Since Mike Webber suggested we talk about CFs, and since I vote for like two dozen every year, I figured I'd start things off with a quick list of how I personally rank the major CFs we've come across. Homers in (). Just because someone comes in numerical order after another doesn't mean they aren't, in many cases, real real close.
(Cobb)
(Charleston)
(Speaker)
(Hines)
(Gore)
(Torriente)
(Simmons)
(Stearnes)
(Hamilton)
1 Duffy
2 GVH
3 Poles
4 Averill
==ballot line==
5 Roush
6 Leach
7 Ryan
8 Bell
==moving, historical consideration set line==
Browning ????
==========================================
(Max Carey)

9 R Thomas
10 Berger
11 Seymour (with pitching)
12 F Jones
13 H Wilson
14 Beaumont
15 Milan
16 Griffin
17 Hoy
18 Stahl
19 Combs
20 L Waner

If I remember my upcoming candidates correctly, only Gene Benson seems like he could slip onto this list.

Also, I feel like I'm forgetting a NgL candidate as well, hope not.

I should have noted upcoming to mean, like, before 1960ish.

I am heading out real soon but here are my CF rankings.

Duffy #8)
Averill #12
GVH #16
Roush #20
Bell #24
Berger #30
Ryan
Thomas
Wilson
F. Jones
Poles

Poles may be higher and I haven't looked at him for a while. I originally ranked him this low because of his offense. I have a peak heavy system but looking at my CF list it doesn't seem to be as off balance, with GVH and Bell rated high, as some other positions. Oh, and the numbers may be a little off.

I’d like to introduce you to my best friend Edd Roush.

Edd stacks up well in Win Shares among the back loggers.


What separates Roush from the other outfielders on this list, is his big seasons. He has three seasons of 30+ MVP Type Win Share seasons (33,33,30). Duffy also had a similar peak, but the others don’t really have a similar string of MVP type seasons.

What about WARP?

Leading career back-loggers by WARP-1:
-old new-


This list is middle infielder heavy, with Leach still ahead of Roush (I like Leach too!) plus Beckley and Ryan depending on which week’s version of WARP you use.

One more list: Top 30 centerfielders by RCAA per the Sinnins encyclopedia



Roush isn’t in the group with the top 10, but he is ahead of Ashburn, Van Haltren, Ryan, Berger and Duffy.

Defensively Roush is an A- centerfielder, with 5 win shares gold gloves.

What are the arguments against Roush?
1.Weak League – No argument about the two Federal League seasons, but I will disagree about the NL of the period. I don’t see any reason to believe that the NL was top to bottom worse than the American League of the same period. Sure the Phillies were terrible, but the Braves of the period weren’t consistently poor. I think maybe the fact that there were many different NL representatives in the World Series leads makes some people think that the league was weak. Was the AL weak in the 1980’s? WARP3 unfairly penalizes Roush for this misperception – at least this week.
2.Missed a lot of games – Yes he held out a lot, and if he didn’t he would be around 350 win shares and a slam dunk. Roush admitted to playing baseball for money first, not a surprise for an athlete of today, but not openly admitted by many then. He was a child of the Federal Leagues era, where competition created a brief spike in salaries, which then shrank up. I won’t argue that his missed time is similar to the Negro Leaguers or War Credit, but even holding out he was a heck of a player whether he was home in Oakland City or in New York putting up with McGraw’s crap. He does deserve a small bump in 1918, he played 113 of his teams 128 games, netting 22 win shares – he led the league in OPS+ that season – a season not shortened by the war could have easily been another 30 win share season.
3.We have a lot of outfielders – True, but I think Roush is clearly a step ahead of backlog, none of them have the combination of his career length sprinkled with MVP type seasons of peak. Roush belongs in the group with Carey, Ashburn, and Bell.
4.Black and Grey Ink – Edd has 14 Black Ink points, 125 Gray Ink, while the averages are 27 and 144. I think that Crosley Field may be a fairly large factor in this. Crosley consistently depressed offense 3 to 6% during Edd’s career. Edd was in the top ten in the league in OPS+ and Slugging 7 times, in the top 10 in OBA 6 times, and batting average 9 times, but only 4 times in RBI and only twice in runs scored. If the environment was little more neutral, he might have picked up significant gray ink in RBI and Runs Scored.

