Inside the change-up

One of the things that PITCHf/x taught me off the bat was that while change-ups may be all about disrupting timing, they are not just about changing speeds. For an offspeed pitch, the change-up is merely average velocity, with the average slider a couple mph faster, and curveballs about five miles an hour slower. But far from being just a “slowball,” most change-ups sink significantly compared to a fastball (due to less backspin, not just velocity), making them hard to do much with even if the batter knows one is coming. So how important is the difference in speeds, or are other factors like drop and control just as important for the effectiveness of a change-up?

To begin, here’s a look at the pitchers who are at the extreme ends in changing speeds. I decided to compare the change-ups only to four-seam fastballs, because although some pitchers throw more two-seamers or sinkers than straight fastballs, it’s the maximum velocity pitch that a hitter has to be prepared for that is really getting him out ahead of the change-up. And so, a guy like Chad Bradford, who PITCHf/x thinks has never thrown a fastball in his life, get overlooked here despite his 69 mph changeup. Fortunately, we can look at more than 400 other pitchers with more than 100 change-ups thrown since PITCHf/x started in 2007.

Greatest difference between change-up and fastball:

Name              Fastball velocity  Change-up difference   Change-up LW
Brian Fuentes   91.5                  -18.7                         0.035
Danny Herrera  88.4                  -16.3                         -0.040
Brian Stokes     95.0                  -15.3                        0.023
Clay Buchholz   92.6                  -14.9                        -0.011
Dallas Braden   87.5                  -13.0                        -0.028

Least difference between change-up and fastball:

Name              Fastball velocity  Change-up difference   Changeup LW
Chris Sampson  88.1                 2.8                           -0.008
David Weathers 87.9                 4.19                          0.023
Nelson Figueroa 87.3                 4.34                          -0.029
Mark DiFelice     85.8                 4.54                          -0.014
Brian Bannister   89.0                4.59                          -0.010

The last column is the average linear weight value for each pitcher’s change-up, using the average value for each ball and strike as well as balls in play. The average linear weight value for a change-up is -.006.

Nobody at the extremes stands out as being a master of the craft (Johan Santana has a 10.1 mph difference with a 93.0 mph fastball and a 83.1 mph change-up), so let’s take a broader look at the entire league:


In general, the more change of speed the better, although the benefit is marginal at the lower levels and starts to take off only once you get above the average, which is 8 miles per hour.

Note that the five pitchers with the least difference between the two pitches also have below 90 mph fastballs. That’s not a coincidence—as you might expect, pitchers with a higher mph ceiling can have more of a gap between their fastballs and change-ups. And that is true at all speeds, meaning it’s not simply that pitchers with very little can’t go any lower without getting into trouble, it is just easier for a fireballer to take something off his fastball. Here’s a graph for the entire league showing the correlation between fastball velocity and how much slower a pitcher throws his change-up.


Now to look at movement. Using the same method as above, here’s the effect of drop, in inches, on the change-up (compared to the drop on each pitcher’s normal four-seam fastball):


Of course it’s no surprise that more downward movement is going to work in the pitcher’s favor, but the effect is much less than the change in speeds, and has diminishing returns once they really start hitting the dirt. And now for horizontal movement, with positive numbers being away from the batter and negative being toward him:


I suspect that the very pitcher-friendly results for the pitches moving more than two inches toward the batter are actually cut fastballs, but otherwise it seems that unlike drop or a change of speeds, horizontal movement isn’t very important except for more than five inches, at which point the pitch is moving almost as much (in the opposite direction) as a normal slider.

Much of the above is either confirming traditional wisdom or common sense; despite impressive movement on some change-ups, the most important thing is disrupting a hitter’s timing. But we’ve also learned that:

—In addition to hitters having to start earlier against a power pitcher to get around on his fastball, they are also going to have to make more of an adjustment to his change-up, which is the most important part of the pitch.

—The average change of speeds on a change is just 8 mph. Almost no pitchers have more than a 15 mph difference between their change-up and four-seam fastball, and the ones that do don’t get any particular advantage from it.

—Movement helps, but only if you can get above-average action on the ball, and not as much as the change in speeds. The average pitcher would benefit more from throwing his change-up slower (with the same arm speed) than improving either the tail or drop on it.

A Hardball Times Update
Goodbye for now.

Comments are closed.