Position Players by WAR: 19th Century

Baseball Prehistory | Deadball Era | Liveball Era | Post-War
Expansion | Free Agency | Modern Era

We’re pleased to welcome Joshua Maciel to the site, and today marks his first post here as an author. Josh designed the original graphic that eventually turned into WARGraphs, and we look forward to more of his interesting visualizations going forward.

Admittedly, I don’t know a lot about players from long ago, save the ones who are famous for one reason or another (I hope I’m not alone). If they weren’t in Ken Burns’ “Baseball” documentary, then I probably don’t know much much about them. Sometimes I’ll see an article like this one by Steve Treder that brings an older player to light. And sometimes, usually around this time of the year, someone will talk about Hall of Fame snubs, like in this article by Adam Darowski about the best players not in the Hall of Fame. But for the most part, I’m willfully ignorant.

Looking at all the players in baseball is a daunting task. There are a lot of them. There are a lot of eras. How do you judge people who played back in the 19th century when you are only familiar with the players of more modern eras? Even if I look at the player page for someone like Cupid Childs, what can I really figure out? He is totally removed from context; how can I judge a guy who played so long ago against his peers if I’m not familiar with them? I get lost easily in the numbers. I can look and stare at stats until the cows come home, but not get anywhere because of the lack of context.

Over the next several weeks, I want to take a look at the top 500 position players of all time by career Wins Above Replacement. I have split them in to seven different eras:

  • Baseball Prehistory (1871-1900)
  • Deadball Era (1900-1920)
  • Liveball Era (1921-1945)
  • Post-war Era (1945-1960)
  • Expansion Era (1961-1976)
  • Free Agent Era (1976-1995)
  • Modern Era (1995-Present)

Here are the eras shown by runs scored per game:

From 1871-1900, baseball was a wild and crazy changing game. There were five different leagues: National Association (which evolved into the National League in 1876), the American Association (which evolved into the American League in 1901), the Union Association, and the Players League. The number of games in each season varied wildly by league and by team, and the whole era is just a giant mess to try to sift through. There are eleven Hall of Fame position players from that era, and the only one I’ve ever heard of is Cap Anson.

In addition to the eleven Hall of Famers, there are twenty other position players in the top 500 for Wins Above Replacement. The following chart shows all 31 players and their results by age. The darker the square, the more WAR they received that season. I intentionally left out total WAR for each player because the goal isn’t to rank these players, but to see how they matched up to each other, in the context of the great position players of their day.

I edited the images when I updated the methodology, the old image is available here.

AboveĀ is what the best position players of the 19th century looked like. If you had a Hall of Fame vote, which 10 players would you put on your ballot? In many ways, this is a lot easier than later eras. There was no World Series (started from 1903), no MVP Award (started from 1931), no All-Star Game (started from 1933), and no Gold Glove (started from 1957). All we have to go on is the record they left on the field. So let your eyes wander, come to your own conclusions, and let me know if you find anything.

Links to Player Pages (Career WAR in Parentheses):

Cap Anson (88.7) George Davis (86.4)
Roger Connor (86.3) Dan Brouthers (80.1)
Bill Dahlen (80) Jesse Burkett (72.8)
Ed Delahanty (71.7) Billy Hamilton (68.9)
Jake Beckley (59.6) Willie Keeler (59.3)
Jack Glasscock (58.7) Joe Kelley (57.5)
Jim O’Rourke (52.2) Bid McPhee (51.4)
John McGraw (50.7) Cupid Childs (48.4)
Harry Stovey (47.7) Buck Ewing (47.5)
Hugh Duffy (46.5) Paul Hines (45.4)
King Kelly (45) Hughie Jennings (44.9)
Lave Cross (44.5) Hardy Richardson (43.7)
Jimmy Ryan (43.4) George Van Haltren (43.2)
Mike Griffin (43.2) Mike Tiernan (42.9)
George Gore (42.8) Deacon White (42.4)
Pete Browning (41.6) Sam Thompson (41.2)
Ed McKean (40.2) Charlie Bennett (39.6)
John Ward (39.1) Kip Selbach (38.8)
Herman Long (37.6) Fred Dunlap (37.6)
Ned Williamson (37.3) Dummy Hoy (35.2)
Billy Nash (33.8) Bill Joyce (33.7)


Print This Post

I'm an expat living in Japan since 2003, doing sales and marketing work. More of my work is available on Henkakyuu, my personal blog. Also feel free to inspire me to use twitter more often @henkakyuu

47 Responses to “Position Players by WAR: 19th Century”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Blue says:

    To my eye, it looks like O’Rourke isn’t at the same level as the other inductees.

