Back Leg Specials (BLS stat)
... the hole in BABIP and xFIP

.

=== (Seager's Double)===

Have you ever noticed, that in jewelry stores,  they are careful to put diamonds onto black velvet when they show them to you?

Contrast, relief, and proportion are important.

Here is the pitch sequence that Brad Peacock threw to Kyle Seager, right before Jesus Montero's monster home run:

Worked him to a one and one count, using two fastballs, and then threw him a beautiful changeup just outside. (If you just joined us, hard stuff in – soft stuff away is the way to take the poison out of a strong player's bat.)

Perfect one and one changeup, and Seager rocked a double the other way.  You are Brad Peacock. What else are you supposed to do?

.

=== Jesus Montero's Blast ===

One extreme to the other, babe.  On the next hitter, Montero, the pitch execution left something much a mortal ton to be desired.

On the second pitch below, they ran the super slo mo. You've heard the expression, "his eyes got wide as saucers"? Somebody back me up here. Montero's eyes literally widened as the ball got halfway to home plate.  One of the sports images of the month.

He set the back foot, cocked the bat, and smote the ball some 450 feet into his power alley.  The ball carried another 150 feet on the carom back into center field.

s -- as was McCracken's original theory.  

.

=== The Metrics ===

Maybe Matty could ride this idea, along with his others, to the assistant GM's chair?  ;- )

The thing is, it may be that there is no way to capture this subset of batted balls statistically, and if not, "scientists" may prefer to minimize its importance.

I can only think of one way to quantify it.  That is to have an experienced observer go back over the tapes and TAG those pitches on which the batter apparently guessed right, and got the pitch in a good location, and loaded up for the pitch, and took a big swing to his pull field.

There would be a certain amount of subjectivity involved -- but that is the case also for "errors," "UZR," and other stats that require scorers.

I for one would be fascinated to see all these LD%, HR/F, xFIP, etc. stats AFTER they were adjusted for mistake pitches.  

Cheers,

Dr. D

............

*by "Hanging Breaking Balls" we mean any bad mistake the batter loads up on.  Two of the three pitches above were fastballs.

 

Comments

2
IcebreakerX's picture

I wonder if there's a difference or pattern in the difference between pitch speed and speed of bat depending on the 'Stupid Factor' of the pitch...?

3

How often have you seen or heard this slogan from a retailer? My first job was at a local men's wear shop and the owner made this his mantra. It's cute, it's supposed to be funny, but taken at face value it's arrogant. And yet that is precisely what strict sabertistas say when a subject like this is brought up.
Ever heard of a doctor telling a patient, "Well we ran all the standard tests and nothing showed up..." with the clear implication that it's all in their mind?
Just because a solution to a problem isn't (yet) known doesn't mean it doesn't exist, and it certainly doesn't mean there isn't a problem.

4

I obviously cannot discuss proprietary things regarding what the Yankees do internally, but I can say that I pitched essentially your idea (with a more complex statistical technique to achieve it using pitch F/X data, rather than with eyeball scouting of the video) as a part of the interview that got me the job as one of their research interns.
I assume that most teams are well aware that the mistakes are the key to winning baseball games. The Mariners, though, were missing all kinds of bad pitches against other teams before this one Houston game. It's not just the location of each pitch...it's the location of that pitch combined with the sequence of previous pitches. You can throw a hanging slider and get a garbage swing if the previous two pitches were murderous 100 mph fastballs in on the hands. So someone needs to figure out what types of sequences lead to allowing the hitter to recognize the mistake fast enough to load up for it.

5

The corollary here is to find pitchers who "avoid" the hanging mistake.
Maddux types.
Would be interesting to see the data on "breaking pitches in the middls of the plate," and how that relates to pitcher success. It's not hard to guess the results.
moe

6

I attempt to account for the difference between "balls hit with authority" and "random-y balls in play."  When you just use BABIP it all goes into the wash and treats all batted balls the same, but obviously the BABIP for the former will be much higher.
My stuff is based on the rough assumption that the number of extra-base hits that "should have" been outs is roughly equal to the number of singles that "should have" been XBH.
So I just assume that all XBH are "balls hit with authority"  a player's XBH% will equal the percentage of "balls hit with authority"and all singles are the result of "random-y balls in play." a player's percentage of singles and ball-in-play outs will equal the percentage of "balls in play without authority."  [EDIT: I think that might be clearer.  Not assuming that that's what actually happens on the field of play.] It's a rough assumption, but it works pretty well, and it's the best one can do with minor-league data anyway.
So my "Plate Skills Index" considers "hitting the 'hitter's pitch' with authority" is a Plate Skill, and so HR% and XBH% goes in there with BB% and K%.
My "Production Index" is centered on ISO (which is just SLG excluding singles), but also factors in BB% and K%, since they increase/decrease the number of opportunities to convert "random-y" opportunies (see the Allegory of the Window).
My "Command/Control" Index is just the reverse of "Plate Skills" in that "avoiding giving the hitter a 'hitter's pitch'" is factored in as a "Command/Control" factor along with walks and strikeouts.
My "Stuff" Index is the reverse of "Production" and it measures how much the pitcher forces the hitter to rely on "random-y" opportunities rather than allowing walks or "balls hit with authority."  Paxton giving up a string of singles last week wouldn't affect his rating.  Paxton giving up a bunch of HR and walks last night does.
Seems like we're all kind of converging on the same things here.

