Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Baseball Newsstand > Baseball Primer Newsblog > Discussion
Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Friday, August 24, 2018

Posnanski: Baseball 100 Rules

In this era of reboots, it was perhaps inevitable that Joe Posnanski would take another crack at the 100 greatest players in major league history. 

The Baseball 100 is more than just a ranking system to me. The difference between my 78th ranked player and my 212th ranked player is so miniscule that it’s mathematically irrelevant. With one slight adjustment, I could have those two players switch places.

Nearly all of the series is to be pay walled, but Zach Greinke is No. 100 on the list.

In the original version of this list, I included a bunch of Negro leaguers — I can tell you that four were in my Top 20. I still believe this. But Negro leaguers will now be a major part of my corresponding Shadowball 100….It’s an eclectic list that includes players who are, in their own ways, larger than life.

No. 100 on this list is Duane Kuiper.

 

 

Rennie's Tenet Posted: August 24, 2018 at 08:01 AM | 1203 comment(s) Login to Bookmark
  Tags: history, joe posnanski, joe posnanski top 100, reboots

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.

Page 11 of 13 pages ‹ First  < 9 10 11 12 13 > 
   1001. Eric J can SABER all he wants to Posted: March 18, 2020 at 04:43 PM (#5931770)
@995: All the linear weights formulas work by adding up offensive events, not defensive mistakes (errors, WP, balks, etc.). The weight given to each event is the one which says "use this coefficient for the league data and it will give you a close estimate of the runs scored in that league". The problem is that this gives too much credit to the offensive events; the coefficients are too high. That's because the offense, per se, is only contributing to *earned* runs. Think of it this way: imagine that there were no errors or other defensive mistakes. We'd have to adjust the coefficients downward because they wouldn't accurately predict the number of runs scored.

bWAR, at least, does this already to some extent (at least, they account for the differences in ROE rates over time and the like). B-R's coefficients for the value of each event vary over time (there's a table at the link), and the coefficients for most recorded batting events are notably lower in, say, the 1908 NL than they were in the 2008 NL. Note (from another table at the link) that Wagner suffers the second-biggest career decrease in batting runs above average from the adjustment - and still has over 130 WAR.
   1002. Mefisto Posted: March 18, 2020 at 05:02 PM (#5931773)
Accounting for ROE is good, but it doesn't reach the entire impact of errors. As I suggested, the easiest fix is to run the regressions against earned runs and revise the coefficients accordingly.
   1003. TomH Posted: March 18, 2020 at 06:03 PM (#5931784)
Thanks for the explanations re: ROE, etc.

I agree errors had a big impact such that the coefficients could be wrong. However, we can do better than simply assuming all batters were equal in accounting for unearned runs. Guys who get on base (and don't strike out) are helped more by errors. Fast guys generated more defensive errors, and thus more runs. The Runs Created formulae account for this.
   1004. Mefisto Posted: March 18, 2020 at 06:15 PM (#5931786)
Yes, that was Eric's point. But I'm not sure we have to try to individualize the errors data. We don't do that for the other components; SB, for example, have the same value no matter who steals them, some just steal more.

My point about the coefficients was a more general one. We got to using linear weights for individuals by first demonstrating that they worked for the league as a whole. That they work for individuals is merely an assumption, since we have no way to prove that. What I'm suggesting is that we can account for errors league-wide by re-running the regressions on earned runs rather than total runs, and then work on allocating ROE and other events to individual players. My own view is that errors are pretty randomly distributed and that it would be a mistake to assume that some players are "better" at forcing them. But that argument has been going on for 35 years or so and I don't think anybody has solved it convincingly.
   1005. Eric J can SABER all he wants to Posted: March 18, 2020 at 06:21 PM (#5931788)
My point about the coefficients was a more general one. We got to using linear weights for individuals by first demonstrating that they worked for the league as a whole. That they work for individuals is merely an assumption, since we have no way to prove that. What I'm suggesting is that we can account for errors league-wide by re-running the regressions on earned runs rather than total runs, and then work on allocating ROE and other events to individual players. My own view is that errors are pretty randomly distributed and that it would be a mistake to assume that some players are "better" at forcing them. But that argument has been going on for 35 years or so and I don't think anybody has solved it convincingly.

I'm not a fan of using only earned runs in the regression, because it assumes that non-error events don't contributed to unearned runs at all. If a batter reaches on an error with two outs and the next five batters all hit doubles, the batting team will score five unearned runs. The fact that the runs were designated as unearned does not mean that the doubles weren't important.
   1006. Mefisto Posted: March 18, 2020 at 06:34 PM (#5931791)
Well, the whole point of distinguishing unearned runs is a "but for" test: "but for" the error, those 5 doubles wouldn't have happened at all.
   1007. Eric J can SABER all he wants to Posted: March 18, 2020 at 06:43 PM (#5931793)
Well, the whole point of distinguishing unearned runs is a "but for" test: "but for" the error, those 5 doubles wouldn't have happened at all.

(A), the counterfactual is an arguable point. But (B), so what? The error and doubles did happen, and contributed to runs for the team. Acting as though the error was the only thing that mattered seems like it obfuscates value rather than clarifying it. And in a game in which 30% of runs were scored as unearned, ignoring any non-error event that contributes to an unearned run scoring seems especially unhelpful. Was Tris Speaker's game-tying single in the bottom of the 10th in Game 8 of the 1912 World Series irrelevant because the run was unearned?
   1008. Jaack Posted: March 18, 2020 at 06:53 PM (#5931797)
I've never been a fan of the earned vs unearned distinction in general. In particular, I don't trust that official scorers in deadball were consistent in what they considered errors either with each other or with subsequent official scorers.

I feel like counting RoE as hits and pretending the error was never an official statistic probably has more statistical value than ignoring all unearned runs, particularly in an era with a lot of unearned runs.
   1009. Mefisto Posted: March 18, 2020 at 06:53 PM (#5931798)
Accepting your point for the sake of argument, yes the doubles happened. The problem is that those doubles had the *opportunity* to happen in 1903 (or in 1912, to use your example) which they don't have today because errors are reduced so significantly. That's not accounted for. There may be other ways to account for it than to use unearned runs, I just suggested that because it's simple to calculate. But one way or the other there has to be such an accounting.
   1010. Mefisto Posted: March 18, 2020 at 07:02 PM (#5931799)
@1008: The problem is that then you're crediting the batter for more value than he's actually creating *as a batter*. Runs don't happen in a vacuum (other than HR). They happen in a series of events. To take this back to Wagner, his hits created X runs under the current formula *only because there were lots of errors*. If he put up the same statistical line today, those hits wouldn't produce as many runs because today there are less than 1/4 as many errors. That difference doesn't matter if we compare Wagner to Lajoie because they played at the same time. It matters a lot if we compare Wagner to Ripken or Bogaerts.
   1011. Eric J can SABER all he wants to Posted: March 18, 2020 at 07:19 PM (#5931806)
Accepting your point for the sake of argument, yes the doubles happened. The problem is that those doubles had the *opportunity* to happen in 1903 (or in 1912, to use your example) which they don't have today because errors are reduced so significantly. That's not accounted for. There may be other ways to account for it than to use unearned runs, I just suggested that because it's simple to calculate. But one way or the other there has to be such an accounting.

I guess I don't really see why? There's already kind of an accounting; deadball players are all assigned a bunch of extra batting outs that weren't actually outs.

If Honus Wagner hits a double that drives in a player who reached on an error, that's a run for the Pirates in 1909 whether it would have happened in 2019 or not. If Pete Alonso hits a solo home run in 2019, that's a run for the Mets whether it would have been a lazy flyout in 1909 or not.

To switch to the other side of the spectrum - do you account for unearned runs allowed by pitchers? The standard modern example would be Curt Schilling vs. Kevin Brown - nearly identical inning totals and both with a 127 ERA+, but Brown allowed over 100 extra unearned runs, largely because errors are more common on ground balls and Brown was a ground ball pitcher.
   1012. Eric J can SABER all he wants to Posted: March 18, 2020 at 07:24 PM (#5931810)
@1008: The problem is that then you're crediting the batter for more value than he's actually creating *as a batter*. Runs don't happen in a vacuum (other than HR). They happen in a series of events. To take this back to Wagner, his hits created X runs under the current formula *only because there were lots of errors*. If he put up the same statistical line today, those hits wouldn't produce as many runs because today there are less than 1/4 as many errors. That difference doesn't matter if we compare Wagner to Lajoie because they played at the same time. It matters a lot if we compare Wagner to Ripken or Bogaerts.