    Williamson, Richardson, Tovey look like early guys who might have gotten missed. Glasscock looks like a clear snub.

    Vote -1 Vote +1

  2. Patrick says:

    Glasscock may be the best name of all time. Should be in the hall-of-fame for his name alone, since I don’t think anyone in recent history could get away with that stitched on their back.

    Vote -1 Vote +1

    • Griggs says:

      Yeah Glasscock probably got shafted. I would have to check his facial hair to be sure though.

      I’m looking forward to the series Joshua, thanks!

      Vote -1 Vote +1

  3. Rich says:

    FWIW, the American Association and American League aren’t related at all other than similar names. American League was originally called the Western League.

    Vote -1 Vote +1

  4. J.Ro says:

    The thing that immediately jumps out to me is Roger Connor’s regular 4-year pattern — good, great, good, off. Maybe he was also a politician, going through campaign cycles.

    Vote -1 Vote +1

    • Joshua Maciel says:

      That is weird. I looked at his wikipedia article, but don’t see any explanation. Anyone know what the deal is with his weird career curve? I also wonder if anyone knows what the deal is with Dan Brouthers and his lone season after years of retirement…

      Vote -1 Vote +1

      • jordan says:

        The best I could find about Brouthers’ 2 games in 1904 is that he hadn’t retired- he was playing in various minor leagues since being cut in 1896 and had organized the Hudson River League, where he hit .385 (ny times obit) or .373 (wikipedia) in 1904. His obit states that, but is missing a couple of years. He was also hired around this time Polo Grounds, as press gate manager (wikipedia) or night watchman (obit), and was lifelong friends with John McGraw. Speculate those 2 games in 1904 as 2nd chance or favor?

        Vote -1 Vote +1

  5. Chad says:

    I notice that SBs were not counted between ’76 and ’85 (and no CS stat for the entire period). Are SB/CS included for these older WAR calculations? If it is, I wonder how much we might be relatively undervaluing some of those early base-stealers (Paul Hines and Hardy Richard both look they had some speed toward the end of their careers, once the SBs were counted).

    How is fielding included in WAR for these players? Herman Long and Glassock both seem to be earning a significant amount of value from their gloves at SS, based on “TZ” (no UZR of course). (despite ~70 errors a year! wonder if those were counted differently back then…).

    Wish I were more enlightened on some of these details, perhaps someone else can illuminate us?

    Vote -1 Vote +1

  6. steve mc says:

    Is there a reason for Sam Thompson left off the list?

    Vote -1 Vote +1

    • Joshua Maciel says:

      He is in the next segment. When careers shifted from one era to another, I had to deal with them somehow, so I just made the cutoff at anyone who finished their career after 1905. Sam Thompson is 1906, so he’ll be in the Deadball Era post next. You’ll see similar stuff for players around 1920, 1945, 1976, and 1995.

      Vote -1 Vote +1

      • Just wanted to point out that although Thompson’s career did end in 1906, he didn’t play at all from 1899-1905. So, other than the 8 games he played in ’06, he was virtually entirely a 19th-century player (1,402 of 1,410 games.)

        Vote -1 Vote +1

  7. Rusty says:

    Nice work. I liked this piece and encourage you in the analysis that you’re doing here.

    BTW, you only recognize Cap Anson’s name? Ken Burns made a pretty big deal about King Kelly.

    Vote -1 Vote +1

    • Joshua Maciel says:

      Thanks for the kind comments, but to be honest, I don’t think there’s that much analysis involved. I’m just showing the data in a different format, without really sharing any conclusions (though I’ve learned a lot in the process).