I know Edd is a long way from being elected, but this is what I am asking you to consider. Are you giving votes to Van Haltren, Ryan, Berger or Duffy? Could you compare my buddy Edd to you centerfielder, and see if there if maybe he is equally if not more meritous?

Are you giving votes to Willard Brown? Did the HOF election of Brown cement your thinking? Hey Edd is a Hall of Famer too. And even if you think the NL of his period was weak, it was still a stronger league than any Brown ever played in except the American League – where Brown hit .179. My argument is Edd was a dominant centerfielder in the majors, and Brown might have been.

So rally around the flag of Old Double D!
Help this overlooked Hoosier take advantage of the brief window of opportunity all back loggers have to become a member of the Hall of Merit.

This message paid for by Friends of Double D, Pat Moran – Treasurer.

He does deserve a small bump in 1918, he played 113 of his teams 128 games, netting 22 win shares – he led the league in OPS+ that season – a season not shortened by the war could have easily been another 30 win share season.

Easily? He needs more than 7 win shares in 25 games, a pace for about 45 in a full season. But, yes, he was a great player then. His HOM thread is much too short.

He does deserve a small bump in 1918, he played 113 of his teams 128 games, netting 22 win shares – he led the league in OPS+ that season – a season not shortened by the war could have easily been another 30 win share season.

Easily? He needs more than 7 win shares in 25 games, a pace for about 45 in a full season. But, yes, he was a great player then. His HOM thread is much too short.

Ok, Easily might be a stretch, but a good fielding centerfielder that leads the league in OPS+ is usually a 30 Win Share Player

Bill James in the NBJHA has Roush and Ashburn side by side, #16 and #17, in his CF rankings - in spite of James's extreme timelining, which we have all noted. The 1982 vote is straightforward, so let's take another look at Edd.

Holy cow! Edd Roush only has 7 posts on his thread? OK, 8? You'd think the double D's would be worth 8 posts right there.

I like Edd Roush a lot more than Van Haltren, Ryan, Browning or Duffy. He's still stuck off ballot at #29.

I would like to revisit the candidacies of Duffy, Waddell (welcome to the top 10, Rube!) and Roush.

And yes, Edd Roush should have a discussion in triple digits, not to mention double.

I am pretty sure that our discussion of Edd Roush are elsewhere.

the man deserves his own, no?

/namedrop

I was eating lunch with Bill James this week and he was telling me about a player in the Red Sox organization that they drafted that he had ranked higher than anyone else on his list on draft day a couple years ago. He suddenly became “his player.” He ends up defending this guy that he wouldn’t recognize had he walked into the room. I of course thought of Edd Roush – “my player”

/endnamedrop

1. Was he ever regarded as the best player in baseball? Did anybody, while he was active, ever suggest that he was the best player in baseball?

Much like Duke Snider or Derek Jeter, Edd Roush was usually the third best player at his position during his career.

2. Was he the best player on his team? – Yes, narrowly over Hall of Merit-inductee Heinie Groh

3. Was he the best player in baseball at his position? Was he the best player in the league at his position?

Speaker and Cobb were the best centerfielders, in baseball. The other great CF in the NL in Edd’s time was Hack Wilson.

From 1917 to 1921 Edd’s NL rank in OPS+ was 2nd, 1st, 3rd, 3rd,4th. Think about that you consecutive peak and peak voters, the gold glove center fielder finishes in the top 4 in the league in OPS+ 5 straight years.

4. Did he have an impact on a number of pennant races? –
1919 his team makes the World Series and of course wins. He likely would have been the MVP, as the Reds won by 9 games.

In 1926 he played well, but his team was 2nd by 2 games. The shortstops in Cincy killed them. In 1927 with the Giants, his team finishes 3rd, just 2 games out and the 34-year old Roush has a mediocre year. His teams had a couple of other 2nd place finishes, but distant seconds.

5. Was he good enough that he could play regularly after passing his prime?

He still played regularly at age 38.

6. Is he the very best baseball player in history who is not in the Hall of Merit?

Among the back loggers, I think so or I wouldn’t be working through this.

7. Are most players who have comparable statistics in the Hall of Merit?

According to his Baseball Ref List - he is fairly unique, closest match an 887, as are most players headed to the HOM. 2 of his top 10 are in Joe Kelley and Henie Manush. Pie Traynor and GVH are also on this list. Cuyler (HOF not HOM), is too.

