P.S. Just a thought: is it possible that one reason for the effectiveness of relief pitchers is that, by the end of the game, the starting players (that is, the hitters who have been playing all game) are getting tired? I’m pretty sure that lots of baseball-statistics experts will know the answer to this.
Much of the effectiveness of relief pitchers comes from the fact that they can try harder on each individual pitch. That increase in effectiveness though is balanced by the tendency to make your best pitching prospects starting pitchers, as a good SP is much more valuable than a good RP. Also RP can be selected for favorable matchups (right handed vs left handed). RH vs LH splits are actually quite material.
I haven’t seen any studies about starting players getting tired. I don’t think it is a big issue, mostly because it just isn’t that tiring to hit once every 2 to 3 innings. You literally get to sit and rest for nearly 50% of the game!
It is well-known that starting pitcher effectiveness decreases more than starting hitter effectiveness over the course of a game. Whether that’s through tiring or hitters ‘figuring out’ how a pitcher is pitching on a given day isn’t clear. Tom Tango has done good work on this. Here is a sample.
The effectiveness of relief pitchers is then a) they are fresh and b) players are (relatively) unprepared for their pitches, in some combination. The tiredness of the starting hitters is less likely to be a factor.
What Tim said.
Patterns of use of starting and relief pitchers have of course changed a lot over the years. I’ve always wondered if you could figure out the optimal usage pattern from the available data. Could be tricky, for various reasons. What’s optimal might well vary from one team to another, one league to another, and over time due to changes in the rules and other game conditions. And despite the fact that usage patterns have changed a lot over the years, many possible usage patterns have never been tried, so figuring out if some as-yet-untried usage pattern would be optimal might involve a lot of risky extrapolation.
A major factor is that the batters can figure out how to time a pitcher after some innings (hitting and watching from the dugout). A new pitcher comes in and that knowledge goes out the window.