      And I’m ashamed to say I didn’t remember King Kelly that much. It’s been several years since I saw the documentary, and clearly some names just didn’t stick. I guess I need to find a blog that focuses on older players or something.

      Vote -1 Vote +1

  8. Adam says:

    My list of prehistory players for the HOF would start with the best two players in the first six years of organized league play: Ross Barnes and A.G. Spalding. Neither is in the HOF as a player (Spalding is in as a pioneer as he later founded the company that bore his name), which isn’t surprising since MLB does not recognize the National Association as a major league. I say screw MLB and put the dominant hitter and pitcher for the Nation Assocation in.

    Vote -1 Vote +1

    • Joshua Maciel says:

      Barnes has only 30 WAR because there are no recorded stats prior to 1871 that I had access to. So he just missed the cut of the top 500 position players by WAR. The cutoff is at 33.6 WAR.

      Spalding only had 8.1 WAR (again, from 1871), and pitching WAR is separate. Even with pitching WAR added in, he’d be under the cutoff due to his years starting prior to 1871.

      Unfortunately, these sorts of exceptions happen because I just don’t know that much about the era, and went with a list to start, and visualized it so that I could get a better understanding. That means some people will fall through the cracks.

      Thank you for the info though — I may learn something about these old-timers yet!

      Vote -1 Vote +1

      • Andross says:

        On a related point, it might be a bit unfair to players of this era by referring to them as either position players or pitchers, since it wasn’t uncommon for them to fill both roles.

        I’m not sure if it actually makes a difference in the calculations, but I just thought I’d throw it out there.

        Vote -1 Vote +1

      • Joshua Maciel says:

        There are a lot of problems with 19th century baseball analysis, and the pitching/hitting thing is one of them. I don’t know of any site that actually adds them together for each player, so I would have to do the work manually, which essentially means calculating them and adding them together for all players.

        I’ll try to figure out if there’s an easier way to do it, and if so we can have our 19th century pitcher/hitter article where I go on about that. But only if I can find the data. I should at least be able to figure out how frequent it was.

        Vote -1 Vote +1

      • Adam says:

        I wasn’t really looking at their performance before 1871, as I don’t know anything about it. As for Barnes, yes, he did not play long, but his peak was amazing; he was basically the best hitter in baseball for six years (1871-76) then fell off of a cliff due to injuries and rule changes. He was only 21 in 1871 so I doubt he had much performance before 1871.

        Spalding shows Andross’s point. His 8.1 WAR was as a hitter; even though almost exclusively a pitcher, he hitting was good enough to produce 8.1 WAR in 6 seasons. As a pitcher, fangraphs doesn’t track his WAR, but he did compile a 253-65 record (still the record high winning percentage but MLB doesn’t recognize it due to being NA), 2.14 ERA, and 2.93 FIP over 2890.2 IP. That has to be worth a ton of WAR.

        Basically, Barnes is borderline as a huge peak but iffy career stats (and given how short the seasons were then, he just didn’t get that many PA’s). Spalding should be a shoe-in, but MLB doesn’t count NA stats and I don’t know if he is even eligible for enshrinement as a player since he has already been enshrined as a pioneer.

        Vote -1 Vote +1

  9. Mr Punch says:

    A lot of the HOF selection is, well, fame, along with personal connections and reputation (the veterans’ committee). Anson (first to 3000 hits), Kelly, Delahanty, Ewing, Burkett, maybe Brouthers, were huge stars. Monte Ward, in multiple roles, was one of the most important figures in early baseball history. Billy Hamilton’s numbers are just ridiculous in a short career. Connor finally made it (1976) because he was the top HR hitter before Ruth. I don’t see Glasscock as better than, or as good as, O’Rourke; but in any case O’Rourke was apparently popular and respected.

    This is one of the better (more consistent) eras in terms of HOF representation, I’d think.

    Vote -1 Vote +1

  10. SF 55 for life says:

    With 10 votes I would go:
    Cap Anson, Roger Connor, Buck Ewing, Billy Hamilton, Ed Delahanty, Dan Brouthers, King Kelly, Jack Glasscock, Deacon White, and Cupid Childs.

    BTW, this was a really great article and I can’t wait to see how the rest unfolds.