Based on the table used in calculating bWAR, the run coefficients of singles, doubles, and triples were all notably lower during Wagner's career than they were 100 years later.
   1013. Jaack Posted: March 18, 2020 at 07:37 PM (#5931815)
@1008: The problem is that then you're crediting the batter for more value than he's actually creating *as a batter*. Runs don't happen in a vacuum (other than HR). They happen in a series of events. To take this back to Wagner, his hits created X runs under the current formula *only because there were lots of errors*. If he put up the same statistical line today, those hits wouldn't produce as many runs because today there are less than 1/4 as many errors. That difference doesn't matter if we compare Wagner to Lajoie because they played at the same time. It matters a lot if we compare Wagner to Ripken or Bogaerts.


I do not think that we can really say anything about what official scorers were considering errors in deadball. There were four times as many errors recorded, but I don't think we have the evidence to say that players were booting balls and making errant throws four times as often. The change in errors could just as easily be official scorers tightening up the definition of an error.

The error as a concept is not great in general, and they hurt a hitters' profile more than they help - they are awarded an out for getting on base. It's like counting a home run as an out because the pitcher hung a curve.
   1014. Best Dressed Chicken in Town Posted: March 18, 2020 at 07:49 PM (#5931820)
My own view is that errors are pretty randomly distributed and that it would be a mistake to assume that some players are "better" at forcing them.

But some are. Derek Jeter led the league in ROE like 5 times, which makes sense because he was a fast, right-handed ground ball hitter.
   1015. Mefisto Posted: March 18, 2020 at 08:14 PM (#5931824)
There's already kind of an accounting; deadball players are all assigned a bunch of extra batting outs that weren't actually outs.


But those are the plays that are actual outs today, not an opportunity for more batters to bat. And it doesn't account for, say, throwing errors on taking the extra base.
   1016. Mefisto Posted: March 18, 2020 at 08:16 PM (#5931825)
do you account for unearned runs allowed by pitchers? The standard modern example would be Curt Schilling vs. Kevin Brown - nearly identical inning totals and both with a 127 ERA+, but Brown allowed over 100 extra unearned runs, largely because errors are more common on ground balls and Brown was a ground ball pitcher.


That's an argument that always struck me as cutting both ways. It may just mean that Schilling had the good fortune to have better defenses behind him, which would make him *less* valuable than if we penalize Brown.
   1017. Mefisto Posted: March 18, 2020 at 08:19 PM (#5931826)
I do not think that we can really say anything about what official scorers were considering errors in deadball. There were four times as many errors recorded, but I don't think we have the evidence to say that players were booting balls and making errant throws four times as often. The change in errors could just as easily be official scorers tightening up the definition of an error.


There are plenty of reasons why errors were more common then: bad gloves; bad fields; replacement level was lower; maybe others. But it's speculative to say that official scorers "might have been" to blame. We have no way to know that.
   1018. Ziggy is done with Dominican discotheques Posted: March 18, 2020 at 08:37 PM (#5931832)
That difference doesn't matter if we compare Wagner to Lajoie because they played at the same time. It matters a lot if we compare Wagner to Ripken or Bogaerts.


No it doesn't, because Wagner actually created value for his team that Ripken and Bogaerts didn't. That doesn't make him a more talented player. And if you DeLorean Ripken back to 1905 he'd have produced that value. But regardless of which was the more talented player, and regardless of whether Ripken would have been able to do it or not, he didn't. And Wagner did.
   1019. Mefisto Posted: March 18, 2020 at 08:52 PM (#5931835)
No it doesn't, because Wagner actually created value for his team that Ripken and Bogaerts didn't. That doesn't make him a more talented player. And if you DeLorean Ripken back to 1905 he'd have produced that value. But regardless of which was the more talented player, and regardless of whether Ripken would have been able to do it or not, he didn't. And Wagner did.


Timelining assumes that if you DeLorean 1908 Wagner to 2020, he'd perform worse than he did in 1908. It's not about producing value in the league of 1908, it's how much value he'd produce if we magically transported him to the present. That's another way of saying that replacement level in 1908 was lower, so we can't compare value created across eras without adjustment.

ETA: Simply accepting value created in a given year, or over a career, isn't a hard problem. We can already measure value pretty well via WAR. We could just take the WAR list and say "that's the ranking".

But that's not really getting at the issue that (IMO) interests people. When someone asks if Cobb was greater than Mays, what they want to know is whether the value Cobb created came against competition equal to or better than what Mays faced. Lots of factors tell us that Cobb's competition was far weaker (see my question above about the second greatest NL position player from 1900-10), and timelining is an effort to measure *how much* weaker it actually was.
   1020. Mefisto Posted: March 18, 2020 at 09:06 PM (#5931838)
On re-thinking it, I should withdraw 1016. My point there is probably wrong.
   1021. Eric J can SABER all he wants to Posted: March 18, 2020 at 09:10 PM (#5931839)
Timelining assumes that if you DeLorean 1908 Wagner to 2020, he'd perform worse than he did in 1908. It's not about producing value in the league of 1908, it's how much value he'd produce if we magically transported him to the present.

If you're magically transporting Wagner to the present (which, for what it's worth, is not my preferred idea of timelining at all), then why assume he'll put up an identical stat line (or a stat line whose value is distributed the same way among its components, even if not identical)? He probably hits for a lower average, steals fewer bases, and hits more home runs than he did 110 years ago, because the game has changed.

Also, you're still assuming that the extra errors in the DBE made hits more valuable back then than they are now. I would contend that the extra homers now make it more likely that a modern hit will lead to a run.

That's an argument that always struck me as cutting both ways. It may just mean that Schilling had the good fortune to have better defenses behind him, which would make him *less* valuable than if we penalize Brown.

Brown was unquestionably more GB-heavy than Schilling, to a ridiculous extent - Schilling's career GB/FB ratio is 0.71, Brown's is 1.45. Errors occur more frequently on ground balls than fly balls regardless of how good the defense is - just look at league fielding percentages by position in any season (for 2019, 3B/SS/2B were .960/.971/.982, LF/CF/RF were .982/.987/.983).

For what it's worth, B-R has Brown's defenses at .05 runs per 9 below average for his career, Schilling's as exactly average. Their unearned runs per 9 innings are separated by over 5 times that much (about .29 per 9 innings). I'm not saying this is definitive as an evaluation of the respective defenses behind the two pitchers. I am, however, saying that if the two of them pitched for a team that had average defense at every position, Brown would almost certainly have allowed more unearned runs because his pitching style lends itself to that outcome.

Edit: Posted this before seeing 1020.
   1022. Mefisto Posted: March 18, 2020 at 09:19 PM (#5931842)
If you're magically transporting Wagner to the present (which, for what it's worth, is not my preferred idea of timelining at all), then why assume he'll put up an identical stat line (or a stat line whose value is distributed the same way among its components, even if not identical)? He probably hits for a lower average, steals fewer bases, and hits more home runs than he did 110 years ago, because the game has changed.


His stat line will be much weaker today. Many of his hits would be caught by today's defenders. He wouldn't be able to get his bat around on today's pitchers. Etc.

Also, you're still assuming that the extra errors in the DBE made hits more valuable back then than they are now. I would contend that the extra homers now make it more likely that a modern hit will lead to a run.


No, my view is that the extra errors made hits *less* valuable.
   1023. the Hugh Jorgan returns Posted: March 18, 2020 at 09:36 PM (#5931844)
His stat line will be much weaker today. Many of his hits would be caught by today's defenders. He wouldn't be able to get his bat around on today's pitchers. Etc.


You're basing this timelining solely on him as a MLB player. If we was born in 1993, had access to better nutrition, better coaches, travel ball, all the advantages that today's kids and athletes get, etc., doesn't he develop on the same level as today's best players? Then he does hit the ball harder, gets around on today's pitchers and still becomes one of the 20 best players ever.

Wagner was 6 foot, 200lbs...a big, strong guy back in his day. Let's say he was born in 1993, with a few more generations of Wagner family evolutionary genetics behind him, you don't even know for sure if this generation's Honus doesn't grow to be 6'3" 225lbs and built like Mike Trout.
   1024. Eric J can SABER all he wants to Posted: March 18, 2020 at 09:39 PM (#5931845)
Also, you're still assuming that the extra errors in the DBE made hits more valuable back then than they are now. I would contend that the extra homers now make it more likely that a modern hit will lead to a run.

No, my view is that the extra errors made hits *less* valuable.


Sorry, by "more valuable" I meant "produce more runs." Poor phrasing on my part.

His stat line will be much weaker today. Many of his hits would be caught by today's defenders. He wouldn't be able to get his bat around on today's pitchers. Etc.

This is assuming a lot, part of which is that Wagner would be limited to deadball hitting tactics if he was transported to the 2010s. Wagner (and Lajoie, and Cobb, etc.) built their games around contact because that was the strategy that worked at the time (partly because the defenses were not very good). In a modern context, I suspect they would adapt to the times and go more for power (which he might well be capable of - he led his leagues in extra-base hits seven times). At the very least, there's no way they'd be using 40-plus ounce bats after the first week.

