This afternoon, we talked about why we chose FIP as the metric to base our pitcher win values on. Now, we turn our attention to the other key aspect involved with understanding how many wins a pitcher was worth – the value of a replacement level pitcher as the baseline.
As we did with position players, we’re defining replacement level production as the expected performance you could get from players who can be acquired for virtually no cost. This pool of players would include free agents who sign minor league contracts or for the league minimum, rule 5 draft picks, guys claimed on waivers, and minor league veterans who can’t shake the “Quad-A” label.
Some walking examples of that group from this off-season would include R.A. Dickey, Clay Hensley, Jason Johnson, Gary Majewski, and Tomo Ohka. Not the most impressive group of guys ever, but that’s why they’ve signed for nothing. They represent a portion of the free talent community, and that’s the group that we want to define as zero value pitchers.
Here’s a good discussion about the historical quality of replacement level pitchers. As Tom notes, it is extremely important to distinguish between roles.
While both involve hurling a ball towards home plate, starting and relieving are still remarkably different. Relievers are, in general, failed starting pitchers who are given an easier task that their skillset will allow them to handle. They are selectively managed to face hitters whom they have the best chance of getting out, and they get to throw at maximum effort on nearly every pitch, giving them greater velocity over their shorter appearances.
Nearly every starting pitcher in baseball could be a useful relief pitcher. Very few relief pitchers could be useful starting pitchers. The distribution of pitching talent is skewed very heavily towards the rotation, and because of this and the extra skills required to pitch 5+ innings per start, we use different replacement levels for starting and relieving in order to capture the additional value added by starting pitchers above and beyond simple run prevention.
What are those replacement levels? Perhaps it’s easiest to understand them in relation to a single game. If we assume that a team has a league average offense, a league average defense, and a league average bullpen, and that they are playing a league average opponent, we would expect them to win any single game started by a replacement level starter 38% of the time.
At the same time, we’d expect a team with a league average offense, league average defense, and a league average starting pitcher, facing a league average opponent, we would expect them to win 47% of any games in which their replacement level bullpen was used.
So, we call a .380 win% the replacement level line for a starting pitcher, and a .470 win% the replacement level line for a relief pitcher. However, league average is different for the AL and NL, thanks to the DH, and of course offensive levels vary from year to year. So, how do we take these replacement level winning percentages and compare them to the RA-scaled FIPs we talked about earlier?
We’ll work through those calculations tomorrow.