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

The "K"-rystal Ball: Batter Strikeout Rates as a Predictor of Future Performance

by Keith Woolner


This study is similar to the study of pitchers I did several months ago, looking at whether strikeout rates in young pitchers told us anything about how they'd fare down the road.I've flipped it around and looked at hitters, comparing 25 year old hitters with similar offensive production who differ by strikeout rate, and looking at how they did at age 30, including survival rates, playing time, and rates of production. The results follow:

Selection criteria:

300+ PA with one team at age 25, during years 1920-1988

Grouping

Legend:
   K: Strikeout category
    G -- Greater than league average
    L -- Less than league average
PRO: Adjusted Production (from Total Baseball) 
      essentially OPS adjusted for park & league (average=100)
#: Number of players in this cohort-subgroup
PA: Plate appearances (AB+BB in this calculation)
AVG: Batting average
OBP: On Base Percentage (H+BB/AB+BB approximation)
SLG: Slugging Percentage
PLAY: Percentage of players in the group still playing at age 30
The data:
|---GROUP---|  |-------AGE 25-------|   |---------AGE 30----------|
K  PRO      #  PA   AVG  OBP  SLG PRO   PLAY  PA  AVG  OBP  SLG PRO 
              
G  40- 59   8  368 .220 .270 .288  54    63% 515 .255 .312 .344  76 
L  40- 59  20  454 .224 .281 .274  52    70% 368 .255 .311 .332  79 
              
G  60- 79  47  494 .236 .299 .329  72    66% 255 .251 .313 .364  83 
L  60- 79  82  512 .248 .300 .323  70    63% 348 .259 .316 .346  85 
              
G  80- 99 104  479 .258 .317 .382  90    63% 348 .257 .328 .383  96 
L  80- 99 188  541 .269 .327 .367  90    75% 365 .271 .332 .378  94 
              
G 100-119 106  530 .268 .338 .429 111    77% 384 .260 .338 .407 106 
L 100-119 182  561 .283 .345 .410 109    84% 427 .281 .349 .409 108 
              
G 120-139  49  603 .289 .360 .481 130    86% 338 .271 .347 .447 117 
L 120-139  95  593 .302 .367 .468 128    91% 480 .287 .360 .453 121 
              
G 140-159  24  651 .305 .384 .536 147    92% 479 .276 .357 .459 127 
L 140-159  39  575 .312 .390 .522 147   100% 474 .290 .372 .464 128 
              
G 160-179   7  562 .296 .391 .554 166   100% 397 .283 .380 .505 153 
L 160-179  12  599 .339 .407 .591 169    83% 538 .285 .362 .467 131 
Results:

The group with lesser strikeout rate at age 25 is more likely to still be playing at age 30 with differentials ranging from 5-12% (except for a 3% deficit in the 60-79 PRO category), averaging 7-8% more likely to be playing at age 30 for the group who was better than average at not striking out. Among those who are playing at age 30, the rates of production (PRO) do not differ significantly. In all PRO grouping, the G/L categories are within 4 PRO points of each other (with the exception of the 160-179 category, which is the smallest size sample in the study).

Playing time at age 30 seems to differ, slightly favoring the group better at avoiding strikeouts. The 4 groups with the largest samples (60-79, 80-99, 100-119, 120-139) showed differences of +93, +17, +43, and +142 PA's per season at age 30 (5% to 42% more PA's). The 160-179 group also showed significant differences (141 PA or 35%), albeit among a much smaller sample. The were no significant differences among the 140-159 group, while the 40-59 group showed a dramatic reverse with those prone to striking out playing more at age 30. Many of the seemingly anomalous results (PLAY totals for 160-179,

PRO for 160-179, PA for 40-59) occur in the extremes of the dataset, and in the two smallest samples considered.

Conclusions:

In general, young players with lower strikeout rates will fare better in the long run than comparably productive players with higher strikeout rates. They are more likely to be in baseball by an average of 7%, and those who are playing tend to have more plate apperances (15-20%). The results are stronger in the average-to-below-average players, probably because good offensive players tend to have easily identifiable hitting value even if they strike out a lot. What's not determinable from this study is whether a manager's perception of strikeouts as a indication of lack of skill hurts high-K players more than comparable low-K players. If a manager has a choice between two comparably performing players, and tends to choose the one with lower strikeout rate, that may tend to artifically depress both the 5 year survival rates and the playing time of high-K players.

[ 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.