Also, it's worth pointing out that if your hypothetical is always based around past players being dumped into the modern game, you're inherently biasing the evaluation in favor of the modern players who were trained in the current style as they were coming up. If you dumped Giancarlo Stanton in 1910, he might have trouble adapting to the mushy balls and terrible fielding equipment. (And yes, I picked Stanton as a current star who is largely built around the home run in a way that Trout isn't. Trout would be fine in any era.)
   1025. Mefisto Posted: March 18, 2020 at 09:50 PM (#5931847)
You're basing this timelining solely on him as a MLB player.


Yes. That's the only way to sort out the lower quality of competition he faced. His opponents were smaller and weaker. They used lousy gloves. The pitchers didn't throw as hard. All of that meant that Wagner faced replacement players who were much worse than those today; accordingly any WAR he produced today would be much lower. He doesn't get to be "reborn in 1993" and have all the modern advantages (see below).

This is assuming a lot, part of which is that Wagner would be limited to deadball hitting tactics if he was transported to the 2010s.


See above. But yes, if we're to get at the problem of producing high value against weak players, we have to do an experiment that takes old style techniques and tests them against the modern game.

If you dumped Giancarlo Stanton in 1910, he might have trouble adapting to the mushy balls and terrible fielding equipment.


I think Stanton would be thrilled to hit against pitchers lobbing batting practice to him. But per the above about how to conduct the experiment, we'd send Stanton back with a modern glove and bat.
   1026. Eric J can SABER all he wants to Posted: March 18, 2020 at 09:58 PM (#5931849)
I see no reason whatsoever that modern players should be given that many advantages in time-travel style timelining. (Even if Stanton tears up 1910 with his 2020 bat, how are you isolating Stanton's actual abilities in that experiment? You could just be discovering that 2020 bats are better than 1910 bats. Quality of equipment and quality of player are very different questions.)

But given that I don't subscribe to that school of timelining to begin with, I'm comfortable concluding that we just have irreconcilable differences on the topic.
   1027. Jaack Posted: March 18, 2020 at 10:05 PM (#5931850)
The time-travel thought experiment is more about the parameters you set than the quality of the players.

If you give a modern day player all the technological advantages and send him to 1910, he'll dominate whether he was Giancarlo Stanton or Yuni Betancourt.
   1028. Mefisto Posted: March 18, 2020 at 10:14 PM (#5931851)
I see no reason whatsoever that modern players should be given that many advantages in time-travel style timelining. (Even if Stanton tears up 1910 with his 2020 bat, how are you isolating Stanton's actual abilities in that experiment? You could just be discovering that 2020 bats are better than 1910 bats. Quality of equipment and quality of player are very different questions.)


You do that because in conducting an experiment you need to control all the factors but one. What we want to know is how much lower the level of competition was in 1908 compared to today. Therefore we hold Stanton constant and put him into the competitive environment back then. Similarly, we'd hold Wagner constant and test him against today's game.

It's quite likely that all those other factors like bats are important too, and we could test each one individually, but really that just reinforces the claim that overall competition was much lower back then.

If you give a modern day player all the technological advantages and send him to 1910, he'll dominate whether he was Giancarlo Stanton or Yuni Betancourt.


Absolutely. And that's exactly why we can't compare Wagner's WAR with that of a modern player unless we adjust it. It's the extent of that adjustment which is the whole debate.
   1029. Ron J Posted: March 19, 2020 at 07:50 AM (#5931889)
My own view is that errors are pretty randomly distributed and that it would be a mistake to assume that some players are "better" at forcing them.


If fact they aren't randomly distributed and we've known this for some time. (You'll find studies that touch on this going back at least as far as Craig Wright's time with the Rangers)

They are however sufficiently rare that it's generally not that important to account for. You aren't chopping runs off team standard error (of runs scored) when you include them. My own feeling is that it's best to acknowledge that (as mentioned) Jeter gets slightly underrated by traditional batting metrics. I don't think this alters his position in history and he's something of an outlier.

Worth noting that the really fast players tend not to be that good at reaching on error. The close plays are more likely to be scored as a hit for the guys who can really fly. And this gets into the problem with errors and accounting for events. There's a discretionary element to reached on error.
   1030. Mefisto Posted: March 19, 2020 at 08:40 AM (#5931895)
I guess I'm out of date, then, on the issue of errors. That's fine, and I don't object to counting them.

Errors create problems in analysis and it's probably too bad they became a category. But they are one, for whatever reason there were lots more of them 100 years ago, and we have to have a way to normalize that if we want to compare across time.
   1031. Mefisto Posted: March 20, 2020 at 08:12 AM (#5932199)
Pos better start publishing the rest or some of the living might not be.
   1032. Rally Posted: March 20, 2020 at 11:30 AM (#5932240)
Mantle #11, I guess putting him ahead of Mays, Musial, Aaron or some of the others to make him #7 was a stretch too far.
   1033. Mefisto Posted: March 20, 2020 at 12:33 PM (#5932256)
Weirdly enough given my comments in this thread, I actually have Mantle at #7. But I rank pitchers and position players on different lists so Pos would need to account for that.
   1034. bbmck Posted: March 20, 2020 at 12:39 PM (#5932257)
Mantle with the same issue as Wagner, who was the 2nd best position player in the AL 1951-1961? How can one player be so much better than the rest of the league? Were they playing 'real' baseball in the 1950s junior circuit?
   1035. Mefisto Posted: March 20, 2020 at 12:54 PM (#5932262)
Not quite the same while Ted Williams was playing. But yeah, the AL back then was pretty weak, both in comparison to what it was earlier and in comparison to the NL.
   1036. TomH Posted: March 20, 2020 at 01:44 PM (#5932272)
Yogi Berra (2nd best AL player), or Ted if he wasn't in Korea.
Yes, AL was weaker, but geee, Mick did OK in the post-season against those great NL teams.

I think we should all agree that in his prime, (lets say 54-61), Mantle was THE best player in the major leagues. Obviously Mays (and Aaron) lasted longer.

   1037. Rally Posted: March 20, 2020 at 02:28 PM (#5932283)
Mantle was what Mike Trout would be with better teammates.
   1038. Mefisto Posted: March 20, 2020 at 03:10 PM (#5932300)
I think we should all agree that in his prime, (lets say 54-61), Mantle was THE best player in the major leagues. Obviously Mays (and Aaron) lasted longer.


Mantle 54-61: 71.1 WAR
Mays 54-61: 71.7 WAR

ETA: I agree with 1037.
   1039. bbmck Posted: March 20, 2020 at 04:36 PM (#5932316)
The 735 players with 3000+ PA and 110+ OPS+ at Age 23-32 collectively average .366 OBP not weighted by individual PA and reach 1st base safely (TOB w/ROE / PA) 37.4% of the time based on total TOB w/ROE and PA. Positive score is reaching 1st base safely higher than OBP, year is first in MLB at Age 23+:

48 players +.015-.019

1924 - Earl Combs, 1928 - Pepper Martin
1930s - 10 players
1940s - 5 players
1950s - 8 players

1960s - 5 players
1970s - 9 players
1980s - 7 players
1994 - Javy Lopez, 1995 - Rondell White

248 players +.010-.014

1918 - Irish Meusel, 1919 - High Pockets Kelly
1920s - 8 players
1930s - 16 players
1940s - 15 players
1950s - 24 players

1960s - 42 players
1970s - 48 players
1980s - 36 players
1990s - 30 players
2000s - 21 players
2011 - Paul Goldschmidt, JD Martinez, 2012 - Starling Marte, Yoenes Cespedes, 2014 - George Springer, Jose Abreu

264 players +.005-.009

1916 - George Burns, 1918 - Babe Ruth, Billy Southworth
1920s - 16 players
1930s - 15 players
1940s - 29 players
1950s - 9 players

1960s - 20 players
1970s - 21 players
1980s - 28 players
1990s - 49 players
2000s - 50 players
2010s - 24 players

All 31 players 1871-1885 are 0, the first player -.005 or lower is 1890 Hugh Duffy, the last player is 1915 Happy Felsch. 1892 Mike Smith and Jesse Burkett, 1897 John Anderson, 1899 Charlie Hickman and 1909 Burt Shotton are -.003 or -.004, every other player before 1910 is -.005 or lower. 1922 Joe Sewell is the most recent below 0.

1910s - 20 players -.005 or lower, 19 players -.004 or +.004

-.004 to +.004

1920s - 12 players
1938 - Roy Cullenbine
1947 - Ferris Fain
1978 - Dwayne Murphy
1985 - Darren Daulton
1988 - Brady Anderson

2002 - Nick Johnson
2003 - Adam Dunn
2004 - Curtis Granderson
2010 - Lucas Duda
2011 - Matt Carpenter
2013 - Anthony Rendon

Basically once Sac Flies are tracked players are reaching first base safely at a better rate than their OBP and pre-Babe Ruth there are a bunch of errors and little tracked play by play data that offsets those ROE.

