The Alternate Inaugural Hall of Fame Class

For some inexplicable reason, I’m just now reading Bill James’ The Politics of Glory : How Baseball’s Hall of Fame Really Works for the first time. Early in the book, James claims the original Hall of Fame class of 1936 was supposed to include five stars of the 19th century, in addition to the five “modern” greats who were so honored: Ty Cobb, Walter Johnson, Christy Mathewson, Babe Ruth and Honus Wagner. In fact, James goes on to say that, according to the Spalding Official Base Ball Guide 1936, the 19th-century greats were intended to be the Hall’s first five inductees.

Well, things didn’t quite work out as originally planned, as the old-timers were voted on but left out of the inaugural class, mainly because the two leaders in the voting were named on only 50.6% of ballots. Of course, with 57 names—including 30 eventual Hall of Famers—receiving at least one vote on a total of 78 ballots, it’s not difficult to understand why. That’s why I’ve decided to try and determine who those five would have (or should have) been.

The distinction between would have and should have is an interesting one. Determining the former would probably be the simple exercise of looking at what five old-timers received the most votes in that 1936 election, or were the first to be elected after the first class. But, that wouldn’t be a whole lot of fun now, would it?

Deciding who should have been elected seems to be a more interesting exercise, but I’m going to add one distinction. Because advanced metrics didn’t exist back then—heck, there wasn’t even a baseball encyclopedia—I’m not going to use them. Instead, I’m going to rely solely on traditional statistics, although I may try to consider context where it seems appropriate.

I’m also making an attempt to be realistic here. No one was talking about George Davis and Deacon White, let alone Bill Dahlen and Jack Glasscock, as potential Hall of Famers back in 1936, so I don’t see them as part of this discussion. Although I might add that Glasscock received two votes, and Dahlen and White one each, while Davis—who was elected to the Hall 15 years prior to White—received none.

Cy Young

Young actually finished 8th in the 1936 baseball writers voting and 4th on the Veterans Committee ballot. I suspect his status as a bit of a crossover candidate is the reason why he didn’t fare better. For the purposes of this exercise, I’ve conveniently decided to consider all players who’d been out of the game for at least 25 years at the time of the election.

I don’t think there’s even a question Young (511-316, 2.63 ERA over 7356 IP), the all-time wins and innings leader, was the greatest pitcher of the first 40 years of the game’s existence, so he’s unquestionably the first player I’d induct from that time period.

Cap Anson

Arguably, the three best hitters of the 19th century were all first basemen. But, it would be unrealistic to consider them three of the five best players from that era. Besides, they certainly weren’t thought of that way, as Dan Brouthers and Roger Connor received a combined two votes—both of them going to the former—in the inaugural election. Brouthers would get into the Hall pretty quickly, in 1945, but it would take the electorate another 31 years to come around on Connor, the career home runs leader from 1895 to 1920.

It’s pretty much not debatable that Anson was the best position player of the 19th century and, with the possible exception of Young, the most deserving of inclusion here. Not surprisingly, he was one of the two players who received just over 50% of the vote in 1936, and his Hall call would come just three years later.

Buck Ewing

With apologies to Deacon White (and Adam), Ewing was probably the first great catcher in baseball history, and the greatest of all-time as of 1936, with a still-active Mickey Cochrane probably the only other player who would enter into the latter debate.

Because he played a significant portion of his career when seasons were half as long as they’ve been throughout most of history, and due to the demands of the position in his day, Ewing only caught 636 games. More than a few 19th-century catchers logged more games at the position, but most were the likes of Pop Snyder (.235 BA, .254 OBP, .299 SLG), Deacon McGuire (.278/.341/.372) and Wilbert Robinson (.273/.316/.346).

Not all were weak offensively—Charlie Bennett and Jack Clements were both quite good in that department—but none could compare to Ewing’s overall game. Over 5772 career plate appearances, Ewing’s triple slash line was .303/.351/.456 and his reputation was that of a Gold Glove receiver with an exceptional throwing arm. According to historian John Thorn, “…the snap throw to second from a crouch position started with Buck Ewing, not Pudge Rodriguez.” Also, while there’s no stolen base vs. caught stealing data for most of his career, the one thing we have to go on is Ewing’s career fielding percentage of .931, which was considerably better than the league average (.905) at his quite-demanding position.

Ed Delahanty

Among outfielders, Willie Keeler finished higher in the 1936 voting, and was inducted six years earlier than Delahanty (1939 to 1945), but I don’t think I need sabermetrics to show Delahanty was a better player.