To me Roush belongs in the group with Carey, Ashburn, and CP Bell, and slightly ahead of the Duffy, Van Haltren pairing.

8. Do the player's numbers meet Hall of Fame standards? -

He is a little short on black and gray ink, 14 of 27 and 125 of 144. I think that Crosley Field may be a fairly large factor in this. Crosley consistently depressed offense 3 to 6% during Edd’s career. Edd was in the top ten in the league in OPS+ and Slugging 7 times, in the top 10 in OBA 6 times, and batting average 9 times, but only 4 times in RBI and only twice in runs scored. If the environment was little more neutral, he might have picked up significant gray ink in RBI and Runs Scored.

9. Is there any evidence to suggest that the player was significantly better or worse than is suggested by his statistics? His park does hurt him offensively.

To illustrate what Crosley Field did to offense, here are a couple of charts.

First guess how many individual seasons a Red hit more than 20 homers in a season before 1950?

Did you guess more than 10?
Did you guess more than 5?

The answer is 3.

CINCINNATI REDS
SEASON
1876-1950

HOMERUNS YEAR HR
1 Hank Sauer 1948 35
2 Ival Goodman 1938 30
3 Ted Kluszewski 1950 25

Knowing that, how about this, how many season did a Red have more than 100 RBI between 1900 and 1950?
Just 8 times they had more than 100 RBI, four of them by Frank McCormick.

That environment can make it tough to rack up black and gray ink, or big runs and RBI totals.

10. Is he the best player at his position who is eligible for the Hall of Merit?

Yes. The centerfielders that have been getting support in HOM voting are Hugh Duffy and GVH.

OPS + - Edd 126, GVH 121, Duffy 122

OWP – Edd 622, GVH 620, Duffy 623

WS per 600 Plate Apps – Edd 23.10, GVH 20.25, Duffy 22.61
(no pitching win shares for GVH)

I like all 3 of these guys, but Roush is just slightly ahead offensively, and comfortably ahead defensively.

OPS+ - Edd 126, Carey 107, Ashburn 111

OWP – Edd 622, Carey 556, Ashburn 600

11. How many MVP-type seasons did he have? Did he ever win an MVP award? If not, how many times was he close?

Edd had 3 30+ Win Share seasons, he led the NL in WS in 1919, and would have been the MVP if an award had been given.

In 1924 and 1925 he finished 10th in the award voting. In 1926 (when the Reds narrowly missed a pennant) his teammate Hughie Critz finished 2nd in the MVP voting. Critz was last among Cincy regulars in both slugging and OBA except the horrific shortstops. I would have thought Hargrave and Roush would battle for the team honors, but well if you think MVP award voting is odd now…

12. How many All-Star-type seasons did he have? How many All-Star games did he play in? Did most of the players who played in this many All-Star games go into the Hall of Fame?
The All-Star game came after Edd, he had 7 20+ Win Share seasons.

13. If this man were the best player on his team, would it be likely that the team could win the pennant?

Yes, and in fact they did.

14. What impact did the player have on baseball history? Was he responsible for any rule changes? Did he introduce any new equipment? Did he change the game in any way?

Does not apply

15. Did the player uphold the standards of sportsmanship and character that the Hall of Fame, in its written guidelines, instructs us to consider?
Yes



On a closing note, in Kansas City last week I tried to get Dr. Chaleeko, KJOK and Ken Fischer to take apart Edd Roush, but they were just too damn polite to tell me why I am wrong (ok, its not like any of them were carrying around their Stats All-Time handbook).

I like Roush, and I have grown to have something of a personal attachment to him, but if someone wants to point out what why you are voting for Duffy or Van Haltren or Oms or Berger or Bob Johnson or Tommy Leach or Sam Rice or Jimmy Ryan or Minoso I’d be interested.

Guys like Keller or Kiner are different types of players, though I’d point out Kiner has just one more 30+ win share season than Roush. In fact their top 5 Win Share seasons go:

Kiner – 37, 35, 30, 30, 23
Roush – 33. 33. 30, 28, 23

Thanks for your time.

I like Roush more than the 1890s triplets but less than the 1940s-50 guys.
He could get back on my ballot near the end of the project...

