Baseball for the Thinking Fan

Login | Register | Feedback

You are here > Home > Hall of Merit > Discussion
Hall of Merit
— A Look at Baseball's All-Time Best

Friday, July 11, 2003

Al Spalding

Probably the toughest candidate to get a handle on.

Joey Numbaz (Scruff) Posted: July 11, 2003 at 03:56 PM | 26 comment(s) Login to Bookmark
  Related News:

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

   1. jimd Posted: July 12, 2003 at 01:33 AM (#515432)
No need to repeat my arguments for Spalding from 1904; I'll just link to them here instead: A, B, C.

If Joe's new Pennants Added metric was calculated using a season-length adjusted WARP-1 instead of the season-length adjusted WARP-2 actually used, I would wager that Spalding would be pretty comparable to Barnes. In another words, a first rank HOMer (at least by current 1904 standards).

FWIW, BP awards the 1877 Gold-Glove/Good-Hands Award at 1B to (drum-roll please :-) Al Spalding. The competition: Deacon White, Joe Start, Juice Latham, Dutch Dehlman, and Charlie Gould. Win Shares gives its award to the Juice. Spalding was also the only acceptable fielder Chicago found for 2B during Barnes injury (alternatives: Harry Smith and Paul Hines), though it's only a 13 game sample.

Spalding appears to have been a very versatile fielder; neither BP nor WS give pitchers any ratings for their own fielding, but this circumstantial evidence indicates that Spalding was quite possibly the best fielding pitcher of this generation. The raw fielding stats at pitcher help back that up, with Spalding making 25% more plays, and committing 29% less errors, than the average pitcher of his league.
   2. jimd Posted: July 12, 2003 at 01:58 AM (#515433)
Ward's raw fielding numbers at pitcher a few years later are 31% less errors on 20% more plays.
   3. Marc Posted: July 31, 2003 at 09:09 PM (#515434)
A lot of discussion has occurred in recent months, but way back when (a year or more ago) we had positional threads which included adjWS--adjusted, that is, for length of season--"normalized" to 162 games. These adjWS did NOT include the NA. So I went one step further and added in NA pitcher WINS, based on the fact that we did not have NA WS at the time, and based on my belief (but not just mine, BTW) that "IN A PERFECT WORLD" a pitcher's WS would probably be equal to his wins. This is in fact quoting Bill James.

Then add in one more thing. Since those original adjWS were posted here, the discussion has long since also taken on the issue of pitching vs. fielding, and a strong consensus has emerged that pitching was less important pre-'93 and fielding significantly more important than post-'93. I happen to agree with that consensus. And so I took the adj WS that were posted here + NA pitchings wins and divided the result by 2.

I think sometimes in our discussion we move beyond a particular analysis to additional, also-useful analyses, but sometimes we forget the earlier also-useful analysis. So following are the adj2WS for 19th century pitchers that resulted from the above analysis and formulae. Note that pitchers who pitched both pre- and post-'93 get the second adjustment (divide by 2) for their pre-'93 work only. I include any pitcher who had significant activity pre-'00.

Cy Young 635
   4. Chris Cobb Posted: August 01, 2003 at 05:14 AM (#515435)
Here's a response to Marc. I'd meant to respond to skeptical comments about the high rating of Galvin based on DIPS-influenced pitching measures earlier, but I was busy with the NA stuff. If Marc is taking us back to using WS to evaluate pre-1893 pitchers, it seems important to restate why this is a bad idea. Others may have a better grasp of the full argument than I do and access to more comprehensive data to support the argument, but I'll state it as I see it.

Since Marc's the only voter so far who has left Galvin off his ballot this year, I may be preaching to the choir here, which is why I've stuck this off on the Spalding thread where it won't be in the way, but there've been a number of comments expressing a bit of skepticism about the measures that rate Galvin so highly, so maybe it's worth talking all this through again.

While reducing pre-1893 pitching WS by dividing them in half puts more accurately reflects pitchers' worth in general, it is a highly imprecise measure of the value of any individual pitcher in the period. We know that, in any single game, pitchers pre-1893 had less impact than modern pitchers, but individual pitchers, had more impact than modern pitchers because they were able to pitch so many innings. The pitcher has less to do with each individual team victory, therefore, but accumulates a lot of smaller units of credit. The WS system, however, is crediting the pitcher with 2/3 of the responsibility on the defensive side for each win or loss. So his win shares will reflect the team record more than it reflects his own contribution to the team's wins and losses. With very good teams and very bad teams, this leads to severe inaccuracies.

