Is there a "Best Pitch?"
Or is it just stick to what you do best ... ?


A little philosophical birdwalk laid out by The Counselor.  


Q by Mojo:  Maybe this is common knowledge, or maybe it is just absurd, but are changeups the most effective off-speed pitch?

I've watched a whole summer of Vargas and now Ramirez embarrass hitters with a change that dives into the dirt. It seems that curves and sliders often get botched, hanged and then hit for extra bases, but the changeup rarely goes wrong and seems just as effective when it goes right. Maybe the curve and slider are more difficult to throw and more unreliable because they rely heavily on a hard spin for the ball.

Is this right baseball thinking?

A:  Why do you say "common knowledge" OR "absurd"?  You're a lawyer.  Who would know better than you not to present those two terms as antonyms!  Oh yeah.  Bill James, 1977.  Well, you're probably second.

It would be incredibly trite to say "it depends on the pitcher."  Mojo axs, normalizing for the pitcher, is one pitch inherently better than another, in any meaningful sense.  It's better to swing away than to bunt.  It's better to shoot an open 3-pointer, if you have it, than it is to run a set 5-man play.  It's better to run a 4-3 defense than to put 10 men in the box.

How would you think about this?  Is a changeup the best pitch?


Q.  Does one pitch rate out better than another?

A.  Limiting ourselves only to 2nd-order thinking, here, we have the following simple outcomes for each type of pitch.  This is a leaguewide stats tabulation from the fantabulous statistical site Fangraphs.  The top row is 2009, followed by 2010, 2011, 2012.

In the first cell (if I understand wFB/C right) -0.15 means that for every 100 fastballs thrown, the pitcher's team suffered an increase of .15 runs scored (above average) by the other team's hitters.

Q.  Wow!  Easy!  The slider is by far the best (normal) pitch and the fastball's the worst, right?

A.  Not so fast.  There's the concept of "pressure" on a pitch type.  If you throw lots of fastballs, and the hitters are always looking fastball -- which is in fact the case -- then a slider is "surprising" the hitter and "playing up."

Fastballs are thrown 58% of the time, while sliders are thrown 14% of the time.  Imagine if Kevin Millwood went out and threw 58% sliders, and just mixed in his 90 MPH fastball once or twice an inning.  The slider's run value would be like -2.00.

So the above table represents what a pitch type is giving its team AFTER it has decided to put almost all the pressure on the fastball.  This implies, obviously, that the fastball is inherently the best pitch, by a wide margin.


Imagine if the 1967 Green Bay Packers controlled football by pulling the guards and running power sweeps.  Imagine that they gained 7.1 yards every time they ran it.  Imagine teams way overplaying it, pulling their safeties, putting in 5 linebackers, and moving everybody to the sidelines.  

Then imagine that the Packers got 4.5 yards per power sweep, and 4.5 yards per line buck up the middle.  Statistically it would then look like the power sweep was no big deal.  But, actually, the sweep was making everything else work, and was by far the best "inherent" play.

The 2005 Seahawks actually made less yards running behind Walter Jones and Steve Hutchinson than they did running to the right.  You guess why.


That's why we can't take the above table at face value...


Q.  The changeup looks about neutral.

A.  There are too many variables here to be dogmatic ... my own observation is that a changeup's wCH/C number is a reflection of how much a hitter respects a pitcher's fastball.  Other things being equal -- supposing that a pitcher has a generic changeup -- I look at Justin Verlander's sky-high changeup numbers and go, "Ah, there's a guy whose fastball is in their heads."  For me, the CH number is really a FA number.  Mostly.

In that respect, I'd say that a generic changeup has the least inherent weight to it of any pitch type -- good OR bad.

Mojo sez, is the changeup the best OFFspeed pitch ...the easiest to throw, no doubt.  The M's teach everybody a circle change, or used to, because it's supposed to be the easiest pitch to add, and easiest on the arm.  People don't even throw sliders and curves unless they're really good at them.  Maybe that implies that a change is the best pitch; if everybody worked as hard at the changeup as they do at the curve, the changeup would be the best pitch?

If they could teach everybody an Erasmo change, they'd have the new market inefficiency.  Pitchers who leap plateaus.


Q.  How about Jason Vargas?  Felix?  Erasmo?

A.  As y'probably know, Vargas' and Moyer's changeups weren't generic.  Vargas' changeup has run values the last three years of +1.6, +0.5, and now the crazy +2.4 runs per 100 pitches that Mojo has been keen to notice.  This is despite putting HUGE pressure on it - he throws it 26 to 29 percent of the time (!!) and nobody respects his fastball.

Vargas' motion and arm action is super deceptive.  The change fades and drops a ton.  And Vargas spots the change.  It's one of the best single pitches in the game, and it keeps Jason Vargas in baseball.


Felix' dry spitter shouldn't be called a change, though he does hold it like one.  He needs a name for it.


Erasmo's change is also extra-class.  Inherently.  It drops a lot, and it drops late, and the arm action on it is very convincing.  Obviously if a pitcher is really good at ANY pitch, that's what he needs to build around.



Q.  No best pitch, except the fastball, in the most theoretical terms?

A.  Dr. D has an idea that, under the current circumstances, a shuuto is a whale of a pitch.

Iwakuma's is a generic shuuto (except for location).  He throws it fully 22% of the time.  People don't respect his fastball that much.  And the run value on it is +2.57.

If there's a market-inefficiency pitch, it might be that.  Terry can tell you about Fernandomania.  :- )



ghost's picture're changing up off of some expectation and you sell it with good arm action...if you teach the average AAA pitcher a quality change-up in terms of movement and velo differential, it will only make him better (and not MUCH worse) if he can sell the change up as something faster.
Moyer is a rare example of an exception that proves the rule. He didn't have a velo differential and yet was still uber-effective because (a) he had ridiculously good command and (b) even with a low velo on his fastball, he was still willing to pitch you inside with the sneaky little stinger at 86 and make you look silly, and those stingers came at completely random intervals and looked identical to his change.

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.


  • 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.