Much like Duke Snider or Derek Jeter, Edd Roush was usually the third best player at his position during his career.

Speaker and Cobb were the best centerfielders, in baseball.

...and Charleston and Torriente. Oms was also a contemporary, and according to the MLEs was better than Roush in prime and career value, though Roush probably had an advantage in peak.

I am a fan of Roush, and vote for him regularly, but I cannot place him above Duffy. 2 essential reasons
1) Duffy had the peak - one of the best I have ever seen.
2) Defense - I find them at least equal. Maybe I overrate Duffy(?) - I'm not convinced that I am, though.
2(a) At the time they each played, Roush was behind Cobb and Speaker (nothing to be ashamed about). Who was Duffy behind?

Brent: Sure Charleston!
Which makes Roush the 4th best CF of his era, maybe more like Ashburn and Tejada than Snider and Jeter right?

Max Carey and Torriente would be in the arguement too. Carey's lack of peak leaves him behind to me, and I think with all the peak crazed voters we have now he would struggle if he hit the ballot today.

Torriente - obviously a great player, not as obvious as Charleston, so he might be ahead of Roush (even a lot ahead of Roush) or not. Hard to tell.

Oms was a corner outfielder, and has some undocumented career gaps - the Sugar Leagues - and just like Torriente comparing him to Roush is tough.

rawagman:

Duffy's peak is awesome, I vote for Duffy too.

Defense - Is Duffy a center fielder though? Games Left/Center/Right 574/676/437


1889 Chicago
Check out this team BTW, the outfield is Van Haltren, Ryan and Duffy, with Anson at 1b, Williamson at 3b and Fred Pfeffer at 2b, and what is team's record? 67-65 - 4th in the league in ERA - 5 potential HOMer you think better then 2 games over .500 right?

Anyway..if you say Duffy is a centerfielder, he's probably 2nd behind Hamilton, with his teammates Van Haltren and Ryan having some supporters.

but if he is a corner outfielder it get crowded - Delahanty, O'Rourke, Burkett, Keeler, Kelley, Thompson - plus several guys with some overlap like Clarke, Sheckard, Flick, Hines. There are always corner outfileders who can hit.

I rate Duffy as a CF in the same way that James (and us) rate Banks as a SS.

His value is highest there.
As I joined the project late, I have not done the research into the 19th C guys who were already elected, so I can't compare him to them. However, eyeballed, I suppose he's 2nd behind Hamilton as CF.

On the ballot thread Tom H wrote:

OWP – Edd 622, GVH 620, Duffy 623
I like Roush, and I have grown to have something of a personal attachment to him, but if someone wants to point out what why you are voting for Duffy or Van Haltren...

If you adjust for schedule length, Van Haltren has advantages of about 1500 plate appearances, AND 430 innings pitched. That's a big difference when comparing two guys of similar offensive value at the same position.

Very valid point, and in fact Van Haltren compares very well to Max Carey - has a similar career shape, and if the schedules were the same I think his total would surpass Carey easily.

When you look at Van Haltren's pitching do you ever wonder if he was spotted against the weak teams? I mean he's clearly the 3rd or 4th pitcher on his team in both starts and ERA. Maybe he was the Saturday double header 2nd game guy.

His pitching definately has value. Van Haltren's California/PCL play in 1904 and 1905 has some value too.

I don't want to run down Carey to me is very good choice - he has 351 career win shares, 92nd all-time. The only guy above 350 win shares that are not really slam dunks are Rusty Staub and Mickey Welch.

The Reds Hall of Fame/Museum-which is an absolute joy-has a wonderful Roush exhibit.

Mike Webber and rawagman:
How do you figure Duffy's peak was great? He had one Norm Cash-style fluke big year, one other good hitting year in the weak 1891 AA, and otherwise a bunch of Beckley-GVH-style seasons. Unless you think his fielding was tremendous (and rawagman isn't sure it was better than Roush's), I don't see how you get a great peak from that. Roush had at least 4 seasons (1917-1919 and 1923) offensively that were better than Duffy's second best season.

Yeah rawagman, how do you figure his peak is better than my boy Roush?? :)

Seriously

Age     WS      OPS+     Gm     Tm Gms   154 eqv
23      26      123     138     138      29.0 
24      28      147     127     139      31.0 
25      29      125     147     152      29.4 
26      28      125     131     131      32.9 
27      33      177     125     133      38.2 
28      23      126     130     132      26.8 

Ok if you let some air out of his age 24 season for a weak league, its still a nice run.