Highest rate of reaching 1st base safely among the group and age range outside the Hall of Fame:

.444 - Mike Trout
.440 - Todd Helton
.436 - Bill Joyce
.431 - Joey Votto
.430 - Barry Bonds
.427 - Ferris Fain

.426 - Gary Sheffield, Albert Pujols
.424 - Brian Giles
.423 - Manny Ramirez
.422 - Joe Cunningham
.421 - John Olerud, Jason Giambi, Riggs Stephenson
.420 - Elmer Valo, Lance Berkman, Bobby Abreu

.400+ Between Joe Cunningham and Barry Bonds:

1959 - Frank Robinson .415
1963 - Carl Yastrzemski .408
1967 - Joe Morgan .413
1969 - Rod Carew .416
1970 - Ken Singleton .408
1974 - Mike Hargrove .408

1977 - Keith Hernandez .406
1982 - Wade Boggs .446
1982 - Rickey Henderson .418
1986 - John Kruk .404
1987 - Edgar Martinez .416

The .400+ trajectory: 14 players 1881-1899, none 1900-1909, 8 players in the 1910s, 14 in the 1920s, 15 in the 1930s, 12 in the 1940s, 7 in the 1950s, 3 in the 1960s, 3 in the 1970s, 5 in the 1980s, 19 in the 1990s, 2002 Nick Johnson, 2003 Albert Pujols, 2004 Kevin Youkilis, 2006 Miguel Cabrera and Joe Mauer, 2007 Joey Votto, 2011 Paul Goldschmidt and 2015 Mike Trout. It seems pretty unlikely to reach 1st base 40%+ of the time with an OPS+ below 110.

Some high BA guys who don't reach 1st 40%+ of the time at Age 23-32: Ichiro Suzuki .331/.390, Tony Gwynn .328/.394, Roberto Clemente .324/.382, George Brett .321/.395, Kirby Puckett .321/.371 and Jose Altuve .320/.377.
   1040. TomH Posted: March 20, 2020 at 07:17 PM (#5932363)
@1038
Mantle DWARFS Mays in Win Shares
Mantle DWARFS Mays in OPS+, 184 to 164. Slightly ahead in baserunning/gidp also, by 15 runs. Oh, and 10 World Series home runs.
I suspect WAR gives Mays a greater fielding credit (77 runs) than WS.
   1041. Mefisto Posted: March 20, 2020 at 07:48 PM (#5932370)
Win Shares doesn't impress me, especially when applied to a dominant team in a weak league. WAR is a much better measure.

Regardless, we do not all agree that Mantle was the best player in MLB in his prime. Here, for example, are the Hall of Merit voting results for those years:

1954: Mays first, Mantle 11th
1955: Mays first, Mantle 2d
1956: Mantle first, Mays 7th
1957: Mantle first, Mays 3d
1958: Mays first, Mantle 2d
1959: Mays fourth, Mantle 5th
1960: Mays first, Mantle 6th
1961: Mantle first, Mays 4th

Even during Mantle's peak, they had Mays ahead in of him in 5 of the 8 years.
   1042. TomH Posted: March 21, 2020 at 05:35 AM (#5932408)
well, we disagree on "WAR is a much better measure"; WAR tells us that the two best AL pitchers last year were Lance Lynn and Mike Minor. Hee hee.

But I appreciate and respect the HOM voters record you posted, thanks. I do note that the voters put them in the same order WAR does each year. Of course ordinal ranking does *not* capture how much a truly dominant season.. Mantle's 56-57 were so far ahead of every other MLB player it's reedikalus.
   1043. Howie Menckel Posted: March 21, 2020 at 08:37 AM (#5932414)
it seems almost impossible to overstate Mantle's advantage in the racist AL vs the integrated NL that took up most of his career.

his greatness cannot be questioned, of course. but nice to have maybe 1 integration stud in a good year instead of 10 to battle in offensive dominance badly skews his career without that filter.
   1044. TomH Posted: March 21, 2020 at 10:59 AM (#5932434)
I dunno, Howie... maybe you just did overstate it :)

Look, Mantle's AL, mostly unintegrated, was not weaker than the Ruth/Gehrig/DiMaggio/Williams AL of the 20s-30s-40s, was it? Can we at least agree on that? Mantle dominated his league more than Gehrig his.
   1045. Mefisto Posted: March 21, 2020 at 11:31 AM (#5932441)
The 1950s AL was among the weakest leagues relative to the contemporaneous NL; it's hard to compare strength of leagues across time (that's the whole problem of timelining). The AL of the 20s and 30s was far stronger relative to the NL. The 40s is hard to judge because of the war.
   1046. TomH Posted: March 21, 2020 at 12:15 PM (#5932467)
oh, I get that, and I agree; I am just saying we should avoid going from "AL 55 was so weak compared to NL 55" (good conclusion) to "AL 55 was maybe the weakest of all leagues" (bad conclusion); as that would mean somehow from Gehrig to Williams the AL got much weaker. Instead we should posit that the NL of 1960, pre-expansion, was MUCH stronger than any league in previous eras. How much stronger? I agree, we don't know. Mays vs Ruth is an impossible argument to solve conclusively.

We *could* put a reasonable CAP on AL-vs-NL in the 50s. Let someone argue the typical NL team that went 77-77 would have gone 85-69 (over .550 wpct). If it was 8 wins per year, Mantle, being only about 6%-7% of a whole team when playing every day, then might be .065*8= 0.5 wins (about 5 runs) a year worse in WAR if he was in the NL.





   1047. Mefisto Posted: March 21, 2020 at 12:41 PM (#5932481)
BBREF already does that for us. The Rrep column in the Player Value table shows the number of runs the average player is above a replacement player for the given year. In the period 54-61, the NL is 24 (heh) or 25 runs, in the AL it's 19-20. For comparison, in period 1925-35 the AL was 25-6, the NL was 20-21. In essence, the two leagues switched their relative strengths.
   1048. John DiFool2 Posted: March 21, 2020 at 04:09 PM (#5932519)
I'd like to see a study involving players who switched leagues. Tho in that time period the sample size I believe was miniscule.
   1049. Mefisto Posted: March 21, 2020 at 04:17 PM (#5932521)
Somebody (Tango?) has done such a study. I'm sure Rally knows it.
   1050. taxandbeerguy Posted: March 22, 2020 at 12:02 PM (#5932628)
So that leaves the top 10 as (alphbetically)

Hank Aaron
Barry Bonds
Oscar Charleston
Ty Cobb
Walter Johnson
Willie Mays
Stan Musial
Satchel Paige
Babe Ruth
Ted Williams

I had Josh Gibson (10, POZ 15) up here and Roger Clemens (5 POZ 13) and Honus Wagner (9 POS 12) rather than Charleston (25), Musial (11) and Paige (14). Mantle's 12 for both Joe and I, so pretty close on both lists. Top 10 is very outfield heavy, even more so than my list. I still believe Mays, Bonds and Ruth are the top 3.
   1051. John DiFool2 Posted: March 22, 2020 at 01:54 PM (#5932644)
Yes, I've read all of the timelining stuff above. [and, partially, acknowledge that it is very likely true, and requires an adjustment]

Still think that if you have Honus Wagner born 100 years later that he'd still be the best player in the league.
   1052. Booey Posted: March 22, 2020 at 04:15 PM (#5932666)
Hmmm...so that top 10 features just 1 player who debuted between 1955-2010. Makes me think Poz might not be timelining enough. Seems almost impossible for a modern player to crack it; Trout's probably only 50/50 to surpass even the WORST player left (Musial, maybe?)
   1053. Mefisto Posted: March 22, 2020 at 04:51 PM (#5932673)
At a pure guess: Paige; Cobb; Charleston; Musial; Aaron; Johnson; Williams; Ruth; Bonds; Mays (or switch Bonds and Mays). The ordering of Cobb, Charleston, and Musial could go any other way among the three and he might bump Musial to 6 just because.
   1054. bbmck Posted: March 22, 2020 at 05:44 PM (#5932680)
Edgar Martinez is the best realistic case for a modern player in their 30s. 5436 PA, 46.6 WAR, 157 OPS+, 3 Rfield, -102 Rpos so if Mike Trout stays in the field he just has to be better than Gary Sheffield -60 Rfield and -65 Rpos in his 30s while hitting as well as Edgar. Edgar did have the 94-95 strike, 42 games in 1993 and 97 games in 2002. Adrian Beltre with 96 Rfield and 12 Rpos in his 30s is an unrealistic path to project. Chipper Jones 5417 PA, 43.9 WAR, 142 OPS+, -1 Rfield, 3 Rpos is another option or somewhere in between.

