The (Lack of A) Conspiracy Against Pitcher Wins

Yesterday, a reader in my chat asked me a question I had no idea how to answer: Are teams increasingly pulling pitchers from games after 4 2/3 innings, even with the lead, in an attempt to cut down on wins and arbitration payouts? Here’s the question in its entirety:
My snap judgment was “probably not.” After thinking about it for a while longer, my answer is still no – but now I have some neat graphs and charts that will hopefully make the point clear. Without further ado, let’s dive into the shape of league-wide starting pitching trends since 1974, the first year in our database of game logs.
In 1974, the concept of a five-inning start existed, but it was almost an insult. More than a quarter of starts went nine or more innings. That’s hard to do, particularly when that’s an impossible feat for a visiting team that trails after the top of the ninth inning. If that’s roughly a quarter of games (it’s not every game the visiting team loses, but road teams lose more than half of the games they play), that means that roughly a third of eligible starts went at least a full nine. That’s downright wild. Here’s a graph of that wildness:
There were a few short starts, even back in the 1970s – 21% of starts went fewer than five innings. More importantly, a pattern we’ll see repeated again and again is immediately evident. Managers like leaving their pitchers in for a whole number of innings. It’s a natural endpoint to the day, mid-inning pitching changes can be tricky, it’s a way of boosting your starter’s confidence – there are plenty of reasons for this to be the case, and I’m not sure which is most true, but that’s just a fact of baseball. Managers like to pull their starters between innings rather than partway through. Read the rest of this entry »