Let me state at the outset that there’s no conclusion to be had in this post, I’m just thinking out loud here.
A friend and I were talking during Sunday night’s game between the White Sox and the Rangers. Steve Phillips had just said that the Rangers had made a mistake trading a bunch of good pitchers for hitters, because they were already the best offensive team in the AL, but they still were a losing team. Joe Morgan responded by saying that a hitter plays for you every day, a starting pitcher can’t win more than one out of five games for you.
My friend was saying how Phillips way outclassed Morgan in that booth. I was surprised to hear that because I think Phillips is even worse than Morgan, and that in this case, Morgan is right. A position player plays almost every game. A pitcher pitches maybe 7 innings every five games.
The more I think about it, the more sure I am that position players are more important than pitchers, at least in the regular season. The playing time imbalance might be part of the reason, but maybe there’s something else. Conveniently, there was a post that was sort of related on Beyond the Box Score a few days ago. Basically, of the top 100 players ever, in wins about replacement, 73 are hitters and 27 are pitchers. So pitchers tend to accumulate less value over their careers. Now, this could be because of any number of things – pitcher careers could be shorter, too, so I don’t want to give the impression that this proves anything, just that it reinforced my view a little.
But I wonder if one of the reasons hitters might be more valuable is that hitting isn’t bounded the way pitching is. A pitcher can’t allow fewer than zero runs. On the other hand, there’s no theoretical maximum. You can win a game 15 to 14, but you can’t win 0 to -1. In other words, at some point, better pitching simply doesn’t make your team any better – if you give up two runs a game but still have a losing record, you’re probably not going to be that much better off only giving up one run a game – you have to score runs to win.
This is probably not a particularly original line of thought – I would assume the people who do serious statistical analysis of baseball have already considered and either rejected or accepted it. But it’s interesting to me nonetheless.