I‘m currently in the midst of writing my college thesis (yes, it’s on baseball), and I recently came across a piece of research that sounded my “wow, everything I thought I knew about baseball is wrong” alarm to claxon-like levels. In the 1915 edition of Baseball Magazine (distributed from 1908-1957), there’s an article written by F.C. Lane that would make even Tom Tango take notice (assuming he isn’t already aware of its existence): “Why the System of Batting Averages Should Be Changed: Statistics Lie at the Foundation of Baseball Popularity — Batting Records Are the Favorite — And Yet Batting Records Are Unnecessarily Inaccurate.”
F.C. Lane wanted to compare two players; Jake Daubert (high batting average) and Gavvy Cravath (sub-.300 batting average):
To make this comparison, Lane looks at the league average figures for singles, doubles, triples and home runs (77.44%, 14.80%, 5.51%, and 2.24%, respectively) and compares those numbers to each player’s numbers. Daubert’s hit breakdown was as follows: 79.47% singles, 13.90% doubles, 5.29% triples and 1.33% home runs. “In other words,” explains Lane, “Jake made more singles and fewer extra base hits than the general average right down the line. Jake had a lot of coins in his pockets, but many of them were nickels and dimes.” Cravath, on the other hand, had the following breakdown: 59.38% singles, 20.80% doubles, 4.69% triples and 16.12% home runs. Lane breaks down the numbers further, assigning the proper (his idea of the correct) values to each hit, thus creating a weighted batting average. Comparing a player’s weighted figures to the league averages seems quite a bit similar to what we know as wRC+ today, wouldn’t you say?
Fascinating stuff!