But I ain't here to argue for Duffy, I'm for Roush!

Duffy left, Roush right - seasons over 400 PA

Year.....Ag.....PA.....OPS +.....Year.....Ag.....PA.....OPS +
1894.....27.....606.....177.....1917.....24.....567.....159
1891.....24.....601.....147.....1918.....25.....492.....153
1897.....30.....621.....127.....1923.....30.....592.....148
1895.....28.....614.....126.....1919.....26.....572.....147
1892.....25.....673.....125.....1921.....28.....463.....143
1893.....26.....611.....125.....1920.....27.....650.....141
1890.....23.....657.....123.....1924.....31.....519.....134
1889.....22.....632.....109.....1926.....33.....631.....124
1898.....31.....645.....107.....1925.....32.....590.....124
1896.....29.....601..... 94.....1915.....22.....616.....123
1899.....32.....641..... 85.....1929.....36.....509.....108
................................1927.....34.....614..... 96
................................1931.....38.....400.....78
                                    
17.....yrs...............122.....18.....Yrs...............126
...............7827..............................8156    

Like Cblau said, after knocking off the top two seasons of each, there does seem to be about 5 Roush seasons that would beat Duffy's #3 season.

All right, here's my offensive system: context-scaled something-like RCAA, sorted best season to worst



This is not adjusted for season length (in games) and I'm not sure it should be. Notice the post above in which Duffy is normally getting more PA per season than Roush. In high scoring times, the lineup turns over more; a batter can have as many or more PA in a short-season, high-scoring league as in a long-season, low-scoring league.

Duffy's 1894 is a nice year, to be sure, and the best year on this list, but accounting for that 7+ R/G context takes a lot of air out of it. Someone called it a "Norm Cash" year. Not to me it isn't. Cash, 1961, weighs in at 100 on this scale.

I've always felt that the closer one looks at Duffy, the less impressed one is.

I don't vote for Roush, either, because the ton of missed games really bugs me. His rate stats for me are right at the HOM cusp, a little like Cash. The games on the bench are what kill both of them, although neither is that far off my ballot anyway.
I like both better than Duffy.

I think it's impossible to look at Duffy's actual career and give him a vote. Seems like he's the red-flag to any 'unified-system' theory, as he weighs in impossibly well on these.
The Duffy CF vs not CF discussion here is great. If Duffy really were as a great a CF as some stats try to claim, he'd have had 95 pct of his games there instead of 45.

And kudos to the Mike Webber observations that could both help and hurt a borderline candidate like Van Haltren...

Mike Webber - the difference for me in your chart is your cut off of 400 PA. If you cut it at 550 PA (closer to a full year, but still not even qualifying for a BA title today), Roush loses 4 seasons.

I know that Roush missed much of that time due to holdouts, and fair game to him for sticking up for himself, but he missed a fair bit if time.

As for fielding, my method is a little crude, but I generally combine a comparison of the player's fielding % and rate compared to league averages in his day, with a little contemporary opinion thrown in.
Both Roush and Duffy score very highly in that system. I give Roush my highest possible grade, Duffy, one step below.

To recap, both are similar. Duffy trumps for me based on his peak. I rate their careers fairly equal. And I don't vote on prime.
This "year," Duffy is 3rd on my ballot, Roush is 10th. I would love for both to be included in the final tally.

I don't have time to enter all the data currently, but I encourage voters to look at BBRef's team pages for Duffy - especially the Boston years. Duffy has great range numbers, better than league average even when playing left. He often has a higher range than the regular centerfielder - Hall of Meriter Billy Hamilton. Also, his fielding percentage compared to the league average is almost always in his favor. Also, his assists to errors is positive in most every year he is a regular.

Does anyone know if Boston's field in the 1890s was one of the many that were asymmetrical - like Yankee Stadium for many years, or Fenway was originally - where a team needed two centerfielders? That would help explain Duffy's high range totals.

I believe Duffy moved to left because Frank Selee saw that he needed two centerfielders. Duffy's numbers do not show any decrease that might be expected by a move from center to left.

Those are my two cents.

