Max Scherzer’s Big Strikeout Numbers
Max Scherzer’s four starts since returning from Toledo have been quite good. Although he has given up a number of runs, he is striking out lots of batters while not giving up a huge number of walks — a recipe for success. Here are his numbers broken down before and after his time in AAA.
Before:
8 starts, 42 innings, 26 Ks (5.6 per 9), 16 BB (3.4 per 9), 9 HRs, 5.89 FIP
After:
4 starts, 24 innings, 33 Ks (12.4 per 9), 10 BB (3.8 per 9), 3 HRs, 3.33 FIP
Scherzer claims to have made some mechanical adjusments in AAA and the results bear this out, as he is throwing his fastball a solid two mph faster (91.8 mph before heading down, 93.9 since). His change is 1.5 mph faster and his slider is the same speed. One would think that given this jump in fastball speed Scherzer’s increased strikeouts would be the result of more swinging strikes against the fastball. Interestingly, this is not really the case.
Swinging Strike Rate
Before After
Fastball 0.053 0.057
Change 0.113 0.191
Slider 0.140 0.229
Breaking and off-speed picthes almost always have a much higher swinging-strike rate than fastball, so that should not be a big surprise. But I am surprsied that Scherzer’s recent jump in strikeouts is largely the result of an increase in swinging strikes on his breaking and off-speed pitches rather than his fastball. I am sure that the work he did on his mechanics improved his slider and change, but it is also possible that the increased swinging-strike rate on those pitches is due to the increase in speed of his fastball. The effect one pitch has on subsequent pitches in an at-bat is complicated, interesting, and not well understood, but Scherzer may offer an interesting case study in which the results from an improvement to one pitch type (increased fastball speed) is seen in other pitch types (change and slider swinging-strike rate).
Dave Allen's other baseball work can be found at Baseball Analysts.
Dave, what do you think is the better metric, swinging strikes/total pitches or swinging strikes/swings? I’ve always preferred the latter, but I’d love to hear what you think.
Lucas,
I am not 100%. I used to use misses per swings exclusively but Jeff Sullivan showed that misses per pitch correlates very well with strikeout rate. I guess the thing to do would be to compare how well each one does at predicting strikeout rate and then go with that one.
Fascinating. I assumed that misses per swings is better since it isolates the variable. Someone (I forget who) explained misses/swings as being a better way of evaluating a particular pitch, and misses/pitches as being a better way of evaluating pitcher performance.