Pud Galvin's early NL seasons provide a good example:

1880 458.2 IP 10.35 h/9, .63 bb/9, 2.52 K/ 9 --> 14 WS
   5. Marc Posted: August 01, 2003 at 11:15 PM (#515439)
Thanks everybody. I understood Chris' post, though playing the devil's advocate I would point out: WS says Hoss was 27% better, and as Chris admits, the raw data suggests that Hoss was a little better (maybe not 27%, maybe 10%--i.e. .6 fewer baserunners, .6 more Ks, 6 more batting WS). Given all of this I'm not sure a 27% edge for Hoss is any more inaccurate than an 18% edge for Galvin.

That aside, I understand about Galvin's season-to-season WARP numbers displaying a more reasonable consistency than his WS.

Now. I was on vacation last week. You probably missed my nagging desparately. But I didn't see the WARP1 post you mentioned, Joe, and frankly there have been a bunch of posts attempting (sort of) to explain WARP (well, generally they have been designed to argue some player's case and used WARP along the way, but that's fine). But what with WARP1, 2 and 3 and then I've seen adjWARP1 as well as arguments that WARP1 works for pitchers, WARP3 for the rest....well, sorry, but I have not internalized WARP and so therefore I do not use it. I went through all of the definitions, etc., but I still don't really get it. (Am I the biggest idiot ever?) I have not rejected it: I don't get it and therefore I don't use it. Though if there is a HUGE timeline built into WARP3, I reject the timeline.

So, if anybody can start on square 1, please do. I'm willing to learn. And along the way if anybody can explain DERA, let's do that one too. What I think I've heard about DERA so far is that the DERA reflects the quality of the defense, and we know what the quality of the defense is by virtue of the DERA. The one refers to the other with no third, independent variable in the equation? Somebody help.

Finally, the WS were not post as definitive. They were meant to generate a serious response and they have done tht. Thanks again. But, Joe, who formulated them? They occupied weeks and weeks of discussion a year ago and I honestly don't remember anybody saying "never mind."

BTW, Spalding was a star and highly sought after in the late '60s. Otherwise how did H. Wright know to make him the highest paid player in the NA right from opening day '71? He played on the "national" stage as early as Pike did.
   6. Howie Menckel Posted: August 02, 2003 at 03:32 AM (#515441)
Well, the Al Spalding Kool-Aid is being copiously consumed...
   7. Marc Posted: August 03, 2003 at 02:29 PM (#515442)
Joe, this is more revisionist history. Every player was a free agent in 1871.

The fact is that the most highly sought after players throughout the 1860s were pitchers, from Jim Creighton to Al Spalding. This seriously undermines our 21st century notion that pitching couldn't have possibly mattered. We are very quick to note that Bill James' WS formulae don't do much to explain 19th century baseball, and suddenly DIPS comes along as the answer to our preconceptions. What's wrong with this picture?

And as for Kool-Aid, another 21st century assumption.
   8. Howie Menckel Posted: August 03, 2003 at 04:06 PM (#515443)
No insult intended; a poor choice of words.
   9. John (You Can Call Me Grandma) Murphy Posted: August 04, 2003 at 03:22 PM (#515444)
Well, the Al Spalding Kool-Aid is being copiously consumed...

Will Galvin get a spike (other than moving up because of a new opening) now because he'll be considered the best pitcher next "year"? He shouldn't on that basis but I'll guess we'll have to wait and see.
   10. Marc Posted: August 04, 2003 at 04:25 PM (#515447)
Spalding ain't elected yet, in fact I predict that we will continue to fight about Al for several more years. And Pud...well, he will probably leapfrog over Al before it is all done. Probably Mickey Welch too.
   11. Howie Menckel Posted: August 04, 2003 at 05:00 PM (#515448)
Oh, I don't know, Marc. If someone voted for Spalding first and somehow left Sutton off the ballot entirely, you'd be surprised what kind of an impact something like that can make.
   12. Howie Menckel Posted: August 04, 2003 at 05:00 PM (#515449)
Oh, I don't know, Marc. If someone voted for Spalding first and somehow left Sutton off the ballot entirely, you'd be surprised what kind of an impact something like that can make.
   13. Marc Posted: August 04, 2003 at 05:22 PM (#515450)
And if the 3rd ballot from the last failed to include Mr. Spalding that would have no effect.