Oh, and I think Roush was totally jobbed by the voters when we elected Carey. In my opinion, one of our biggest mistakes. Despite being 26th on my current ballot, he is a PHOMer and could easily make my ballot again. Roush, Fox, Cepeda, Cash, Elston, and the left field knot of Kiner/Minoso/Howard/Burns are going to be reexamined again come August.

5. Paul Wendt Posted: April 20, 2006 at 03:26 PM (#1977173)
. . . yes, he was a great player then. His HOM thread is much too short.
8. sunnyday2 Posted: July 10, 2006 at 11:58 PM (#2094997)
Holy cow! Edd Roush only has 7 posts on his thread? OK, 8? You'd think the double D's would be worth 8 posts right there.
. . .
10. rawagman Posted: July 11, 2006 at 10:52 AM (#2095278)
. . . yes, Edd Roush should have a discussion in triple digits, not to mention double.

Mission accomplished.

Mike Webber, sabr-kc in my eddressbook. Is there a Cinci-KC connection? Yes, Covington had a franchise in the Federal League but it was a business failure, replaced by KC before going major in 1914.
sabr-kc wrote:
3. Was he the best player in baseball at his position? Was he the best player in the league at his position?

Speaker and Cobb were the best centerfielders, in baseball. The other great CF in the NL in Edd’s time was Hack Wilson.

I think you need to put Max Carey here.

7. . . . Joe Kelley and Henie Manush. Pie Traynor and GVH are also on this list. Cuyler (HOF not HOM), is too.

Manush is HOF not HOM.

9. . . . CINCINNATI REDS
SEASON
1876-1950

"1876-" is a throw-away reference regarding Crosley Field, I know, but you asked for it.

Lip Pike


It looks like Cincinnati boosted a declining home run king.

And last, because pennant race analysis is out of favor here.
4. Did he have an impact on a number of pennant races? –
1919 his team makes the World Series and of course wins. He likely would have been the MVP, as the Reds won by 9 games.

In 1926 he played well, but his team was 2nd by 2 games. The shortstops in Cincy killed them. In 1927 with the Giants, his team finishes 3rd, just 2 games out and the 34-year old Roush has a mediocre year. His teams had a couple of other 2nd place finishes, but distant seconds.

7.0 and 4.5 games behind in 1922-1923 compared to 2 games in 1926-1927-and-1928. Granting that pennant race analysis belongs, it can't be right to dismiss all but '27-28 as distant seconds.

New York to Cincinnati in 1917, that was certainly a move in the wrong direction for the moment. In Cincinnati 1918-1926, it was a strong team with one horrible season in 1921. In New York 1927-1928, Roush was mediocre --simply because of age (it can't be leaving Crosley Field)? In 1929 like Cin 1918, the team was strong but Chicago was dominating. New York 1930 was a close third but without Roush, essentially done, finale back in Cincinnati again, now a weakling team.

Why did he play only one-third of the 1922 and 1928 seasons?

How long were all these teams contenders?

--
1889 Chicago
Check out this team BTW, the outfield is Van Haltren, Ryan and Duffy, with Anson at 1b, Williamson at 3b and Fred Pfeffer at 2b, and what is team's record? 67-65 - 4th in the league in ERA - 5 potential HOMer you think better then 2 games over .500 right?

What did they get for Clarkson? $10000? Of course, they wouldn't have won a pennant with Clarkson, only a lot more games and maybe more fans. Boston didn't buy a pennant with that $10K, on top of $10K for Kelly a couple years earlier. Spalding & Bros. probably opened a few more stores or bought a rival.

--
rawagman:
the difference for me in your chart is your cut off of 400 PA. If you cut it at 550 PA (closer to a full year, but still not even qualifying for a BA title today)

it isn't? It's about 3.4 per 162 games, 3.6 per 154.
What do they say in Londonstan, probably not "my indiscretion beats your indiscretion"?

if he is a corner outfielder it get crowded - Delahanty, O'Rourke, Burkett, Keeler, Kelley, Thompson - plus several guys with some overlap like Clarke, Sheckard, Flick, Hines. There are always corner outfileders who can hit.

