1871 - 2008

Trying to Get the "Lay of the Land" Regarding the History of Our National Pasttime

Michael Round

July 22, 2009

Players come and go - big and small - old and young - and it's easy to lose track of what has taken place.  Let's play with the data and see if we can get a glimpse of "the numbers behind the numbers".

For example, Randy Johnson, southpaw, is 6'10".  We think this is unusual, because there aren't any other pitchers with this height.

What about in the history of the game?

Of course, looking at statistics requires a number of assumptions.  Players weight changes over their careers.  In the beginning, player height / weight was not recorded.

But let's take a look at what we do have:

You see Johnson's physical characteristics are indeed unique.  And to check the credibility of the database, we see Eddie Gaedel's statistics in the upper-left corner!  Here is the exhibit from Cooperstown with the two standing side-by-side.  Isaac and I will be there in a few days!

Thinking of the history of the game, how have pitchers changed over time - right-handed vs. left-handed?  To answer the question requires a bit of an assumption: how should a right-handed pitcher with 300 innings-pitched in a season be compared with a left-handed hurler throwing 5 innings?  Merely counting pitchers, they're both equal, though we, knowing a bit about the game, know they're not.

Therefore, let's quantify our statistic as "distribution of innings pitched by throwing arm".  What does the data look like?

About 70/30, and fairly steady for the past half-century!

What about hitting?  Here, let's use "at-bats" across batter to view the distribution.  Of course, with batting, we have another alternative:  the switch-hitter:

Recognizing left-handed hitters occupy about 30% of the plate appearances, we see the game - structurally - as not changed much over time, pitchers and hitters alike!

These merely quantified at-bats and innings-pitched by right/left/switch-hitter (for batting).  What about performance?  Batting Average?  ERA?  Let's see:

BATTING AVERAGE

By Hitting Stance

We see the batting average "regressing towards the mean", where all three - left, right, and switch-hitters - all bat "about the same".

What about pitching?

PITCHING ERA

By Throwing Arm

Remarkably, pitching ERA has always been about the same, regardless of the throwing arm of the pitcher.

Just playing around.

Let's play around a bit more.  The "steroid-era" produced remarkable home-run output.  McGuire.  Sosa.  Bonds.  The era was tainted.

But was it - really?

Aside from these three players, what was the rest of the league doing?  One thing is striking to begin with, for example:  all three of these players were in the National League.  What was going on in the American League?

Additionally, McGuire and Sosa destroyed Maris' mark in 1998.  Bonds destroyed that mark in 2001.  What was going on in 1999 and 2000?  If steroids was responsible for this output, I'd expect these two years to display as great an output as the other years.  What does the data look like?

Finally, if steroids were as rampant as some people claim, surely the entire top-10 would rise dramatically, not just the performance of the top three. 

Let's answer all of this with a simple graphic:

What happened in 2000?  The league-leader had 50 (National League), but the top-10 was bunched together.  Why is this?

And why is the American League dramatically different from the National League?