BaseballStuff.com | Baseball Scholars Forum |
Dan Levitt's Baseball Page
Speed Scores and Reaching Base on Errors
In a previous essay I examined the
effect of team speed on opposition errors.
This essay expands the analysis by looking at individual players
I examined the effect of speed on
reaching base on error by looking at all 219 batters who had at least
300 at bats in 1980 using the 1980 Retrosheet files.
As a proxy for player speed I used the Bill James Speed Score.
The table below summarizes the data by breaking it into
quintiles. I also looked at
correlations based on the 219 individual player records.
RBoE.GB Opp.GB /Opp.GB
Speed Score Quintile
A very modest correlation of .14 exists between reaching base on
error per opportunity and a player's speed score.
Because nearly all times first base is reached on error comes on
a ground ball, much of this correlation can be explained by the
relationship between speed scores and the propensity to hit ground balls
(correlation = .30), i.e. faster players tend to be ground ball hitters.
When one looks at times reached base on error as a percentage of
non-basehit ground balls the correlation between speed and reaching base
on error evaporates to basically zero (correlation = .04).
As a further inquiry I examined
whether reaching base by error is related to whether a hitter bats
left-handed or right-handed.
RBoE.GB Opp.GB /Opp.GB
The side of the plate the at bat took place from.
Switch hitter at bats are therefore split. All other columns are
the same as above.
Summary: Here it seems fairly
conclusive that those hitting from the right side are much more likely
to reach base on error than those hitting from the left side.
I would suppose righties are more likely to reach base on error
because they are more likely to hit to the SS or 3B who typically make
more errors (mainly due to the longer throw) than first or second
A note on Speed Scores
Bill James developed Speed Scores as
a method for determining a player's practical game condition speed.
It is calculated by averaging the top five of six different
statistics. The scale for
players ranges from approximately one for the slowest players to ten for
the fastest. In my
analysis, for simplicity, I use only five of the statistics (the one not
used is Range Factor).
by Dan Levitt
Comments or Suggestions
Thanks for visiting the BasebaLL Think Factory. For information about contributing, to make a comment or suggestion, or to report a problem at my WWW site, please complete my Feedback Form. If your browser does not support forms, send e-mail to firstname.lastname@example.org.
Original material Copyright © 1996-1999