As to "somehow," I've said from day one that I like a nice high peak, in fact that's a threshhold. First show me the peak, then we'll talk about the rest. And a peak is not "X" number of WS, it is, was this player ever the best player in the game? If there are 15 yes's available, the rest are gonna have a tough time. If there are not 15 yes's available, then there's room for the rest, and Ezra has been among the rest on my ballot on several occasions.

But he's no Joe Start.
   14. sean gilman Posted: August 04, 2003 at 05:43 PM (#515451)
So, say player A had 4 35+ win share seasons, but in those seasons players B, C, D, and E happened to have 1 more WS than him each year. You would then say that player A had no peak?
   15. Marc Posted: August 04, 2003 at 05:48 PM (#515452)
I use a 3 year peak + 5 year peak/2, so, no, sean, I wouldn't say that unless B, C and D were the same player.
   16. sean gilman Posted: August 04, 2003 at 05:52 PM (#515453)
Then I don't understand. . . If you use "3 year peak + 5 year peak/2" then what does this mean?: "First show me the peak, then we'll talk about the rest. And a peak is not "X" number of WS, it is, was this player ever the best player in the game?"
   17. John (You Can Call Me Grandma) Murphy Posted: August 04, 2003 at 05:55 PM (#515454)
But he's no Joe Start.

Dealing with his documented stats, how in Doc Adam's name does Start (who I love) have a better peak than Sutton (and Ezra's 18 years at third is comparable to Start's 27 at first because of the higher attrition level at the "hot corner")? How does Dave Orr? You can't possibly be comparing Sutton to his peers at the position. If your mindset for the position is that of the modern era, your underrating the position pre-Mathews.

With that said, Spalding is a terrific choice.
   18. John (You Can Call Me Grandma) Murphy Posted: August 04, 2003 at 06:08 PM (#515455)
BTW, I hope I didn't sound too harsh, Marc. I wasn't meaning to.
   19. Howie Menckel Posted: August 04, 2003 at 07:15 PM (#515456)
   20. Marc Posted: August 04, 2003 at 07:21 PM (#515457)
At the end of each season (1878-2002) I construct a little list of the top active players (actually 2 lists, one for pitchers and another for position players) in terms of WS over the previous 3 years and WS over the previous 5 years. Then, basically eyeballing the lists and then doing whatever additional statistical analysis seems called for based on who is on the list (e.g. if I have a couple middle infielders I'll look a little further into the defensive numbers or if I have a couple "sluggers" I'll worry about that, etc. etc. WS, in other words, is the start but not the end of this analysis.), I then finalize the list of the best active "peak" players at that time. There is no set number of players who make the list, it will depend more on how closely bunched they are.

That list is a sort of threshold of players who qualify as "great" players by virtue of having been "great" at some particular time and place. If you don't make that list, then you can't really "become great" by hanging around.

Now, not every player who ever made the "great" list ranks ahead of every other player. There are tradeoffs. But so far, and I think probably in the future as well, a player that never made the "great" list is gonna have a hard time getting into my top 10. In another 40 years, the number of players "on the list" will only get larger. But Ezra Sutton did make my ballot a few times, and Bid McPhee even made my top 10 (though I have some remorse about that). But I place a heavy burden of proof as to why a player who didn't make the list of "greats" should rank highly on my ballot.

That's how I do it. In summary, I regard a 3 to 5 year peak as the first and foremost measure of greatness. It is during that peak period that a (merely mortal great) player (i.e. other than the Ruths, Cobbs, Bonds, Mays, etc.) really really really impacts pennant races.

I also believe that when all is said and done, statistical systems are not foolproof and contemporaneous reports by people who actually saw the player(s) in question are also valid data.
   21. Marc Posted: August 04, 2003 at 07:30 PM (#515458)
Howie, I categorically deny making a "strategic" vote. You can check my previous ballots. I have been consistent. And even at the beginning of this "year," who knew that Sutton (the #4 holdover from last "year") and Spalding (#3) would both leapfrog ahead of Galvin and McPhee and end up head-to-head for the top spot. Certainly not me. (If McPhee phinished first and Galvin second, I guess that would be my fault too.)