Hines, CF

"In 1925, Ty Cobb and Tris Speaker each played his 17th season as a regular centerfielder in the Major Leagues, which surpassed Paul Hines 16 seasons, 1874-1891." Whose in Center? Long Careers beginning before 1920

Oms was a corner outfielder, and has some undocumented career gaps - the Sugar Leagues - and just like Torriente comparing him to Roush is tough.


While we don't have precise data on outfield positioning for the Negro League/Cuban League players, if you review the data listed in the Oms thread you'll see that he played more center field than anywhere else. During his 1921-30 prime, Holway lists him in CF 6 seasons, LF 1 season, RF 1 season, and no position listed the other two -- and data posted by Gary A indicates that the one season in RF is in error; Oms actually played all games in center field that year (1928) and led the east in range and in fielding percentage. Gadfly posted data that Oms was still playing center field in Cuban in the mid-1930s when Oms was about 40.

Because of Oms's extensive play in Cuba and the overlap of Cuban play with white Cuban major leaguers, I think his MLEs (at least the latest version of them) are among the most accurate we have. Oms was compared to Paul Waner and his MLEs at the same ages are actually very close to Waner's. You're right that his statistical record has a few gaps, but that's true for many or most Negro League players and I think to treat them fairly we have to extrapolate a bit to fill in the gaps. Although I can see a peak argument for Roush, on the whole I think Oms is the best eligible center fielder from the 20th century.

I want to say that Duffy may deserve some credit fo rplaying on teams that really overrachieved not just according to their runs scored and runs allowed but also according to their expected rusn scored and runs allowed. If you buy that players deserve soem credit for their teams 'extra' wins, like Win Shares does, then Duffy, and the rest of the 1890's BoStockings, will look better. Duffy's WS peak is mighty impressive, almost as good as Jennings in my system. Of course he wasnt' as good as Jennings, but there is a reason to think he had a very good to great peak.

On Roush,

I have stated this before, his inability to play full seasons, for wahtever reason, is what has hurt him in my system. I do beleive that he was better than Carey and that Carey was a 'mistake'. Of course I had Carey at #6 when he was elected so he is a 'mistake' that I contributed to quite a bit. I have Roush a few spots above Carey, if Roush had played full seasons instead of 4/5 seasons, then he would be solidly ahead of Carey.

Also,

I am not so sure that Jeter will go down as the 3rd best SS of his generation as I think he will look better than both Tejada AND Nomar. Even on peak, Jeter isn't a slouch and was the game's best in 1999. And being behind ARod is like being behind Cobb or Mays, it really shouldnt' be a factor.

A lot of Roush's missed time was due to holdouts. He wanted more money. He would never go to spring training.
I don't want to take points away from Roush for missing time, as I feel that he would have had more ink and whatnot had he played full seasons. Punishing him for missing time would be a double strike.

Wags,

Was that annwer to my post? If so I am not double counting his missed time since i dont 'relaly look at black and gray ink much. The reason that Roush gets docked for his missed time shows up in the fact that his best seasons don't have as much value as they would have if he had played most games per year. sorry if I wasn't more clear on that earlier.

schmeag - 'twas.
It makes sense either way.

I was trying to clarify how less time played is a symptom but not a cause.

But let's be clear on one big thing. The problem with Roush is not ultimately that he's better or worse than Ryan/GVH/Ashburn, et al. That's all debatable stuff. What's not debatable is that he's a borderline candidate, no matter how you cut it. If he wasn't he'd be in already.

One quick way to look at it. WS sees Roush as the best position player in the game once (we'll call this the MVP below), but also as one of the top ten position players four other times. It also sees him as the best CF thrice and the second-best three times (we'll call this All-Star below). That's pretty good and it puts his prime in possibly electable territory.

Here's a WS-centric look at some other players in history who
a) may or may not have a similar pattern of excellence
b) are competitors in the backlog
c) are future potential backlog competitors



Roush is obviously getting votes for good reason. Among HOMers, he looks like a more dominant player than Ashburn, but it's hard to say he dominanted his leagues more than anyone else on this list, with the possible exception of Jennings, who he runs neck and neck with. In addition, several of the current HOMers own a considerable career-value advantage. Roush seems to be running among the lower tiers of HOMers. Even within this group, he's slotting somewhere among the last four guys on the list, none of whom are showing great dominance over league or position. And he gives back a lot of value on the career side to several guys in the HOM ranks.

