[ Home ] [ Library ] [ Bookstore ] [ Contact ] [ Search ]

Forecasting pitchers: The "K"-rystal ball

Keith Woolner

March 23, 1997

One commonly accept bit of baseball wisdom is that power pitchers -- those who rely the fastball and strikeouts to
earn their keep, survive longer in the majors than do control pitchers (with the possible exception of crafty,
one-batter lefties, who apparently can find a job in the majors until they qualify for Social Security). The repeated
chances given to young fireballers who impress scouts with their velocity is a testament to the love affair with power
pitchers that has persisted for years.

But is there any basis for this belief? Do pitchers with high strikeout rates have any better chances for success
than do comparably performing same-age pitchers with low strikeout rates? I decided to answer the question by
researching the fortunes of such pitchers in the study presented below.

The study compares groups of pitchers with similar ERA's at age 25 who differed by strikeout rate, and seeing how
they did 5 years later. The goal was to determine whether high-strikeout rates pitchers aged better as a group, as is
commonly believed.

In this study, I took pitchers who pitched 150 or more innings at age 25 for a single team (pitchers who split
time with 2 or more clubs at age 25 were excluded for computational simplicity). I compared each pitcher's
strikeout rate to the league's average in that year. Pitchers at or above the league average fell into one
group, those below league average fell into another.

I broke each group into ERA+ cohorts -- that is, all pitchers with similar park and league adjusted ERA's were
grouped together. ERA+ is defined in Total Baseball, and the chart below shows the ranges I used to divide
the groups (assuming a league average ERA of 4.50):

ERA+ 
Range    ERA Range
-------  -----------
 60- 79  5.70 - 7.50
 80- 99  4.50 - 5.70
100-119  3.80 - 4.50
120-139  3.20 - 3.80
140-159  2.80 - 3.20
I totalled the performance for each strikeout group's ERA+ cohort at age 25, and looked at how many of them were still pitching at age 30. I also looked at their age 30 performance -- averaged innings pitched, ERA, and ERA+.
Range : ERA+ Cohort range (the upper and lower bounds on
           ERA+ for that group)
K : Strikeout rate group
    G -> Greater than or equal to league average
    L -> Less than league average
#Pit : The number of pitchers is this group
IP : Averaged number of innings pitched
ERA : Collective Earned Run Average (unadjusted) for the group
ERA+ : Park and league normalized collective ERA
% Active : Percentage of pitchers from this group who
     appeared in the majors at age 30
Data is through 1993, and excludes then-active pitchers who hadn't reached age 30 as of 1993. Minimum 10 pitchers per strikeout rate/ERA+ group.
           Age 25                    Age 30
Range   K  #Pit IP  ERA ERA+         %Active  IP  ERA ERA+
-----------------------------        ----------------------
60-79   G   16 189 4.50   74             38% 210 3.22  116
60-79   L   19 186 5.05   74             37% 131 3.29  116
80-99   G   89 215 3.95   92             62% 159 3.58  111
80-99   L  130 213 4.02   90             52% 129 3.91  100
100-119 G  121 243 3.34  109             61% 168 3.63  105
100-119 L  143 229 3.40  110             55% 167 3.78  105
120-139 G   84 254 2.85  128             75% 177 3.20  117
120-139 L   45 244 2.91  126             69% 167 3.53  108
140-159 G   37 261 2.53  147             76% 195 3.19  121
140-159 L   10 252 2.33  148             70% 180 3.69  102

The comparison between the G/L groups for each ERA+ cohorts are striking in their similarity -- in almost all cases, their ERA's and innings pitched are close enough to be all but equal. High strikeout pitchers tend to pitch a few more innings (the most significant gap being the 14 IP gap among the 100-119 cohort), but otherwise it's hard to distinguish the two groups at age 25 in any way other than strikeout rate.

The look at age 30 shows how the groups have diverged. The percentage of pitchers still active is higher in every cohort, ranging from the negligible difference in the 60-79 cohort to 10 percentage points in the 80-99 cohort, averaging out a consistent gap of 5-7% in favor of the high strikeout pitchers.

What's even more striking is the differences in performance at age 30 among those still active. In every case, the
high strikeout pitchers are throwing more innings -- and the gaps are significant with the exception of the 100-119 cohort. High strikeout pitchers are throwing 20 or so innings more per year than their low-strikeout counterparts. The results are more impressive too. In three of the groups, the high strikeout pitchers are posting adjusted ERA's 10-15% higher than low-strikeout pitchers were. And in the other two cases (60-79 and 100-119) the rates of run prevention were identical.

There is *no* case in any cohort where the high strikeout pitchers were throwing fewer innings, or at a worse rate of performance than low strikeout pitchers were at age 30. And in several instances, there was a distinct trend towards such pitchers being more likely to have a job, to throw more innings, and to post better ERA's.

The combined impact of each of these findings makes it clear that, while not a foolproof indicator by any means, we can make better predictions about how pitchers will age by looking at their strikeout rates when they are young.

[ Home ] [ Library ] [ Bookstore ] [ Contact ] [ Search ]

Last Updated: Contact webmaster@stathead.com for corrections or problems

Copyright 1997-2001 by Keith Woolner. All included authors retain the copyrights to their original works.