Further, I don't know what the results are this time, but last year there were 4 voters who did not vote for Spalding and 4 who did not vote for Sutton. Blame them.
   22. Howie Menckel Posted: August 04, 2003 at 07:51 PM (#515460)
   23. Marc Posted: August 04, 2003 at 09:20 PM (#515461)
Howie, you make it sound like 40 people voted for Sutton and 1 for Spalding. That's not accurate. Every single voter who had Spalding 9 places ahead of Sutton swung it the other way. And also, since we are only electing one this year, our system coulda been--vote for 1. In that scenario, Al wins 14-12. Or if our ballots only went x deep, here are your winners.

The most 1sts--Spalding 14
   24. Howie Menckel Posted: August 04, 2003 at 09:31 PM (#515462)
Okay, four other voters 'swung it the other way,' then.
   25. jimd Posted: August 04, 2003 at 09:49 PM (#515463)
Marc doesn't need my defense, but his ballots appear to have been consistent over the years.

FWIW, there were 6 other ballots that could be termed "decisive", three 1-9 splits, a 1-8, and two 2-13 splits in favor of Spalding, and OTOH, there were two 1-7, a 1-8, three 1-9, and a 1-12 split going the other way. Combine that with three ballots where both candidates were 9th or lower and there is a significant block that is not yet convinced of the Merit of either candidate (for that matter, there is similar doubt about every candidate on the 1906 ballot).
   26. Marc Posted: August 05, 2003 at 12:37 AM (#515465)
Joe, there are only so many hours in a day. I don't assimilate everything I see here, no question about that. I've said on numerous occasions why I don't use WARP numbers *at all*. Maybe someday I'll understand them well enough to do so. But as with DERA, I'm not there.

I have moved Joe Start up substantially over the years, fortunately you don't need a Ph.D. in math to get the Joe Start story, a history degree will do (which is what mine is). I never would have voted for Charlie Bennett until I read about him here, I had him 5th last time.

Anyone who thinks they've got 19th century baseball pegged, god bless you. I make no such claims. (But from 1878-1880, I only see Paul Hines among position players as clearly better than Charley Jones. He was indistinguishable from O'Rourke and Kelly, [better than Kelly for the 5 years] and I mean overall, not just as a hitter.)

Nevertheless, I am refining (or at least revisiting my methods and my rankings) as we speak. The HoM certainly has helped me to better articulate and understand in my own mind how (I think) one should go about doing this, so I am constantly refining. A major refine is on the way, but like I said, there are only so many hours in a day.

You must be Registered and Logged In to post comments.



<< Back to main

BBTF Partner

Support BBTF


Thanks to
The Piehole of David Wells
for his generous support.


You must be logged in to view your Bookmarks.

Hot Topics

2016 Hall of Merit Ballot Discussion
(64 - 9:47am, Jan 26)
Last: DL from MN

Most Meritorious Player: 1902 Discussion
(11 - 9:41am, Jan 26)
Last: DL from MN

Vic Willis & Sam Leever
(58 - 2:28pm, Jan 25)
Last: Kiko Sakata

Ernie Banks
(71 - 8:32am, Jan 24)
Last: djrelays

Trevor Hoffman and Billy Wagner
(77 - 2:22am, Jan 23)
Last: Rob_Wood

Rube Waddell
(30 - 5:45pm, Jan 21)
Last: W. G. Braund

Most Meritorious Player: 1901 Ballot
(8 - 1:15pm, Jan 20)
Last: Michael J. Binkley's anxiety closet

Tommy John
(16 - 11:03am, Jan 20)
Last: DL from MN

Toby Harrah
(14 - 11:01pm, Jan 18)
Last: Howie Menckel

Most Meritorious Player: 1901 Discussion
(39 - 3:36pm, Jan 18)
Last: TomH

Ben Taylor
(87 - 11:44am, Jan 17)
Last: Chris Cobb

2015 Election Results - Randy Johnson, Pedro Martinez & Curt Schilling elected to the Hall of Merit!
(31 - 11:20am, Jan 17)
Last: Mike Webber

Cristóbal Torriente
(86 - 7:54am, Jan 16)
Last: Chris Cobb

Ken Griffey Jr.
(48 - 1:30pm, Jan 15)
Last: Jeltzandini

The Baseball Hall of Merit Plaque Room: Home Page
(1464 - 3:45pm, Jan 14)
Last: theorioleway


Page rendered in 0.4369 seconds
43 querie(s) executed