Billy Hamilton (1888-1901)1591758421584073616909120.3440.4550.4320.887
Ed Delahanty (1888-1903)183583892596101146415994550.3460.4110.5050.916
Willie Keeler (1892-1910)2123959429323381017194950.3410.3880.4150.803

I added Billy Hamilton for good measure, but other than stolen bases and the fact he was a center fielder, it’s hard to see that a case could be made for him over Big Ed.

Beyond what the statistics tell us, we know Delahanty was a feared hitter with power, whereas Keeler could probably best be described as “pesky,” albeit to an exaggerated degree. You would think Keeler had a better defensive reputation–the advanced metrics don’t confirm this, but that’s a moot point as far as this post is concerned–but that’s really not the case. Delahanty was considered a great all-around player who was good enough to play center rather than left, while Keeler was a converted infielder who held his own in the outfield, but didn’t necessarily shine.

Of course, most of us probably know how Delahanty’s career (and life) ended, so perhaps his tragic downfall—pun intended—was held against him by the voters.

Kid Nichols
(1890-1901, 1904-1906)

Perhaps the best pitcher of the 19th century if you discount the fact half of Young’s career occurred after 1900, Nichols was passed over by the Hall’s electorate five times (with very minimal support) before finally getting the call in 1949. Fellow 19th-century pitcher Old Hoss Radbourn got in 10 years earlier, so perhaps I should justify my choice.

In fact, let’s take a look at all of the 300-game winners from the 19th century to see if one of them stands out:

Pud Galvin (1875/1879-92)3653100.5412.85646576003.11807107
Kid Nichols (1890-1901/1904-06)3612080.6342.96532465067.11881140
Tim Keefe (1880-1893)3422250.6032.63554395049.22564126
John Clarkson (1882/1884-1894)3281780.6482.81485374536.11978133
Old Hoss Radbourn (1880-1891)3091940.6142.68488354527.11830119
Mickey Welch (1880-1892)3072100.5942.715254148021850113

It’s hard to really find much separation between these guys when looking at the traditional statistics.

Pud Galvin‘s significant edge in innings pitched also gives him an advantage in complete games, shutouts and wins, but he also has clearly the lowest winning percentage of the group.

Tim Keefe tops them all in ERA and strikeouts, but certainly not overwhelmingly, and I’m otherwise not seeing a compelling reason to favor him. The latter sentiment also applies to Mickey Welch and Radbourn, although Old Hoss does have his 678-inning/73 CG/59-win 1884 season going for him. Beyond that campaign, though, the full body of work is a little underwhelming–certainly Hall-worthy, but not inaugural-class-worthy.

I added ERA+ to the chart, although I’m aware voters didn’t have this statistic at their disposal in 1936. But, were they able to recognize that only one guy on this list didn’t pitch in the low run-scoring environment of the 1880s? I’m not sure they were making such comparisons back then, but they were certainly capable. Considering their collective experience seeing any of these guys play was pretty minimal, it seems pretty irresponsible for them not to.

Looking at just the National League, here are unweighted average ERAs, breaking the era out into four smaller sub-eras which reflect significant statistical shifts:

1879-1885: 2.78
1886-1892: 3.48
1893-1900: 4.32
1901-1906: 2.95

Of course, frequent rule changes throughout the era come into play, but the most significant one is the change in the distance between the pitchers’ mound and home plate. In 1893, it moved from 50 feet to its current 60 feet, 6 inches.

More than half of Nichols’ career was during the 1893-1900 stretch in which the league average ERA was well over 4. In fact, the NL’s composite ERA fell below 3 in only one season during his prime (1904). With the exception of John Clarkson, the remaining pitchers played about half or more of their careers in the 1879-1885 period where the league average was almost a quarter run below 3. Additionally, all of them—except Nichols, but including Clarkson—were done by 1893, the year the NL allowed earned runs at a rate of 4.66 per nine innings.

It becomes pretty clear by even this simple analysis taking into account the years each pitcher played, that Kid Nichols was the best 19th-century pitcher not named Cy Young.

19 thoughts on “The Alternate Inaugural Hall of Fame Class

  1. 1

    Great piece, Dan. I’m surprised I haven’t seen an analysis like this before.

    I’m not surprised you found room for the best left fielder of the era, though you may dismiss Sliding Billy a little too easily. A .455 OBP and 912 steals gave him the edge in career wOBA (.433 to .428; if you’ll allow me to break your “existing stats” rule). Throw in the positional adjustment and one might expect Hamilton to out-WAR Delahanty. But wait… Hamilton’s career Rpos was -86, while Delahanty gets away with -76. Might that be because b-r couldn’t glean which outfield position these guys played from box scores and gave each OF position the same adjustment? Ty Cobb’s career Rpos is -96. Anyone know if this is true?

