2. Should the EloRater's Top 5 pitchers all be men who played before World War II ended?

Arguably, yes. The importance of the starting pitcher has been in continual decline for as long as baseball has existed. You could make a well founded argument that the top five pitchers of all time should all be men who played before World War I started.

Although one could argue the opposite too. That starting pitchers pre-WWI were less important than other factors such as quality of fielding given virtually every ball was in play. A good catcher with a strong arm might have been crazy valuable back then with the number of stolen bases occurring, as would be quality middle infielders. First basemen were perceived as high end defensive importance too due to the low quality of gloves (thus easier to have a ball get past them).

Just taking a team at random, the 1890 Cincinnati Reds. Their ace was Billy Rhines at 21 who threw 401 innings over 45 starts and 1 relief appearance. His ERA+ went from 186 that year to 115 to 61 to 51, then the rules changed from underhand to overhand and 60'6" and he missed a year then was 103-189-111-101-64 and done. Crazy jumps there. Checking the 1901 team the leader was Noodles Hahn who was in his 3rd year with a 119 ERA+. His career was from 1899 to 1906, with ERA+'s of 145-112-119-169-141-142-118-78. Clearly above average but one bad year and that was it. But note how he allowed 574 earned runs and 821 runs overall, or an extra 247 runs. Wow. Defense was _big_ back then. If you could get a top quality defender it could change everything back then, far more so than now I suspect.

In 1901 3681 runs were scored in the NL, 5193 overall. 4.63 runs/game vs a 3.32 ERA. Errors and missed plays led to a 39% increase in offense. If you check the top ERA team that year, Pittsburgh (yup, a long time ago), you see a spread for guys with 100 IP from 2.18 ERA to 2.86 (4 starters) - 31% or less than the amount errors would add to offense.

Pitching was important but a lot of effort should go into figuring out just how vital defense was back then. One might find that pitchers with 400+ IP then were not as valuable as guys with 150 today due to how big defense and other factors could've been.

Doesn't Catfish Hunter belong pretty squarely in the outliers section as well? Only three very good seasons, in the HOF because he had a great (constructed) nickname and played for a bunch of good teams. Neither a peak nor career argument.

I seem to recall the Neyer/James pitching guide (which I have at home) claiming that it was almost a different sport during its early years.

(I may be misquoting, so apologies as applicable).

I take the pitching side of early baseball with a huge number of salt grains.

Checking the 1901 team the leader was Noodles Hahn who was in his 3rd year with a 119 ERA+. His career was from 1899 to 1906, with ERA+'s of 145-112-119-169-141-142-118-78. Clearly above average but one bad year and that was it.

Hahn retired because he hurt his arm.

Hahn retired because he hurt his arm.

Well, to be precise, he got released because he hurt his arm, and he retired because nobody else wanted to pick him up. He did play a little semi-pro ball on the side.

I am back with another long post where I screed about the 19th century.

With respect to the EloRater, having looked at the 19th century extensively, to me, having Keefe so far above Clarkson is just as bad, if not worse than Hunter that far above Marichal. People often consider them nearly equal, but Clarkson was much better than Keefe. Clarkson was better just on raw numbers, and he was better at the peripheral skills you could argue. Finally, Keefe had about 3 seasons where his ER/R ratios were really anomalous. In the 19th century it was really hard for my system to dock pitchers for an unusual amount of UER b/c the base rate of ER/R was so low, and Keefe was basically the only elite pitcher I ever docked, and I did it 3 times. The only other pitcher who has that much docking for UER before 1941 (which is where I'm at) is Ned Garvin. Meanwhile, Clarkson to me is the obvious best pitcher up to Nichols and Young.

And yes, fielding was much more important in the 19th century than today, as said in an earlier post. When I rate players, I diminute the earliest pitchers by 55% based on ER/R ratios since only a third of runs were earned, and I dock early fielders by 15% based on lower fpcts. But since the fielding pcts were lower, the relative values of a better fielder are enhanced. My fielding rating (and all other ratings) is based on relative failure rates. In the 19th century, fielding leverage was huge. All of my top fielding ratings are outfielders from the 19th century: Mike Griffin, Ned Hanlon, Paul Hines, Tom York were the best. 3B could also have a huge impact if you had Bob Ferguson or Jerry Denny. 2B/SS were not as leveraged but there were some great individuals like Pfeffer, McPhee, Barnes, Wright who lapped the field every year. C/P there were never consistent winners, and 1B has low leverage obviously but there were a lot of players with nice streaks of 4-5 years as best in the league like Joe Start, Cap Anson, Herman Dehlman, etc. People are turned off by the low fpcts, but I think when people talk about the 19th century they've got to consider fielding a lot more.

