2009 12:10 AM PST up reply actions 1 recs

This is very eloquent

and matches how I feel about called strikes. Well said.

by appleshampoo on Mar 5, 2009 4:37 PM PST up reply actions 0 recs

You're almost certainly right

I was about to start a bunch of speculation about how the correlation of StS% and K% correlates because there’s more dispersion in the StS% and blah blah blah. Then I figured better than speculate, test the hypothesis.
Hypothesis Total strike percentage is more important to K rate than StS.
Test method: Determine correlation coeffcients between K/100pitches and total strike rate and K/100P and StS for all pitchers with at least 500 pitches as starters in 2008. (This method was chosen because that’s the data I had really handy in an excel file.)
Result: the correlation between strike outs and swinging strikes is .84 and between strike outs and total strikes is .79. However, the correlation between looking strikes and strike outs is only .1. This strongly suggests that if I were to sit down and do a proper two variable regression on K/100P with StL% and StS% the coefficient for StS% would be the big one, suggesting that the ability to get a swinging strike is much more of a performance indicator. Additionally, the correlation between St% and StS% is .68 while it’s just .57 between St% and StL%. The hypotheis is false

by philosofool on Mar 5, 2009 9:12 AM PST up reply actions 0 recs

One thing I always wonder about all the correlation studies people do for saber stuff

We almost always look at individual peripheral stats by themselves in a linear fashion. That means that I look at an equation that is Strikeouts = a*(swinging strikes) + b*(called strikes). This assumes that a strikeout can be had by just getting a shear number of strikes on a batter (in this case 3 strikes). That seems pretty logical until you think about the individual matchups you see. A pitcher can overpower bad hitters and get three swinging strikes but often strikeouts are dependent on fooling a batter to not swing at a pitch in the zone (curvy things) or throwing well located pitches that are called strikes in addition to swinging strikes to “finish him off”.

If a hitter can defend against called strikes for the 3rd strike by taking a different approach where he becomes more aggressive and sacrifices power for contact we would expect the pitcher to need to throw a really good pitch to get it by him to rack up a strikeout. Earlier in the count it is probably easier to get called strikes than swinging strikes because the hitter will only swing at pitches they think they can hit and appear to be in the zone. I might argue it might be worth looking at K% = a*(StS)+b*(StL)+ c*(StS)*(StL) because a good pitcher should need to be able to get both called and swinging strikes to get lots of strikeouts.

I don’t know if it would show any better results but I think all too often we try to model baseball as simply counting up numbers and multiplying them by a weight to get a value. There is no specific reason we always should ignore higher order terms in stats. For example FIP~ a*K + b*BB + c*HR. If a picher gives up 1 BB per game the run value of that walk will be much different than if the pitcher gives up 9 BB per game. Since we use a linear relationship all of that is lost. Although it might not be perfect a relationship, FIP ~ a*K + b*BB + c*HR + d*BB^2 might actually model it slightly more correctly. The major problem with adding more parameters is that you can model anything given enough adjustable parameters. For most things we can probably ignore the higher order terms (science does this all the time) but they do contain information and as we strive for more accuracy I think at some point we will need to take a closer look at them.

by Edgar for Pres on Mar 5, 2009 10:19 AM PST up reply actions 0 recs