The 7 MLB position players in the Top 10 by position player WAR in their 30s:

88.4 - Barry Bonds 6181 PA, 206 OPS+, 40 Rfield, -54 Rpos
86.0 - Babe Ruth 6016 PA, 199 OPS+, 33 Rfield, -63 Rpos
79.9 - Willie Mays 6110 PA, 157 OPS+, 96 Rfield -7 Rpos
66.7 - Hank Aaron 6126 PA, 161 OPS+, 31 Rfield, -64 Rpos

58.3 - Ty Cobb 5524 PA, 160 OPS+, -6 Rfield, -33 Rpos
56.5 - Ted Williams 4452 PA, 191 OPS+, -11 Rfield, -41 Rpos
54.5 - Stan Musial 6005 PA, 157 OPS+, 5 Rfield, -64 Rpos

Other than Barry, in the 1955-2010 debut range, Mike Schmidt and Rickey! make really compelling cases for "only" 14 years which isn't long enough for serious Top 10 consideration. "Only" F-Rob, Yaz, Cal or A-Rod in your 30s and you're losing a lot of ground to those seven and have to be considered much better in your 20s and/or 40s.

In your 30s being Chipper, Beltre, Edgar or a healthier Larry Walker while also not having late development delay the start of your career is certainly possible. Assuming Trout doesn't lose too much time to Covid-19, injuries and/or the end of the world he's already done the really hard part of having an impressive resume in his 20s. But he's still only positioning himself as the next Frank Robinson with a reasonably high degree of likelihood. Of course Albert Pujols and Ken Griffey Jr had also positioned themselves to have a reasonable high degree of likelihood of being the next Frank Robinson.
   1055. TomH Posted: March 22, 2020 at 06:04 PM (#5932682)
1050 taxandbeerguy and I are muy sympatico on our top 10 lists
   1056. alsep73 Posted: March 23, 2020 at 07:42 AM (#5932731)
10. Satchel Paige.

Given Joe's love of the Negro Leagues, I wonder just how highly Oscar Charleston is going to rank. I can imagine a circumstance where he's top 5, maybe even top 3, just to make a statement about how great the greatest Negro League position player must have been.
   1057. EddieA Posted: March 23, 2020 at 10:43 AM (#5932764)
so that top 10 features just 1 player who debuted between 1955-2010.


So top 10 by debuts:
1900s - 2
1910s - 2
1920s - 1
1930s - 1
1940s - 1
1950s - 2
1960s - 0
1970s - 0
1980s - 1
1990s - 0
2000s - 0

In a definite dry spell.
   1058. Booey Posted: March 23, 2020 at 12:10 PM (#5932783)
Top 10 all timers in a 50 year span:

1905-1955 = 9
1955-2005 = 1

Yeah, call me skeptical. I'd personally put Clemens and ARod in my top 10...and of course both of them - and Bonds - are all PED guys, which helps illustrate that it's basically impossible for a clean modern player to dominate his era to the degree that the old timers could.
   1059. Mefisto Posted: March 23, 2020 at 12:22 PM (#5932787)
My Bingo card gets its first hit! Very exciting.

The big gap in the 60s and 70s is really obvious, but I'm not sure why it's there. Morgan and Schmidt are the 2 top players from those decades, but both fall just outside the top 10 (I have Schmidt at 11 and Morgan at 14 [I erroneously said 13 earlier]). Are we not timelining enough? Is there just a cycle there? Was competition especially difficult over that time? I don't know, but I think it's worth a study.

I agree with 1058.
   1060. Slivers of Maranville descends into chaos (SdeB) Posted: March 23, 2020 at 12:45 PM (#5932791)
Yes. That's the only way to sort out the lower quality of competition he faced. His opponents were smaller and weaker. They used lousy gloves. The pitchers didn't throw as hard. All of that meant that Wagner faced replacement players who were much worse than those today; accordingly any WAR he produced today would be much lower. He doesn't get to be "reborn in 1993" and have all the modern advantages (see below).


But Wagner suffered under all of those restraints as well. Don't we have to control for that if we want to determine who was more talented, Wagner or Stanton?
   1061. Jeremy Renner App is Dead and I killed it Posted: March 23, 2020 at 12:46 PM (#5932792)
Totally agree with 1058. Just in terms of common sense it's implausible that the majority of its very best performers of a sport come from an period more than 60 years in past. And yet with baseball there are a LOT of people who get super indignant when you employ common sense. Like totally outraged. So weird
   1062. EddieA Posted: March 23, 2020 at 12:50 PM (#5932795)
most of us have only seen 1 all-time great if top 10 is the criterion. I recall 2 Hank Aaron home runs in games I attended, one fence-skimmer in an All-star game. I was small. I recall 1 Barry Bonds home run in a game I attended, to left field off his fists! My memories of Willie Mays play are minimal. I'm sure he never homered in games I attended.

Watching old baseball clips, the oldest guys don't look impressive at all. It's probably something about the filming. But by Willie Mays, the footage gets pretty impressive.
   1063. Mefisto Posted: March 23, 2020 at 12:55 PM (#5932798)
Don't we have to control for that if we want to determine who was more talented, Wagner or Stanton?


Not for purposes of this discussion, no. The basic issue here is whether we can use WAR totals to judge players across eras. We can only do that if WAR is giving us comparable values, that is, if the replacement level player in each era is the same ability. My position is that in order to test that, we need to conduct a thought experiment in which we hold the player constant, but change his environment. Thus, for example, we'd have to hold Wagner (or any other player) constant in every respect if we could magically bring him into the modern game. He'd have to use his old glove, his old bat, etc. Similarly, if we wanted to test the other way, we'd have to hold Stanton constant if we magically sent him back to 1908.

Sure, if we had a Honus Wagner clone born in 2000, he'd get modern nutrition, modern training, and would play the modern game. But that wouldn't answer the question about the replacement level back then. For that, we need to hold him constant in order to limit the variables.
   1064. What did Billy Ripken have against ElRoy Face? Posted: March 23, 2020 at 01:25 PM (#5932809)
Watching old baseball clips, the oldest guys don't look impressive at all. It's probably something about the filming. But by Willie Mays, the footage gets pretty impressive.
It's that the guys were physically smaller, of course, but more that their mechanics were far from optimal. Most hitters were just trying to dink the ball somewhere. Even the power hitters of the Mays era - him, Aaron, Banks, Robinson, etc.* - hit off of their front foot, diving over the plate with their back foot bailing out behind them. If you had wrists like those guys, you could generate a lot of power, but they were of course the exceptions.

I don't really know the right terms for what I'm talking about, but before the widespread use of weight training, players' mechanics look like they relied on torque and leverage to generate power where possible, rather than bodily strength and efficiency of motion. This includes pitchers with their elaborate windups and slinging motions.

*RH hitters, that is - I haven't noticed it among LH hitters, but (a) there are a lot fewer, so maybe just not as memorable, and (b) LH swings just always look better.
   1065. bbmck Posted: March 23, 2020 at 05:57 PM (#5932918)
The 9 year average of 10th place in position player WAR is pretty stable once you account for work stoppages and a change in schedule length. 6.8 is the peak 1996-2004 and it's 6.5-6.8 for 1992-2000 to 2015-2019. It's 6.0-6.4 1953-1961 to 1991-1999. It's 5.5-6.1 1919-1927 to 1952-1960.

1918 drags things down with George Sisler 6.8 leading to Heinie Groh and part-time Babe Ruth tied for 9th at 4.7 being among the lowest values since ~150+ games a season started. 1898-1906 to 1918-1926 is 5.1-5.4. 1883-1891 to 1897-1905 is 4.6 to 4.9, prior to that it's a pretty steady drop with schedule length being a major factor.

Based on the 9 year average of 4 years prior and after during Honus Wagner's peak of 1899-1912:

#1 - 127.7 (5 are Honus)
#2 - 115 (2 are Honus)
Honus Wagner 111.4
#3 - 106.4 (3 are Honus)
#4 - 95.1 (1 is Honus)
#5 - 90.1

#6 - 84.1 (2 are Honus)
#7 - 80.7
#8 - 77.7 (1 is Honus)
#9 - 74.9
#10 - 72.7

Based on the 9 year average of 4 years prior and after during Mike Schmidt's peak of 1974-1987:

#1 - 128 (2 are Mike)
#2 - 116.1 (3 are Mike)
#3 - 109.8 (1 is Mike)
#4 - 105.8 (1 is Mike)
Mike Schmidt 103.5
#5 - 100.4 (2 are Mike)

#6 - 96.2 (1 is Mike)
#7 - 91.5
#8 - 89.1 (2 are Mike)
#9 - 87.1 (1 is Mike)
#10 - 85.6

Mike Schmidt "only" 5.0 in 1985 when 10th is 6.4. Honus precedes that with 4.9 WAR in 2 years, follows up that 14 year run with "only" 14.6 WAR in 4 years and a final season -0.1. Mike precedes that with 1.9 WAR in 2 years, follows up that 14 year run with a season of 1.8 WAR and a final season of -0.4 WAR.