I dislike how much writers seem to dismiss the 19th century/DBE and overrate today and especially the 60s-80s. Yes, they are better today but not much better. Yes, a small advantage makes all the difference but when you are comparing across eras you shouldn't just say the old timers would be crushed in today's game out of hand. Barry Bonds' 73 HR season drops to 28 if you overlay the DBE Polo Grounds for his home dingers and extrapolate the home/away split. Had he played in the DBE he would have had less than 300 HR, which would still be amazing but still.

I am going to bring up Griffin in the HOM voting if I can. He was the best defensive player of the 19th century by my system. He fielded 956 when the league average was 916, and his RF was like 2.5 to 2 for the league, or something like that. Just blew the league away every year. I have him as best in the league 5 of his NL years, and I didn't even do AA. I was shocked when I looked at the HOM plaque room and Mike Griffin wasn't there (especially after seeing Hines and Gore in the inaugural vote). He was also a good hitter and had significant SB if I recall correctly, which differentiates him from some of the other top fielders.

As far as pitchers, at the beginning of course they were the most important. You had one pitcher, he pitched all game every game. And he hit, well if he was named Al Spalding or Jim Whitney. And he fielded more than pitchers do now. In the 19th century, my MVPs are mostly pitchers. After that, it's pretty constant. In the old days, you had more innings pitched with less extreme superlativity values. Now, you have less IP with more extreme superlativity values. I think it evens out.

to add to my point about players not being much better today: it pretty much became asymptotic after integration

finally i also want to add that ruffing like pennock had 2 careers one of which was much better and for the yankees.

one could argue the opposite too. That starting pitchers pre-WWI were less important than other factors such as quality of fielding given virtually every ball was in play. A good catcher with a strong arm might have been crazy valuable back then

That's perhaps a closer description of pre-1884 baseball than 1893-1916. When pitchers' deliveries were limited, and catchers played well off the bat (before 1884, IOW), catcher was a supremely valuable defensive position, and pitchers were somewhat interchangeable. By the early 1900s things had reversed. The stardom of some very early catchers, and the more generic nature of the top catchers in the deadball era (contrasted with the stardom of pitchers like Mathewson or Walsh) testifies to that.

I'm comfortable with the notion that single pitchers tend to contribute less and less to a team in a pretty much straight decline since Old Hoss Radbourn. Of course it's not a totally direct path; the 1960s and 70s saw a bit of a relapse.

Doesn't Catfish Hunter belong pretty squarely in the outliers section as well? Only three very good seasons, in the HOF because he had a great (constructed) nickname and played for a bunch of good teams. Neither a peak nor career argument.

I was confused by Hunter and Ruffing not being listed as outliers, but I think he's using ELO as an indicator of popular support, and then explaining the ones who got in through BBWAA despite not being thought of as great pitchers.

So, were Tinker/Evers/Chance really better than previously thought? From what I've read, given the number of unearned runs that scored in those days, a great infield defense could have a huge impact and make the pitchers on the team look crazy good.

Don't we sort of have to track this all along by eras as regards to glove use and technology? According to Wikipedia gloves were ubiquitous by the mid 1890s, though obviously there were many technical advancements to come (the first gloves were literally gloves). A pitcher in the 1880s was in front of a significantly different defense than one in the 1890s, then again in the 1900s and 1910s.

I dislike how much writers seem to dismiss the 19th century/DBE and overrate today and especially the 60s-80s. Yes, they are better today but not much better. Yes, a small advantage makes all the difference but when you are comparing across eras you shouldn't just say the old timers would be crushed in today's game out of hand.

Given the advances in every other sport in the same time frame, I think this is wholly unsupportable. Given the measured improvements in sports like track & field, swimming, cycling, etc., how can you not conclude that a good H.S team would beat the dog crap out of the average 1892 N.L. team?

Given the advances in every other sport in the same time frame, I think this is wholly unsupportable. Given the measured improvements in sports like track & field, swimming, cycling, etc., how can you not conclude that a good H.S team would beat the dog crap out of the average 1892 N.L. team?

