clock menu more-arrow no yes

Filed under:

Visualizing GB/FB Ratios, Split By Batter Handedness

New, 42 comments

In November, I developed a way of visualizing GB/FB ratios as vectors. After thinking about Brad Ziegler's mammoth platoon splits the other day, I had a thought—why not go back to the vector visualization, but split it by batter handedness? Here it is. Click for a much bigger version.

Bbsplits500

Some thoughts and methodology notes after the jump.

 

  • First off, methodology. In this visual, a hypothetical pitcher who generated 100% groundballs would have a horizontal line, and a pitcher who got all flyballs would have a vertical one. Everybody else is in between, obviously. (For the math inclined: the arctangent of the FB/GB ratio gives the angle of each line, with zero as the horizontal.)
  • To avoid cluttering up the chart too much (and since I couldn't find contact numbers split by batter handedness), the length of each line is constant for everybody.
  • It's in the legend, but I'll repeat it here—for each pitcher, the green line corresponds to the GB/FB ratio against left handed batters. Gold is against righties.
  • Lefties are on the top row, and the righties are on the bottom. By organizing it like that, the pattern becomes more evident—pitchers, even the flyball ones, tend to generate more groundballs against batters of the same handedness as themselves, although some pitchers don't seem to exhibit very much difference.
  • The big outlier? Gio Gonzalez. His batted ball splits look like a RHP's, and I'm not sure why. Michael Wuertz is another one, but the I'd argue that the difference between facing lefties and righties for him is small enough to be about the same.
  • Also, note that for all of these pitchers, they have enough of a sample size with each split for the results to be statistically significant. Groundball and flyball rates stabilize far quicker than something like batting average or OPS.