The Phillies play 2214 regular season games during Schmidt's 14 year run, the Pirates and one year of the Colonels play 2107 games during Honus' 14 year run. ~8 more WAR in ~100 fewer possible games 75 years earlier, enter time lining. If you place much emphasis on Honus having a few extra productive years he's going to rank better than Schmidt and you would never project Mike Trout, Ozzie Albies or whoever to have more than 14 highly productive seasons until they have played at least 12 of those seasons.

By and large the "real" rankings simply reflect an emphasis on a few extra seasons, if only MLB had been depleted of talent near the end of Mike Schmidt's career so he could have tacked on a few extra productive seasons. Schmidt #20 and Rickey! #24 are both uniform numbers and have the most similar 14 year runs to Honus with Schmidt not playing much and Rickey! not playing well outside of those 14 years limiting the number of direct comparisons but for the most part the Poz rankings are just career WAR with some non-MLB players and "creative" rankings mixed in and the top of the list is players that pushed their way past similar players in their late 30s.
   1066. Mefisto Posted: March 23, 2020 at 06:25 PM (#5932928)
If you had wrists like those guys, you could generate a lot of power, but they were of course the exceptions.


I remember reading somewhere that Arizona State tested its players for forearm strength and that Barry was by far the best at that measure.
   1067. SoSH U at work Posted: March 23, 2020 at 06:42 PM (#5932936)
Yeah, call me skeptical. I'd personally put Clemens and ARod in my top 10...and of course both of them - and Bonds - are all PED guys, which helps illustrate that it's basically impossible for a clean modern player to dominate his era to the degree that the old timers could.


So Trout is obviously juicing?
   1068. Booey Posted: March 23, 2020 at 08:45 PM (#5932951)
#1067 - No, but Trout has a long way to go before he's anywhere near the top 10 all time.
   1069. SoSH U at work Posted: March 23, 2020 at 09:04 PM (#5932954)

#1067 - No, but Trout has a long way to go before he's anywhere near the top 10 all time.


He has a long way to go to rack up the counting stats of those guys, but he's already just as dominant in his era as many of those guys were in theirs.

He has five Top 10 position player WAR finishes, one fewer than Williams, the same as Cobb, one more than Musial, and three more than Aaron. Not surprisingly, he's got a ways to go to catch the consensus Big 3.

Moreover, pre-juicing Bonds was well on his way to the all-time Top 10 with a normal decline phase (nearly 100 WAR by his age 33 season). Clemens was on a similar trajectory to where he is now even if he'd never met Brian McNamee.

   1070. Booey Posted: March 23, 2020 at 10:17 PM (#5932967)
#1069 - Maybe "almost impossible" was an oversell, but that level of dominance is definitely much harder/less common than it used to be. Hence the need for stronger timelining.
   1071. SoSH U at work Posted: March 24, 2020 at 12:48 AM (#5932984)
#1069 - Maybe "almost impossible" was an oversell, but that level of dominance is definitely much harder/less common than it used to be. Hence the need for stronger timelining.


Obviously, it's going to be somewhat harder to dominate the same way because of the size of the leagues, but I think if you look beyond the Top 10, you don't see the same kind of edge to the early stars.

On BBRef's All-Time WAR Position Player* leaderboard, these are the Top 24 (the ones with photos). It's a pretty nice distribution.

Number of players by decade of greatest impact:

1880s: 1
1900s: 2
1910s: 3
1920s: 2
1930s: 2
1940s: 2
1950s: 2
1960s: 4
1970s: 1
1980s: 2
1990s: 1
2000s: 2

* Obviously, the same doesn't apply to the full leaderboard, where the early baseball pitchers dominates. But as Poz's list shows, no one really uses the early baseball WAR pitching leaderboard in the conversation about all-time greatness.
   1072. Booey Posted: March 24, 2020 at 09:12 AM (#5933032)
#1071 - That list only has a good balance if you consider "top 24" to be essentially the same as "top 10". My original post was about the top 10, and 9/10 (save Bonds) and 15/18 (add ARod and Rickey) of the top position player WAR totals debuted from the 1900's-1950's. The list doesn't start balancing out until the bottom when you get Schmidt, Pujols, Morgan, and Yaz in the 18-24 range.

It's also ofc not including Negro League players, which would tilt it even more in favor of the old timers.
   1073. SoSH U at work Posted: March 24, 2020 at 09:21 AM (#5933040)
#1071 - That list only has a good balance if you consider "top 24" to be essentially the same as "top 10".


Your original point was that it was almost impossible to dominate now sans steroids, which does look like that if you only look at the Top 10. But if you expand your view into the Top 24, such dominance fades a bit (though the Negro League inclusion is worth noting). I'm not saying the Top 24 is equal in importance, whatever that means, only that it's more information to consider when looking at the question.

I never said your original point was totally wrong. Just that it was, as you subsequently acknowledged, an overbid.
   1074. Ron J Posted: March 24, 2020 at 09:27 AM (#5933046)
#1062 It's even more disconcerting to watch hockey games from the 50s and 60s. They're playing in a completely different way. Really long shifts (and shorter benches) meant that they rarely went all out. And even at maximum effort don't look all that impressive to a modern eye.
   1075. Booey Posted: March 24, 2020 at 09:42 AM (#5933049)
Only 11 position players have reached the 120 WAR benchmark. 6 of them debuted in a 19 season span between 1897-1915. 4 more debuted in a 16 season span between 1939-1954. Only 1 has debuted since 1955. And that's not even taking into consideration that the old timers were playing 154 game seasons.
   1076. alsep73 Posted: March 24, 2020 at 11:52 AM (#5933085)
The "favorite player" series continues with an ode to Eddie Murray, who made the top 100 in a previous version of the list.
   1077. Mefisto Posted: March 24, 2020 at 12:30 PM (#5933096)
Only 1 has debuted since 1955.


And he's a bit controversial, though I'm confident he'd have hit that level regardless.

   1078. bbmck Posted: March 24, 2020 at 01:03 PM (#5933104)
Using the Poz standard of "real" baseball starting around 1895, min 1000 PA and fewest PA per position player WAR. Among the Top 10 most dominant players, the ones who debuted after 1955:

Age 20-24: Mike Trout
Age 21-25: Mike Trout
Age 22-26: Mike Trout
Age 23-27: Mike Trout, Ken Griffey Jr
Age 24-28: Mike Trout, Barry Bonds

Age 25-29: Mike Trout, Barry Bonds, Albert Pujols, Mookie Betts
Age 26-30: Mike Trout, Barry Bonds, Albert Pujols, Carl Yastrzemski
Age 27-31: Barry Bonds, Joe Morgan, Mike Schmidt
Age 28-32: Barry Bonds, Joe Morgan, Mike Schmidt
Age 29-33: Barry Bonds, Joe Morgan, Mike Schmidt

Age 30-34: Mike Schmidt
Age 31-35: Barry Bonds
Age 32-36: Barry Bonds
Age 33-37: Barry Bonds
Age 34-38: Barry Bonds

Age 35-39: Barry Bonds
Age 36-40: Barry Bonds
Age 37-41: Barry Bonds, Adrian Beltre
Age 38-42: Barry Bonds, Joe Morgan
Age 39-43: Barry Bonds, Carlton Fisk, David Ortiz, Brian Downing
Age 40-44: Barry Bonds, Carlton Fisk, Darrell Evans, Edgar Martinez, Graig Nettles
   1079. base ball chick Posted: March 24, 2020 at 02:53 PM (#5933129)
looks like this bonds guy was something special

even watching old games from the 70s is so completely different. the pitches, the swings - just no ocmparison

but i like them better because the pitchers take about 6 seconds between the time they get the ball and the time they pitch the ball.
   1080. TomH Posted: March 24, 2020 at 07:06 PM (#5933206)
Here is a brief explanation of why I “only” rank Walter Johnson as the 3rd best pitcher of all time, which is a position not many others would take. It is safe to say that Walter Johnson has been a consensus, although by no means unanimous, choice by many as the greatest pitcher in baseball history.