    • 4
      Lawrence Azrin says:

      I’ve read that Sean Forman has said that the defensive ratings on B-R for 19th centuy/early 20th century players have a lot of guesswork in them, and are nowhere as acurate as more modern defensive ratings, and thus shouldn’t be taken too too literally…

      Going that far back, contemporary opinion needs to given a lot of weight.

      Billy Hamilton – very very fast, of course a good not great defensive CFer

      Cobb – very good early, but hurt his arm in a throwing contest and it was never quite the same. Never in Speaker’s class, but good to excellent most of his career.

      One reason I would question his defensive greatness is that Sam Crawford played center Cobb’s first four full years, while Cobb played right. If he were that good, wouldn’t he have displaced Crawford?

      Delahanty – good enough defensively to play 2B and SS regularly his first few years, he was an very good but not great OFer

      BTW,the 2006 Baseball Encyclopedia has complete defnsive outfield position breakdowns for all three of those players.

    • 5

      Thanks Bryan, and you’re right. I was a little too dismissive of Hamilton. Perhaps it’s because when you Google his name, all you get are results for a much-lesser accomplished player of the same name.

      No, seriously. I had it in my mind that I was arguing Delahanty vs. Keeler and for some reason didn’t think I needed to make as strong a case vs. Hamilton. No, I will not allow you to use wOBA for your argument, but I suppose you could have simply used OBP.

      B-Ref does seem to know how many games each outfielder played by specific position, but perhaps they haven’t incorporated them into their Rpos calculations. Honestly, I’ve always assumed the positional adjustments have changed over time, and for some reason, center field was considered equally as important to the corner outfield positions back then. That’s another reason I didn’t consider the fact Hamilton was a center fielder as a huge factor.

    • 7
      baltimorechop says:

      I, too, have wondered about rPos for quite some time. CF seems to vary greatly from Duffy 1884 -7, i think there are some -4 around Max Carey, -1 for Roush, now it’s +2 i think. Adam posted some great links on Hall of Stats (the update earlier this month I posted a similar question), but it didn’t really discuss 19th century rPos.

      You can find a good amount of sway in other positions as well. Look at Catchers like Lombardi, then look at Carter. Or 2B: Doerr gets a ridiculous amount of rPos for 2B.

  2. 2
    Lawrence Azrin says:

    You didn’t state it directly, but a big problem for 19th century players with the first HOF election was structural.

    As you note, there was both a “Baseball Writers” vote, which was not limited to 20th century players, but was mostly 20th century players, and also a Veterans Committee vote, which was indeed limited to 19th century players.

    Problem was, the writers didn’t seem to be instructed which players should be voted _only_ on the 19th century list; a number of players (Delahanty, Collins) got votes on both lists. So, serious candidates such as Cy Young and, Willie Keeler got their votes split. So it’s likely that some writers didn’t vote for 19th century players at all, because they were not sure which list they should be on.

    I agree with your first four likely 19th century HOFers of Young/ Anson/ Ewing/ Delahanty. However,I think Keeler would’ve been elected ahead of Nichols, due to the whole ‘fame’ thing, and also his association with the 1890s Orioles.

    • 6

      Totally agree with your assessment that there was a structural problem with the voting, Lawrence. Like you said, I implied it but didn’t say it directly. Thanks for a little more detail.

      I won’t argue that Keeler would’ve been elected ahead of Nichols because…well, he was elected ahead of Nichols. My goal was to show who should’ve been elected and to show that with traditional statistics.

    • 9
      Lawrence Azrin says:

      @6/Dan M –

      Thanks; indeed, I think that the HOF voting had a very long-running structural problem.

      One of the main things I learned from Bill James’ HOF book was that the whole concept of a “First Ballot HOFer” shouldn’t be taken seriously until 1956 (or later). Since there didn’t seem to be strict enforcement of the five year waiting rule, all-time greats such as Grove (4 appearances), Foxx (7 appearances!), Ott (3 appearances), and Dimaggio (4 appearances) were on the ballot a number of times before being elected.

      Feller and Jackie Robinson in 1962 were the first “first-ballot” HOFers since the original 1936 class.

    • 12
      Lawrence Azrin says:

      @9/addition –

      Phil Rizzuto in 1956 and Warren Spahn in 1958 got HOF votes while they were still active (Spahn wouldn’t retire for seen more years), so maybe the five-year waiting rule wasn’t totally enforced until after 1958. Or maybe there were two stubborn (idiotic?) voters who knew the rules, but still ignored them.

      I can’t take the whole “First Ballot HOFer” thing too seriously when Lou Brock and Dennis Eckersley made it, but Tris Speaker, Yogi Berra, and Pete Alexandeer didn’t.