    Vote -1 Vote +1

  11. Mike Rogers says:

    Cap Anson was kind of a Babe Ruth before Babe Ruth in how much he separated himself from his peers. Anson could’ve retired after 1889 with 19 seasons under his belt, he would’ve had 76.3 rWAR (baseball-reference as a source since I have that data handy) and no one else through 1889 in their careers had more than 54.4 (Roger Connor). That’s quite the feat considering from 1871-1879 he averaged only 295 PA’s due to the shorter seasons. Absurdly good.

    I really liked Deacon White’s numbers because he was a very good all-around player. Very good hitter, but one of the best defenders of the era and above-average on the bases according to B-Ref WAR, as well.

    Dan Brouthers was 6-foot-2, and around 215-pounds which I’d guess was what would be the equivalent of when Frank Thomas broke on the scene — huge dude playing baseball compared to the average citizen. He was one of the best offensive players of the decade.

    Also, Brouthers and Roger Connor are kind of a 19th century Frank Thomas-Jeff Bagwell in that they have a few weird connections. Born around 6th months apart, I think, finished near one-another in WAR, both big time sluggers for the era and bigger-than-average bulks and played first base.

    Thomas and Bagwell, of course, were born on the same day of the same year, nearly equal in B-Ref WAR, perhaps the best first basemen of their eras, etc etc.

    Love this stuff.

    Vote -1 Vote +1

    • SF 55 for life says:

      Cap Anson pretty much established the segregated nature of baseball. Crazy that his intolerance towards other races kept the game separated for over 60 years.

      Vote -1 Vote +1

      • baycommuter says:

        If Anson hadn’t, someone else would have within a few years, given the era (Jim Crow really came into force in the 1890s). Ty Cobb certainly wouldn’t have played with blacks.

        Vote -1 Vote +1

  12. For a great book about baseball in the late 19th century, I recommend Burt Solomon’s Where They Ain’t, about Ned Hanlon’s Baltimore Orioles, one of the first baseball dynasties.

    Vote -1 Vote +1

  13. KJOK says:

    One thing you’ll need to adjust for in the 19th century is length of season. Teams thru the early 1880’s played 80 game seasons or less, then they bumped up to arouns 140 games thru about 1903.

    Vote -1 Vote +1

  14. Josh Gibson says:

    What constitutes replacement level for WAR when potentially some of the best players in baseball were excluded? Is it lowered, thus making a good player artificially higher in value?

    Vote -1 Vote +1

    • Joshua Maciel says:

      Do you mean because non-white players were excluded (from 1890), or because the pre-1871 players stats weren’t included?

      Vote -1 Vote +1

  15. bcp33bosox says:

    Cool article and welcome to fangraphs, I am relatively new as well. I have been reading articles for a couple months now (Chad Finn and his readers had mentioned you guys quite a bit), but I just made my account this week to start posting.

    My comment, maybe I am in the minority here and I know you don’t want to list the totals to discourage straight ranking of players, but I have trouble differentiating the shades of blue, i.e. 2-4 and 4-6 or 4-6 and 6-8. Obviously if they are all lined up in a row in descending or ascending order I can distinguish them just fine, but where they are so scattered it makes it a bit difficult for me (though it should be known I have never been able to do a stereogram, I *could* however, surprisingly always find Waldo…lol).

    Were different colors objected to, due to not wanting us to rank them? If completely different colors are out, would it be possible to at least use a wider range of color to make each value a bit more visible or would a different color allow for more definitive shading?

    Also, I assume less than 2 (10 is amazing just about any way you look at it. Plus it seems to be, at least in this group of names, much less likely than a <2, so the high end is fine. However, a 1.9, though not very good, is still helping the team win and therefore is better than any negative WAR even if it is -0.1. Stiil, with the intended ambiguity there has to be some cut off point, but if all negatives at least had their own value/color it *might* make for an interesting for argument sake.