I do believe Walter Johnson was the most valuable pitcher ever; he helped his teams win more games than any MLB hurler. And, even with some timelining, there are no obvious candidates to overtake him, unless you wish to give Roger Clemens full credit (no ‘steroid discount’). Now, there are many great pitchers who primarily pitched in the deadball era; besides the Big Train, you might name Kid Nichols, Cy Young, Christy Mathewson, Pete Alexander. Do we know with much certainty how ANY of them would have fared in 1955 or 2010? No. So why would I pick on Johnson, who, like Alexander, actually DID pitch part of his career against the Babe and other power hitters? Because we DO have data on Johnson in live-ball circumstances, and it indicates pretty clearly that he was not the same dominant pitcher.
Walter Johnson has a career ERA of 2.17, and an ERA+ of 142; and 417 wins. Phenomenal. I would attribute his success to many factors:
- Struck out lots of guys
- Walked only a few
- The ball was dead
The first two apply well across all eras. The last one changed around 1920. How did Johnson do from 1920 on? From 1920-27, Johnsons’ ERA+ was 118. While this isn’t Maddux or Seaver, it is Phil Niekro, or Bob Gibson sans 1968; not bad considering he was an older guy at this point. So Tom, where is your point then? Here it is:

Johnson only allowed 97 home runs his entire career; almost 6000 innings! If you don’t walk anybody, and don’t allow home runs, and if your big KO totals limit the singles (and some doubles), it is hard to score many runs off you.
Johnson allowed 31 of his career home runs from 1907-1919, and 66 after that. Pre-1920, he allowed about as many HR home and away; I suspect most of them were in-the-park types. But 1920ff, he allowed 21 HR at home, and 45 on the road; even while pitching 20% more innings at home. In the live ball era, Johnson’s home park was death to home run hitters. It was 388 down the line to LF; and generally large, except for the RF corner.

His ERA from 1920 on was 2.83 at home, and 3.97 on the road. It would seem that his uber-fastball in the strike zone was still a great weapon at home, but on the road a few (and there were still only a few!) power hitters could tag him; so he became an average pitcher when in a neutral environment. In road games from 1920-27, Johnson allowed home runs at a pace significantly greater than the league average. In today’s game, where most batters hit for power, he might have become a Curt Schilling-like (“pound the strike zone”) pitcher… excellent, but with dinger-vulnerability.

Ultimately, I would want to pick a pitcher who could have been a great pitcher in the conditions over the past 100 years. So in my book, I would put the Big Train just outside the top 10 overall.
   1081. bbmck Posted: March 24, 2020 at 08:39 PM (#5933243)
Age 32-39

Greg Maddux: 33.2 pitWAR, 1808 IP, 132 ERA+, 76 OPS+, 1.4 posWAR, 643 PA, 11 OPS+
Walter Johnson: 32.2 pitWAR, 1823.2 IP, 119 ERA+, 76 OPS+, 4 posWAR, 790 PA, 74 OPS+
Tom Seaver: 27.3 pitWAR, 1649.1 IP, 112 ERA+, 84 OPS+, 1.2 posWAR, 526 PA, 29 OPS+

Walter Johnson: 2245 BF or 29.2% are first 9 BF in an appearance, 2073 BF or 27% are 10th-18th BF, 1842 BF or 24% are 19th-27th BF, 1264 BF or 16.5% are 28th-36th BF, 240 BF or 3.1% are 37th-45th BF, 18 BF or 0.2% are 46th-54th BF during the regular season. Playoff start game scores: 82, 75, 64, 37 and 26. 4 IP, 3 H, 3 BB and 0 R in his playoff relief appearance.

Tom Seaver: 2128 BF or 31.4% are first 9 BF in an appearance, 2060 BF or 30.4% are 10th-18th BF, 1805 BF or 26.6% are 19th-27th BF, 762 BF or 11.2% are 28th-36th BF, 23 BF or 0.3% are 37th-45th BF during the regular season. Playoff start game score: 67.

Greg Maddux: 2453 BF or 33.2% are first 9 BF in an appearance, 2436 BF or 32.9% are 10th-18th BF, 2043 BF or 27.6% are 19th-27th BF, 458 BF or 6.2% are 28th-36th BF, 5 BF or 0.1% are 37th-45th BF during the regular season. Playoff start game scores: 64, 60, 60, 59, 57, 56, 56, 52, 50, 50, 25 and 19. Retired 3 batters and walked one in a playoff relief appearance and walked the only batter he faced in his other playoff relief appearance.

Walter Johnson wasn't Lefty Grove or Randy Johnson at those ages, maybe even not Phil Niekro or Gaylord Perry the all-time IP leaders at those ages, Bob Gibson if you exclude his Age 32 season in 1968 is 36.4 pitWAR, 1675.2 IP, 122 ERA+, 80 OPS+ and is another similar pitcher.
   1082. Mefisto Posted: March 24, 2020 at 09:29 PM (#5933257)
This discussion of the top 10 made me realize that BBREF has updated its data, including recently, and I haven't kept up. I went back and re-calculated my top players and now the modern players look better. F. Robby moves into the top 10 and Schmidt is now tied for 10th. Morgan and Pujols are 13 and 14. Chronologically, the top 14 are now Cobb, Ruth, Williams, Musial, Mays, Mantle, Aaron, F. Robby, Morgan, Schmidt, Henderson, Bonds, ARod. Remember that I rank pitchers separately.

This makes my priors tingle.
   1083. Mefisto Posted: March 24, 2020 at 10:21 PM (#5933261)
Oops. Left Pujols out of my chrono list. He should come after ARod (duh).
   1084. alsep73 Posted: March 25, 2020 at 10:16 AM (#5933355)
Number 9 is Stan Musial. Between Mantle at 11 and this, it's clear the numerology gimmick is out the window when we get this high up.

The big remaining questions: 1)How high does Charleston finish? 2)Is Ruth the top player, or is it Mays? Or is it Charleston, so Joe can make one last grand statement about the Negro Leagues?
   1085. DanG Posted: March 25, 2020 at 11:45 AM (#5933380)
Ultimately, I would want to pick a pitcher who could have been a great pitcher in the conditions over the past 100 years. So in my book, I would put the Big Train just outside the top 10 overall
How does Johnson compare to other pitchers at ages 32-38? Here's everyone with 30 Pitching WAR at age 32-38 since 1919:

Rk         Player  WAR WAAERA+   W   L From   To
1   Randy Johnson 53.0 40.1  175 125  42 1996 2002 H
2     Lefty Grove 52.9 36.4  149 125  63 1932 1938 H
3      Bob Gibson 47.9 32.1  137 123  76 1968 1974 H
4     Dazzy Vance 46.5 32.0  133 129  78 1923 1929 H
5     Phil Niekro 45.1 28.6  123 112  95 1971 1977 H
6   Gaylord Perry 43.9 25.0  127 128 103 1971 1977 H
7   Roger Clemens 42.7 28.5  140 108  52 1995 2001
8  Curt Schilling 41.6 30.1  136 108  54 1999 2005
9  Pete Alexander 39.9 25.7  131 123  79 1919 1925 H
10  Steve Carlton 37.9 22.2  130 132  74 1977 1983 H
11   Warren Spahn 37.6 18.7  127 145  81 1953 1959 H
12    Kevin Brown 36.0 25.3  151  92  47 1997 2003
13      Red Faber 33.9 18.0  124 100  81 1921 1927 H
14    Jim Bunning 32.7 16.4  119 101  85 1964 1970 H
15 Walter Johnson 32.5 16.0  123 115  82 1920 1926 H
16     Luis Tiant 32.2 16.3  116 119  76 1973 1979
17  Bert Blyleven 31.2 16.4  115 102  81 1983 1989 H
18 Jstn Verlander 30.4 21.3  147  73  40 2015 2019
19    Greg Maddux 30.4 17.9  137 121  66 1998 2004 H
20   Carl Hubbell 30.1 16.4  125 117  66 1935 1941 H
21   Mike Mussina 30.0 17.0  112 103  63 2001 2007 H 
   1086. Ron J Posted: March 25, 2020 at 12:23 PM (#5933397)
Not sure how much Walter Johnson's results in live ball tell us how peak Walter would have done in those conditions. He lost a fair amount off of his fastball in a 1920 injury and was a noticeably different pitcher.

Yeah all pitchers lose their stuff late (Bar Randy Johnson, Nolan Ryan, Dazzy Vance I guess). The point though is that we can't infer from his later results how the younger version would have done. He did take a bigger hit than most but also had to make a bigger adjustment than anybody on the list.

One thing we do know however. He threw a spitter in the teens. Denied a low effort, highly effective pitch there's no way he could have sustained the workloads he handled in his prime.
   1087. Ron J Posted: March 25, 2020 at 12:30 PM (#5933401)
The way I used to describe prime Walter Johnson was an extremely healthy (prime) Bob Tewksbury who could at will and without warning become Goose Gossage any time he needed to. Tewksbury was a perfectly serviceable pitcher and that trump card is huge.