      • 17
        Ed says:

        Re: the structural voting problem. There are two other, related issues. I’ve posted the link before but I don’t feel like digging it up again. Basically, there was no physical ballot until the mid 50s. Certainly in the very first election, voters were basically voting blind (which is one of the reasons why some players got votes in both the 19th and the 20th century lists).

        A second issue is something Bill James mentioned. Until the first MacMillan Baseball Encyclopedia was published in 1969, there was simply no single source that people could go to to look up players’ stats. It’s not that the stats didn’t exist (though I’m sure in some cases they didn’t) but voters didn’t have easy access to them. So they probably relied on a hodge podge of information to make their decisions.

  3. 3
    Dalton Mack says:

    Really enjoyed this. What I found most interesting, and certainly this wasn’t the aim of your piece, was just how high Tim Keefe’s K/9 was for his era, especially in the context of other great pitchers.

    Not completely sold on Buck Ewing as one of these esteemed five–surely a HOFer, but maybe suited for the Second Alternate Hall of Fame Class.

    • 8

      Thanks Dalton. Keefe ranks 7th in the 19th century in K/9, but is by far the era’s strikeout leader.

      What do you know about Buck Ewing? The guy played his final game like a century before you were born. 🙂

  4. 10
    no statistician but says:

    I agree entirely about Kid Nichols, but, Old Hoss’s achievement in 1884 makes him second among those pitchers, to my doddering mind. Somehow B-ref has Galvin with a higher pitching WAR for the year despite trailing Radbourn in adjusted ERA+ 155 to 205, pitching for a team with an OPS+ of 101 vs Radbourn’s support of 93, and 48-22 to Radbourn’s 59-12. I’m sure there’s a perfectly good explanation statistically for this judgment, but that doesn’t mean it isn’t horseradish.

    In the 3-game 1884 postseason Radbourn was 3-0, all complete games, with an ERA of 0.00, by the way, so he actually won 62 games that year with 76 CGs in 76 starts.

  5. 14
    Hartvig says:

    Doesn’t this all just tie in nicely with the Circle of Greats voting that we are currently doing as well? Am I correct in what I recall that we were going to treat players born before a certain year (188? sticks in my mind for some reason but that seems a little late, it must have been 186?) like the original veterans committee was supposed to and select a certain number of players from those years in addition to the 112 voted in by the BBWAA? I think all but maybe a couple of these guys were born 1867 or before.

    Another thing I noticed when I was checking on birthdays was that about half of these guys didn’t live to see 50 which is kind of a sobering thought when that’s nearly a decade in your rearview mirror.

    • 15
      Hartvig says:

      I realize that I forgot to mention anything about what a great job you did in your article so: Well done Dan. Even though I just reread the Politics of Glory a couple of months ago I’m tempted again to pick it up and reread yet again the parts about the boondoggle with the initial voting.

      • 18

        Thanks Hartvig. Re: the age thing, our life expectancy is much longer than these guys, so please don’t get any (more?) gray hair over the fact they all checked out prematurely (by our standards). 🙂

  6. 16
    John Autin says:

    Nice work, Dan. I’m not surprised that the original voters had trouble; these are hard choices.

    Myself, I would not include Buck Ewing. Maybe the first great catcher, maybe the best catcher of the 19th century. But when you’re taking just 5 players in all, I can’t see Ewing being that much more worthy at that position than some contemporaries who caught more than twice as many games as he did. I mean, Deacon McGuire caught almost 1,000 more games, 1,612-634. Bennett caught 50% more than Ewing, and was a comparable hitter during the 1880s.

    But if not Ewing in this inaugural class, then who? Three of the greatest were first basemen — Anson, Dan Brouthers and Roger Connor. But if we take another 1B, how to pick between the other two? And taking all three … I can’t do it.

    It’s tempting to take just one pitcher, to get more positional balance. But how could you leave off either Young or Nichols?

    Ultimately, I would take Hamilton over Ewing. But after Anson, there are no easy choices.

    • 19

      Thanks John. I can understand why you and Dalton (@3) are both underwhelmed with Ewing. You’re not the only ones. Two of the members of SABR’s 19th Century Overlooked Legends Committee (one of them writes for this site) teamed up on me on Facebook to advocate for Deacon White over Ewing. I guess, ultimately, I didn’t do as good a job convincing you as I did myself that Ewing was the best non-active catcher in baseball history as of 1936.

      If I had to choose five additional guys, they would probably be Connor, Brouthers, Hamilton, John McGraw and John Clarkson.

Leave a Reply

Your email address will not be published. Required fields are marked *