    Vote -1 Vote +1

  16. bcp33bosox says:

    Not sure what happened to the last part of my post, but it should be:

    Also, I assume less than 2 (10 is amazing just about any way you look at it. Plus it seems to be, at least in this group of names, much less likely than a <2, so the high end is fine. However, a 1.9, though not very good, is still helping the team win and therefore is better than any negative WAR even if it is -0.1. Stiil, with the intended ambiguity there has to be some cut off point, but if all negatives at least had their own value/color it *might* make for an interesting for argument sake.

    Vote -1 Vote +1

  17. bcp33bosox says:

    Also, I assume less than 2 could also be negative WAR…which might not be so common with the all-time 500, but it would be interesting to see which players actually had seasons of negative WARs. Which again, maybe I am in the minority, but to me, seems much worse than just having a low positive WAR.

    Obviously, not listing the exact numbers and totals, some vagueness is intended. On the other hand, my reasoning is, >10 is amazing just about any way you look at it. Plus it seems to be, at least in this group of names, much less likely than a <2, so the high end is fine. However, a 1.9, though not very good, is still helping the team win and therefore is better than any negative WAR even if it is -0.1. Stiil, with the intended ambiguity there has to be some cut off point, but if all negatives at least had their own value/color it *might* make for an interesting for argument sake.

    Vote -1 Vote +1

    • Joshua Maciel says:

      I could color negatives in a similar shade (it’s just a little extra work), and I understand that the shades are tough to tell apart, but my general reasoning was that I didn’t want the negative seasons to stand out too much (even a -0.1 in red would be an eyecatcher), and anything below 2 WAR is below average anyway.

      I can use any colors I want, but were I to put yellow for 2-4, and green for 4-6 and whatnot, it would become an eyesore, and it would actually (believe it or not) become harder to determine which player was better. The human eye is not great at distinguishing too many shades of color, but if the colors are different, we start adding up the different colors rather than taking it all in and making a conclusion.

      If you are really interested in seeing one that is colored, just let me know how you want it done and I will make you a quick sample and you (and everyone else) can let me know what you think.

      Vote -1 Vote +1

      • bcp33bosox says:

        Thanks for the quick reply, I assumed there was some intent with both issues I had. I too, know little about the early eras, so that probably has some to do with it…this one it is basically a complete blank slate as far as my knowledge, so maybe it just seems a bit overwhelming, due to lack of familiarity with the players. A few players certainly stood out, but for me the rest kind of blend together.

        Without making it completely multi colored, would using a wider range of the same color allow you to keep the aesthetic you’re going for, but at the same time make each value a bit more distinctive? For example starting with a light shade of a primary color, blue, and then darken it to a more standard blue and then add yellow making it greenish blue? Or with the way you have it now, would outlining each box with a thin black line possibly allow my eyes to have a reference color that would assist in distinguishing, without becoming an eyesore?

        The one other thing that would be great, would be if somewhere in the article there was a list of the players names linked to their individual fangraphs player page, so for players I am unfamiliar with, I can easily right click them and take a closer look.

        Regardless, I will certainly keep an eye out for the next article in the series!

        Again, I certainly appreciate the personal reply and the fact that you took the time for that is great…thanks

        Vote -1 Vote +1

      • Joshua Maciel says:

        The comments are appreciated — I’m always looking to improve.

        For picking colors, basically there are three things you can change:
        – Brightness (how much black is mixed in)
        – Saturation (how washed out it looks)
        – Hue (what base color it starts with)

        In general, humans aren’t that good at making fine distinctions between shades of colors, so I tried to limit the amount I used. Swapping to another color at a certain point (or mixing something in) may make it easier to distinguish, but making more shades won’t.

        The outline idea is absolute genius. Here is the same chart with the outlines, and what it looks like:

        With the palette shift, this is how it came out:

        Let me know which one you prefer — I think the first is much easier to read, because, as you said, it’s much easier to get an idea of the shade when you have something to compare it to.

        In regards to most players just kind of blending in to each other, it is intentional. Most players don’t stand out, and that’s part of what the exercise is supposed to show — which players from this era really stood heads and shoulders above the other good players of the era.

        I will also add in the players with links to their pages to the main post (and to successive ones too). Thanks again for the comments!