Feel free to update the example with more recent examples. But prime Johnson was pitching to contact with quality strikes and had the ability to just completely change style.
   1088. JJ1986 Posted: March 25, 2020 at 12:30 PM (#5933402)
I'm going to guess: Cobb, Johnson, Aaron, Charleston, Ted, Ruth, Bonds, Mays.
   1089. TomH Posted: March 25, 2020 at 01:28 PM (#5933426)
RonJ, that's a fascinating analogy.

Re: Walter's place all-time, my point is not that he wasn't all that great in his 30s. My point is, he wasn't that great when home runs became a thing, but his home park totally masked that fact. And since 1950ish, home runs have been much more a thing than they were in the 20s.
   1090. Gch Posted: March 25, 2020 at 01:37 PM (#5933432)
Not sure how much Walter Johnson's results in live ball tell us how peak Walter would have done in those conditions. He lost a fair amount off of his fastball in a 1920 injury and was a noticeably different pitcher.


Not to mention he was still performing at an all-star level, led the league in FIP twice, and ERA/ERA+ once, all with approximately a gazillion miles on his arm.

Something I found out yesterday while I was looking at Johnson's numbers: 41% of the home runs he surrendered were given up to batters who are in the Hall of Fame... okay, it helps that he gave up 10 out of 97 to Babe Ruth, but that's still (AFAIK) a remarkable total.
   1091. Ron J Posted: March 25, 2020 at 01:40 PM (#5933436)
#1089 I don't precisely disagree with the point, just not sure it's valid in that he lost so much of his stuff around 1920. And had to make a bigger adjustment than most.

Still, I've made a similar kind of point with prime Koufax. Who took such extreme advantage of Dodger Stadium that he personally moved the park factor a couple of points.

Not sure how to deal with this. The ability to take extra advantage of the home park results in real world wins even if (say) Koufax would have been slightly less valuable in other contexts.
   1092. bbmck Posted: March 25, 2020 at 02:14 PM (#5933445)
Excluding the Senators, Hall of Famers hit 29.5% of the HR in the AL from 1907-1927, 1490 out of 5050. Babe Ruth hit 7.5% of the HR, 416 out of 5516.
   1093. Eric J can SABER all he wants to Posted: March 25, 2020 at 05:33 PM (#5933507)
1918 drags things down with George Sisler 6.8 leading to Heinie Groh and part-time Babe Ruth tied for 9th at 4.7 being among the lowest values since ~150+ games a season started.

This is at least in part because 1918 was shortened by the war to roughly 120-130 games in the regular season.
   1094. alsep73 Posted: March 26, 2020 at 11:00 AM (#5933696)
Today's Favorite Player is Roy Hobbs, which gives Pos an excuse to once again tear apart bumbling New York Knights manager Pop Fisher:

Fisher loved the game, but his managerial ineptitude is the stuff of legend. As one of his players said, “Pop was a nice guy, but he had no presence. Nobody really listened to him.” He also was known to hire a crank psychologist who would give long and bizarre talks to the players about the “disease of losing.”

When Hobbs showed up, the Knights were buried in last place and were in desperate need of any kind of spark. But Fisher initially refused to play Hobbs. He even refused to let Hobbs take batting practice. It’s hard to know the politics of what was happening; Fisher was in the middle of a nasty battle at the time with his co-owner, Judge Goodwill Banner, who had been a New York circuit court judge. Fisher might have been convinced that Banner had signed Hobbs as a trick to make him look bad — and so the team would lose more games, freeing Banner to buy out Fisher and own the team himself. Fisher, in addition to his other failings as a manager, was also a bit paranoid.
   1095. alsep73 Posted: March 27, 2020 at 08:44 AM (#5933958)
8. Ty Coibb
   1096. alsep73 Posted: March 27, 2020 at 09:28 AM (#5933970)
8. Ty Coibb


AKA Ty Cobb.

As a ballplayer, Ty Cobb was his own species, his own category, a snarling, whirling, cunning, brilliant hit machine who won 11 or 12 batting titles back when that was more or less the only thing that mattered in baseball. He stole home 54 times in his career. Fifty-four! He retired with a .367 batting average (since adjusted to .366 after a couple of hits were found to be phantoms) and it is not just a record that will never be touched, it is an absurdity. Nobody has hit .367 in a season for 15 years.

Last year, Christian Yelich won the National League batting title at .329. For 19 seasons, from 1909 to 1927, Cobb never hit as low as .329.

But that’s nothing. From 1909 to 1919, Cobb never hit as low as .367 in a season.

He was not just the dominant ballplayer of Deadball. He was the only ballplayer of Deadball who mattered. He determined exactly how the game was to be played, how the bases were to be run, how a batter was to swing, how a ballplayer was to compete, and all anyone else could do was follow. In 1920, when the baseballs were boosted and the spitball was outlawed, a big lug named Babe Ruth took the game away from Cobb, made it more about hitting baseballs over fences. Cobb never quite forgave him (or the baseballs) for that.

“Nothing means much today,” he grumbled years after he retired. “Some second-rate hinky-dinky can come up and pop a ball clear to the fence, a hit that with the old ball would have barely got out of the infield.”
   1097. gef, talking mongoose & vexatious litigant Posted: March 27, 2020 at 10:20 AM (#5933986)
But that’s nothing. From 1909 to 1919, Cobb never hit as low as .367 in a season.


I know all sorts of important factors are no longer the same, but .367 is an excellent week.
   1098. Howie Menckel Posted: March 27, 2020 at 10:27 AM (#5933991)
8. Ty Coibb

there's no 'I' in Ty Cobb, said no one ever.
   1099. Mefisto Posted: March 27, 2020 at 11:18 AM (#5934007)
Hah. My Bingo card is looking pretty good (though I fudged a bit).
   1100. Booey Posted: March 27, 2020 at 11:29 AM (#5934013)
#1096 makes me realize how much I miss the days when players still cared about batting average. It's obviously not the best stat for determining value, but it sure is one of the most fun, and the game is significantly less exciting now that it's gone out of vogue.

No one's hit .350 in 10 years. Only 10 active players have a lifetime BA over .300, and I'd bet on only 2 of them retiring above the mark (Cabrera, Altuve).
Page 11 of 13 pages ‹ First  < 9 10 11 12 13 > 

You must be Registered and Logged In to post comments.

 

 

<< Back to main

News

All News | Prime News

Old-School Newsstand


BBTF Partner

Dynasty League Baseball

Support BBTF

donate

Thanks to
Sheer Tim Foli
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Hot Topics

NewsblogPirates pitcher Steven Brault releases showtunes album ‘A Pitch at Broadway’
(20 - 1:41pm, Apr 05)
Last: AndrewJ

NewsblogUS-based pro sports leagues monitoring coronavirus outbreak
(4271 - 1:38pm, Apr 05)
Last: tshipman

NewsblogFields of green: Mild winter, no games aid ballpark grass
(1 - 1:23pm, Apr 05)
Last: Russlan thinks deGrom is da bomb

NewsblogRinger:The Strangest, Funniest, and Most Fascinating Baseball-Reference Pages
(38 - 1:16pm, Apr 05)
Last: Ziggy is done with Dominican discotheques

NewsblogBaseball Question of the Day: What’s your favorite baseball movie?
(65 - 1:05pm, Apr 05)
Last: Gch

NewsblogWho Would You Build an MLB Team Around in the 2020s?
(13 - 12:56pm, Apr 05)
Last: What did Billy Ripken have against ElRoy Face?

NewsblogCoronavirus: MLB debating playing games at spring training ballparks without fans, per report
(3 - 12:51pm, Apr 05)
Last: What did Billy Ripken have against ElRoy Face?

NewsblogAn Activist and a Bookworm, Sean Doolittle Is the Conscience of Baseball
(24 - 7:43am, Apr 05)
Last: Jarrod HypnerotomachiaPoliphili (TeddyF.Ballgame)

NewsblogJudge nixes fantasy players' lawsuit over sign-stealing scandal
(4 - 2:39am, Apr 05)
Last: Sunday silence

NewsblogBaseball Question of the Day: Which historical game would you want to see?
(51 - 12:31am, Apr 05)
Last: Cleveland (need new name) fan

Hall of Merit2021 Hall of Merit Ballot Discussion
(205 - 7:38pm, Apr 04)
Last: Jaack

NewsblogThe Official Mets Fan Self-Immolation Thread
(953 - 6:37pm, Apr 04)
Last: Howie Menckel

NewsblogBryce Harper, wife Kayla donate $500,000 to coronavirus relief
(25 - 5:30pm, Apr 04)
Last: Baldrick

NewsblogPosnanski: Baseball 100 Rules
(1203 - 4:43pm, Apr 04)
Last: alsep73

NewsblogOT - NBA Thread 2020
(2031 - 3:36pm, Apr 04)
Last: BaseballObscura

Page rendered in 0.8854 seconds
46 querie(s) executed