Posted by ArlenWilliam on 12/22/2018 5:29:00 PM (view original):
Posted by PennQuaker on 11/25/2018 5:11:00 PM (view original):
It would be even nicer if deadball pitchers didn't shut off most of the HR power and if 1920s-1930s pitchers who have great normalized stats could be as effective in competitive leagues as they should be.
From what I've seen PQ, there is also special favor granted to the better and best deadball pitchers in terms of avoiding extra-base hits, compared to later pitchers. I don't know how it happens, I suppose it's probably based on the entire picture including ERC# -- which of course is also based upon, thus skewed by, home runs given up, in the entire collective mix of stats.
Just play Jake Arrieta, for example, vs. a deadball pitcher that is comparable aside from the home runs; note the number of doubles and triples given up by each.
I agree with this, but don't have the data to back it up. However, let's look at two pitchers with very comparable OAV+, ERA+, HR/9+, and ERC#. Both of these pitchers are starters.
Pitcher 1: OAV+ 123, ERA+ 161, HR/9+ 137, ERC# 2.01, WHIP# 0.98, Salary ~ $8.1 million
Pitcher 2: OAV+ 122, ERA+ 162, HR/9+ 144, ERC# 1.83, WHIP# 0.94, Salary ~ $9.4 million
In my view, these two pitchers should perform very similarly in the sim with pitcher 2 having a slight edge. Certainly the power allowed by these two pitchers in the same environment should be very similar.
However, pitcher 1 ('28 Garland Braxton) tends to get hit hard if used in $100-$120M cap leagues and will give up the long ball. The second pitcher ('02 Bill Bernhard) is usually very effective at those salary caps and is a popular choice on more rosters. He also will tend to shut off the power. According to the performance history, Bernhard has been used in 400 seasons compared to Braxton's 78, which tells me that others recognize the inherent difference in value.
The biggest difference between these two pitchers is the real life environments in which they performed. The '28 AL had a total OPS of .741 with an average team scoring 4.77 runs per game while hitting 1 HR every 110 PA. The '02 AL had a total OPS of .700 with an average team scoring 4.89 runs per game (due to 294.8 errors/team compared to 192.5 in the '28 AL) while hitting 1 HR every 179 PA. Since the sim doesn't take into account park factors for individual player seasons in determining value, I'm not doing so here.
Both pitchers were 4th in FIP with Braxton being 20% better than league average compared to Bernhard's 14%. Bernhard's WHIP was 30% better than average compared to 29% for Braxton. And, Braxton posted 223/264/319 for a .583 OPS on a team with an average of .690 in a league that averaged .741. The park factor at Griffith Stadium was between 99 and 101, so there's no substantial impact there even if the sim considered it in determining player value.
All in all, '28 Braxton and '02 Bernhard should perform very comparably over time in the sim. However, the level of usage for each player tells me that this isn't the case. I've been burned by Braxton in numerous leagues until I learned I wasn't really getting a bargain. I feel strongly that if we want to see the sim do a better job of reflecting baseball as it should be played, valuation across time needs to be addressed. If the dead-ball pitchers are going to shut off power and XBH, then price them accordingly so they don't proliferate on most strong teams. Modern pitchers who give up fewer HR/9 than league average should shut off the power more than dead-ball pitchers whose relative numbers aren't as strong.
Separately, the sim needs to recognize that '13 Walter Johnson isn't a dead-ball version of Robin Roberts or Camilo Pascual. I know that he led the league in HR allowed with 9 (virtually all of which were inside the park), but he clearly shouldn't be giving up all other hits more than other pitchers. It's one of the best seasons ever by an ERA title qualifier and it's very hard to use him at the highest caps because the sim seems to say that his HR/9+ of 57 means that he should give up lots of doubles and triples as well.