7
tjm's picture

. . . is the single largest factor that distinguishes good players from bad. There are tons of guys in AAA who can throw a great breaking ball, but can't throw it all the time. The ability to throw it is what keeps players in organizations. The inability to throw it all the time keeps them out of The Show. Consistency is the great separator. I don't think there's any known way to quantify this; it's why you have scouts and minor league coaches.
This is true in every sport. I've hit 320-yard drives (most of them into second growth forests) probably five times in my life. Tiger hits them at will. I've also made 28-foot jumpers every once in a while. Fred Brown made them in his sleep.

9

Thanks for sharing Matt!
What would you CALL it?  That shouldn't be proprietary.  You can do versions with, or without, the batter actually pulling the ball in the air ... obviously a mistake could be MISSED; the hitter could just get under it.
From the pitcher's standpoint I'd want to know when he threw a mistake and the batter let the bat fly; from a FIELDING standpoint I'd want to know when the batter deliberately launched one, to "correct" BABIP etc ... from a hitter's standpoint either one would be good.
BLS category ... Back Leg Specials :- )

10

And considered this one of its Five Pitcher Skills.  They've gotten away from it since, unfortunately, buying too much into the BABIP, GB%, and HR/F theory.
For now Mo' we can just keep an eye on HR/9 and HR/F, tracking the trends across seasons.  Without looking, I'm sure that Pedro's numbers will be an example of how to do it right...

11

As you're talking about it here Terry, the tumblers in the keyhole start clicking on every sport I've ever watched.
The problem is, the players at the highest levels have already FILTERED this factor so finely that it's hard to perceive... once knew a girl not interested in sports, and the only thing she ever commented was during a NBA game once.  "I don't like watching it.  They never make a misTAKE!"

13
tjm's picture

Carlos Peguero. His "potential" is tantalizing but I just don't see there being any chance he ever becomes a consistent MLB hitter. I'll bet few people in the organization do either, but it's so damn hard to give up on a guy with such a huge upside.
Remember Bobby Ayala? Great stuff, could make hitters look horrible on one pitch then Hall of Famers on the next. You can imagine team execs saying to one another: "If he could just be more consistent." Right, but he couldn't.

14

Man, this is a paparazzi thread ... "They blinded me with SCIENCE!"
While I could definitely buy into the concept of mistake-hitting ability being a MAJOR component of what makes great hitters great ... and mistake pitch frequency being a key factor in pitching performance ... I think it is important to remember that by definition, mistake pitches are not intended, nor predictable. They are mistakes. The pitcher is NEVER "trying" to throw a mistake pitch.
From a psychological standpoint, I think it would deadly to have pitcher or catcher attempting to "protect" against mistake pitches. Because, what you are really doing is encouraging both to be thinking in terms of mistakes, (which will typically mean manifesting more of them - not less). The importance of confidence in the pitches you are throwing is HUGE - (I would think anyone remembering the Wash/Johjima debacle would have that one ingrained).
Where I see overlap with some of Matt's previous work is in the examination of the "randomness" of the mistake pitches.
I think we all understand pitchers have different mistake frequencies. "In general", the higher frequency, the more you get punished.
But, hitters also have varying skill in punishing mistakes.
Pitcher A could have a 5% mistake freq and pitcher B could have a 10% mistake freq.
But, if pitcher A is "unlucky", he might happen to make mistakes against the wrong hitters and get overly punished.
Pitcher B might get "lucky" and make a large portion of his mistakes against weaker hitters that don't punish him.
The outcome is pitchers whose overall stats are fairly similar --- but one is likely to improve and the other likely to swoon next season.
Of course, it is possible that some pitchers actually tend to make their mistakes against weaker players - so you would want to profile pitchers to determine if their mistakes ARE random - or if they actually have trends that are measurable, (the Noesi "gets tight" under pressure vs. the Moyer "dials up the control" when it matters most).
But, imagine the production difference year to year if Fielder sees 50 mistakes one year, but 25 or 75 the next?
What is the year-to-year variance in "mistakes seen" from player to player?
I'm thinking the TTO players are probably kings of mistake mashing BECAUSE they look at so many pitches. Pity you can't go back in time and track every pitch somebody like Wade Boggs saw. The way he fouled off pitches until he found one to his liking ... it's fun to contemplate the concept of "inducing" mistakes through superior patience.
And THAT said ... what happens if you have a hitter (or hitters) whose history has been to take LOTS of pitches (and draw lots of walks), but they punish mistakes. What happens if you tell them - you're taking too many pitches. You need to be more aggressive?
I wonder what that kind of advice would have on hitters like ... oh ... Smoak or Ackley ...
I wonder ...

Add comment

Filtered HTML

  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd><p><br>
  • Lines and paragraphs break automatically.

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.

shout_filter

  • Allowed HTML tags: <a> <em> <strong> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.
  • Web page addresses and e-mail addresses turn into links automatically.