Well there are a couple things here. Nutrition and training have clearly improved now versus then. So the raw physical aspects have changed. Of course equipment has also changed, so unless you plan on having each team play with its stuff I think there is more equality than you do - for example do both sides get the video monitoring stuff, or just the modern players - especially since baseball is more about skill than atheletics (which minimizes the impact of the training and nutrition).

More importantly what are you measuring? Baby Babe Ruth growing up in modern times almost certainly would have the skills to be much better than the average "good H.S. team" and thinking otherwise is silly. Removing folks from their context, dropping them in another context (keeping all their upside and none of the other contexts downside) and declaring something like you have is pretty meaningless.

So, were Tinker/Evers/Chance really better than previously thought? From what I've read, given the number of unearned runs that scored in those days, a great infield defense could have a huge impact and make the pitchers on the team look crazy good.

Bill James did a fine job of explaining this way back when he did the historical baseball abstract.

how can you not conclude that a good H.S team would beat the dog crap out of the average 1892 N.L. team?

Extreme hyperbole. I think that's as unlikely as an 1892 team travelling through time and winning the 2013 world series. The level of play has probably increased to the point where only the very best from 1892 would be playing MLB, and only a portion would be able to play professional baseball. But the top professional adults of 1892 should still have no problem beating the crap out of a group of 16-18 year olds. It would be even easier if the high schoolers were required to use the 1892 style gloves.

Well there are a couple things here. Nutrition and training have clearly improved now versus then. So the raw physical aspects have changed. Of course equipment has also changed, so unless you plan on having each team play with its stuff I think there is more equality than you do - for example do both sides get the video monitoring stuff, or just the modern players - especially since baseball is more about skill than atheletics (which minimizes the impact of the training and nutrition).

More importantly what are you measuring? Baby Babe Ruth growing up in modern times almost certainly would have the skills to be much better than the average "good H.S. team" and thinking otherwise is silly. Removing folks from their context, dropping them in another context (keeping all their upside and none of the other contexts downside) and declaring something like you have is pretty meaningless.

How in the world can you respond to a quality-of-play comparison wanting to start by equalizing the quality of play? Better nutrition, training, coaching, bigger participatory segment of a bigger population, strategic evolution, technique evolution all make todays players better at baseball. That's the point. The equipment factor does have to be considered, because you'd be mentally throwing them out there together and letting them hash it out, but the other stuff is what causes the improvement over time that occurred in every other sport along the same time frame. (Tried to find a track/swimming example and yards/meters makes it hard. Found the women's world long-jump record from 1922 - 30 later than we're talking about, mind you. Would not have qualified for the 2012 Michigan High School state meet. Not placed - qualified. That's the kind of athletic advancement the post to which I responded was trying to hand-wave away with "not much better".)

How in the world can you respond to a quality-of-play comparison wanting to start by equalizing the quality of play?

Because I am pointing out a fundemental flaw in this sort of exercise. There is a whole lot that goes into quality of play and pretending you could pick up one team out of context, ignore everything but the raw numbers, and proclaim superiority might be true according to the hypotheticals of the exercise, but that doesn't make the exercise any less silly.

(Tried to find a track/swimming example and yards/meters makes it hard. Found the women's world long-jump record from 1922 - 30 later than we're talking about, mind you. Would not have qualified for the 2012 Michigan High School state meet. Not placed - qualified. That's the kind of athletic advancement the post to which I responded was trying to hand-wave away with "not much better".)

You realize equipment has a huge impact on this right? Shoes and quality of the track are just two factors. If Jesse Owens had had modern shoes, modern tracks (as oppossed to the ash crap he ran on), modern starting gates and so on he would have run much faster and thus been superior to himself. Sure I guess, but so what?

Found the women's world long-jump record from 1922 - 30 later than we're talking about, mind you. Would not have qualified for the 2012 Michigan High School state meet. Not placed - qualified. That's the kind of athletic advancement the post to which I responded was trying to hand-wave away with "not much better".)

Because 1922 was such a boom time for women's athletics

Because 1922 was such a boom time for women's athletics

Title IX has been a huge boon for Women's sports. And Men's sports seem to be doing fine as well (depending on your opinion of college atheletics I suppose).

EDIT: There might be more Women doing sports (organized and trained) in Michigan this year than there were world wide in 1922.