        Vote -1 Vote +1

  18. bcp33bosox says:

    Yeah, now that I see it, the outline is exactly what I was looking for…Thanks for working it out with me. I am actually quite surprised how much better I can *see* the chart. Anyway, as I said, I am happy you took the time…so again, thanks and now I seem to be able to fully appreaciate the chart…

    Vote -1 Vote +1

  19. delv says:

    cool article and great discussion. there are gonna be problems with these analyses, but it’s worth the effort!

    Vote -1 Vote +1

  20. Ryan says:

    For the best analysis of 19th century baseball, and any era, visit:


    Post 93 gives an approximate ranking:


    10 best:
    1 Cy Young
    2 Cap Anson
    3 Kid Nichols
    4 Roger Connor
    5 Dan Brouthers
    6 Ed Delehanty
    7 George Davis
    8 Bill Dahlen
    9 Billy Hamilton
    10 Amos Rusie

    Vote -1 Vote +1

    • Sorry Ryan! I thought I’d included a link to the Hall of Merit in there (I know I do in one of the following installments, and I know I do at the pitcher version on my blog). It’s a great resource, but a little number-intensive for me — like a sabermetric version of the real Hall of Fame voting. I just wanted to keep things simple.

      For anyone who wants to see a lot more information on these players, for most of the top players, there is a thread over at baseball think factory explaining the logic of why a player should/shouldn’t be inducted.

      Again Ryan, sorry for neglecting to link to you guys! Totally not intended.

      Vote -1 Vote +1

  21. Ryan says:

    11 Buck Ewing
    12 John Clarkson
    13 King Kelly
    14 George Wright
    15 Deacon White
    16 Jesse Burkett
    17 Jack Glasscock
    18 Paul Hines
    19 Charley Radbourn
    20 Jim O’Rourke

    Vote -1 Vote +1

  22. Ryan says:

    21 Tim Keefe
    22 Al Spalding
    23 Ross Barnes
    24 Hughie Jennings
    25 John Ward
    26 John McGraw
    27 Willie Keeler
    28 Joe Kelley
    29 Jimmy Collins
    30 Charlie Bennett

    Vote -1 Vote +1

  23. Ryan says:

    31 Ezra Sutton
    32 George Gore
    33 Bid McPhee
    34 Clark Griffith
    35 Joe Start
    36 Dickey Pearce
    37 Pud Galvin
    38 Bob Caruthers
    39 Ed Williamson
    40 Hardy Richardson

    Vote -1 Vote +1

  24. Ryan says:

    41 Cal McVey
    42 Charley Jones
    43 Frank Grant – Negro who played in International Association
    44 Jim McCormick
    45 Jake Beckley
    46 Lip Pike

    Pioneering great:
    Jim Creighton – amazing as a teen, passed away at 21.

    In my estimation, The 46 players I have ranked meet or exceeded the standards for election into the Cooperstown Baseball Hall of Fame.

    Spots 1-10 could fit nicely into a small Hall, consisting of ~100 members.
    Spots 11-21 are better than the average hall of famer, and have no blemishes large enough that they should be omitted from cooperstown.
    Spots 22-38 easily meet the hall standards in my world, but there resumes are not as strong, and or, there is not as much evidence to support greatness (Spalding, Barnes, Start, Pearce)
    Spots 39-46 are more subjective and are more open to debate, but I ultimately find them worthy.

    Players who are close and notable:
    Tommy Bond
    Fred Dunlap
    Davy Force
    Harry Stovey
    Sam Thompson
    Jim Whitney

    Terrible Hall of Fame mistake:
    Tommy McCarthy – arguably the worst player selected.

    Worst Hall of Fame oversight:
    Bill Dahlen – great hitter and fielder of the 1890s and 1900s criminally overlooked. Similar in career value to Robin Yount.

    Vote -1 Vote +1

  25. Ryan says:

    Oops…I failed to mention Cupid Childs…who would slot in at 46, ahead of Lip Pike.

    Vote -1 Vote +1

  26. al baseball says:

    I think John Ward is unfairly scored. He was a star pitcher and than a star infielder. I think the scoring system has trouble dealing with his dual talents. He [I think] is the only player in history with 200 wins & 2,000 hits!]

    Vote -1 Vote +1