Both Tyler Cowen and Russ Roberts have interesting posts on the rise of perfect games in baseball. Randy Johnson’s perfect game last week was the 17th rule-book defined “perfect game” in the history of major league baseball. How does this compare with past feats? First, there are a few caveats to these numbers. Two of the perfect games occurred prior to the historian-defined “modern era” in which the game was so very different from today’s game. This knocks off two games for a total of 15 perfect games. In my view, baseball is not really today’s game until we reach the post-spitball and Ruth-as-hitter (not pitcher) era of the 1920s, which lops off one more perfect game for comparison. On the other hand, it just so happens that in the era of comparison two pitchers have thrown 9 or more perfect innings (meaning the pitchers pitched perfect games in length equal to those recognized as throwing perfect games). Thus, in my mind there have been 14 perfect 9 inning games thrown by pitchers that are consistent with the rules of today’s game.
Second, what is so odd about the perfect game is that, despite its rarity in very recent times, perfect games have been much more frequent than in the past. The first post-spitter perfect game was in 1922 (Robertson). A full 33 seasons passed before Don Larsen’s game in the 1956 World Series. The next two came eight and nine years later in 1964 (Bunning) and 1965 (Koufax), with another happening in 1968 (Hunter). Thirteen years passed before Barker’s game in 1981,when perfect games started popping up with some regularity. The next perfect games were in 1984(Witt), 1988 (Browning), 1991 (Martinez), 1994 (Rogers; edit see comments), 1998(Wells), 1998 (Cone), and Johnson (2004). From 1981-present, there has been a perfect game almost every three years. Why is this?
My first inclination was that changes in the talent dispersion of players might explain it, but now that I look at the numbers, I doubt it. Roberts thinks it may be due to better fielding from groundskeeping and better gloves. Using errors as quick metric I whipped up this little chart from the Lahman Database, and the numbers seem to support the story.
Although, I think most of the improvements in fielding have to do with better athleticism, improved scouting, and positioning. Additionally, pitchers seem to be doing a better job at keeping the ball out of fielders’ hands by striking out more batters.
Tyler adds that more perfect game capable pitchers are playing, which increases the likelihood of a perfect game occurring. This is especially important given the number of games played has increased. I am sure there are some other reasons as well. Please feel free to suggest some.
Addendum: Jon over at Talking Baseball also has a post on the issue, which I missed. Thanks to Dave for pointing it out.