Edwin Jackson's no-hitter yesterday got me thinking about pitch counts (and the beauty of Baseball Prospectus data). There's obviously a greater focus these days on in-game pitch counts and season inning limits now than there was 20 or 30 years ago, making performances like Jackson's all the more remarkable. So I did some digging through some BPro statistics, and was a bit surprised to find that the average number of pitches per game has stayed virtually the same or actually risen since the late 1980s. What's more intuitive is that the average of maximum pitches thrown per start for all pitchers has decreased steadily over the past two decades. Here is the graph that shows average number of pitches per game, and average maximum number of pitches by a pitcher in a season.
There's more fluctuation for average number of pitches; there's no clear trend in one direction or another. I would figure that with more and more attention being put to the 100 pitch mark, starters of this generation are being asked to throw 100 pitches or so (and not much else) than those of previous generations were. It's possible to test this, at least rudimentarily, by splitting up starts by a defined criterion. This next graph shows the percentage of starts that went a certain amount of pitches.
Most notable is the steady rise in appearances between 101 and 110 pitches, and the steady decline in appearances of more than 111 pitches. So, it seems that the the recognition of "century mark" as being meaningful has had some sort of significant impact on how managers handle their pitchers.