Comparing pitchers from different eras (derived from Koufax vs. Kershaw thread)
Tl;dr: existing pitcher metrics are unsatisfactory.
In the sports memorabilia forum, there is a thread asking whether Kershaw or Koufax is the best Dodger pitcher of all time, though IMO, a very strong case can be made for Dazzy Vance being better than Koufax (still much worse than Kershaw). Anyway, one of the arguments for Koufax' superiority is that he had more complete games in both 1965 and 1966 than Kershaw has so far had in his entire career.
Obviously it is impossible to compare pitchers of different eras using traditional stats. For example, which season was best of Dutch Leonard's 1914, Lefty Grove's 1936, Bob Gibson's 1968, and Pedro Martinez' 2000? And yes, I know that three of the four are Red Sox. All four great seasons, but two are much better than the other two, and using traditional stats won't tell you which.
But while advanced stats do a pretty good job with hitting, and are horrible (but better than gold gloves) when it comes to fielding, pitching has changed so much that advanced stats can't keep up. Some try to solve this by only considering the "modern" (post 1920) era, refusing to evaluate guys like Cy Young, Christy Mathewson, and most of Walter Johnson. Baseball Reference uses something called WAR7Adj which scales each season to a maximum 250 IP which has the effect of reducing, say, John Clarkson from 11th to 29th. This is, of course, not satisfactory because a) it's a pretty arbitrary cutoff, and b) Clarkson really did pitch 623 innings in 1885 (68 complete games! with a 53-16 record and a 1.85 ERA) and it's just not fair to normalize it to, say, Chief Bender's 1910. Both players are then left with a 5.1 (Clarkson 5.17) adjusted WAR. It's probably not right to equate the two seasons, but by how much should Clarkson's season be considered better?
Charlie Radbourne had 108 wins in 1883-4. During that time, Providence only played 220 games. Clearly, baseball was a very different game then. Today (last year) there were only 45 pitchers to even pitch 162 innings, that is the minimum to even qualify for the ERA title. None even came close to the "normalized" 250 innings. Does that mean that in the near future we'll need to recalculate WAR7Adj to reflect the much greater number of innings pitched in the 1980s (109 seasons of 250 or more)?
1884 is a fun year. Only 62 players stepped on the mound in the National League, including some guy who may have been named pitched to five batters, all of whom reached base. 42 pitched more than 9 innings. 8 teams, but Boston and Cleveland used only five pitchers each.
So, how many, or which, of the 19th century pitchers deserve the HoF?
I don't know the answers here. I hope I've at least brought up the question of how to fairly compare not only Koufax to Kershaw, but also Dazzy Vance and even Nap Rucker or Bob Caruthers, neither of whom are under serious consideration for best Dodger pitcher of all time, but who should be considered best of his era. And of course this can, and should, be extended to other teams and the major leagues as a whole.