Among backloggers, Duffy, Browning, and Berger present more formidable dominance credentials than Edd. Not a ton more, but more. Wilson's very close to Roush. Like Jennings above, it's close and depends on your POV. Ryan and Thomas are also extremely close. GVH falls back---Roush doesnt' lap him, but he's obviously got some advantages in this area of analysis. Only Ryan and GVH exceed Roush in career value, while Duffy's career value is comparable. Roush mostly holds his own in this group by balancing off his career value against others' slightly more dominant primes. He doesn't stick out as obviously better.

In the future eligibles group, I wanted to run several moderns, but their careers are not yet far enough along to present as a good point of comparison. The two I could include are Jimmy Wynn and Willie Davis. Davis looks like a simple second-tier star by this analysis, but Wynn might have a little more to offer dominancewise. Despite lacking the MVP year, The Toy Cannon has a couple more years of being his position's best player. It's very close between him and Roush and though Roush has the career advantage, it's not huge. Anyway, Roush figures well here.

This is just one piece of the puzzle, but the issue is that Roush is not as dominant as middle and lower-tier HOMers, which is, I think, being recognized by the electorate in addition to his missing much time. He does present some career/outside-prime value that some other candidates don't, but then he presents less of it than some similar candidates too. Ultimately I don't like to contradict Mike since he is a really nice guy and he's smarter than I am about baseball, but I don't see much to distinguish Edd from the knot of very good CFs around him. How you measure peak, how you value it in comparison to career, and what you think about holdouts is ultimately going to push Roush further up or down your ballot than anything else. But all that said, I do think he's being given a fair shake by the electorate.

I’ve been a long-time Friend of Edd Roush, and I’m happy to see his discussion thread getting some attention, but the following comparison between Roush and a contemporary NL OF, George Burns, has always concerned me:

WARP1 Best 1/3/5/7/10/total
Roush 12.1/31.7/48.4/63.8/84.5/107.9
Burns 12.2/34.1/51.9/67.8/87.6/98

Win Shares Best 1/3/5/7/10/total
Roush 33/96/147/191/250/314
Burns 34/97/146/193/257/290

As the argument for Roush is mostly based on his peak/prime, I have trouble seeing how either WARP or Win Shares supports a vote for Roush over Burns. Roush was a better hitter—or at least had more power—and his rate stats are better than Burns’s, but Burns appears to have been a better baserunner and a better LF than Roush was a CF. And while Big Edd rarely played more than 140 games a season, Silent George rarely missed a game between 1913 and 1923.

I like both Roush and Burns more than most, and I have justified placing Roush ahead of Burns on the assumption, possibly wrong, that when playing Roush was a better player than Burns, but I don't see much difference between the two. And though I have supported Roush's election to the HoM, when it isn’t apparent that you were a more valuable player than an exact contemporary at your position who is currently receiving zero HoM votes, well, as I said, it’s a concern.

Doc,

Decent Analysis on Roush. I would say that by looking at that chart it confirms to me that Jimmy Wynn, someone I ma likely to support when he is eligible, is going to be a serious contender for the HOM. Five times as one of the ten best players in baseball, five times the best CFer and 8 times as one of the top 2 CFers (if I read your chart right) is pretty similar to Averill's 4,7,10.

Here's a comparison of nine prime seasons of Roush and Oms. For Roush, I'll use ages 24-28 and 30-33, his 1917-21 and 23-26 seasons (in 1922 he missed most of the season due to holding out and injuries). For Oms I'll use ages 26-34, his 1921-29 seasons, based on the Oms MLEs compiled by Chris Cobb.

As hitters, in their prime they were very similar. Roush had a slight edge in SLG+, but I think it's offset by the games he missed. Also, age 26 is the first season for Oms for which we have statistical data, but we know he played professionally as early as age 22 so I think it's reasonable to assume at least one prime season before age 26. Oms continued to play at a good major league level into his early 40s, so he has the career bulk too. On fielding win shares lists Roush as an A- outfielder, while Chris Cobb concluded that Oms was probably an A. Roush does have a slight edge on peak (best season with OPS+ of 159 compared to 153 for Oms), but overall I see most of the advantages going to Oms.

Five times as one of the ten best players in baseball,

jschmeagol,

my bad, let me clarify: that's only in his own league. he's among the 10 best in his own league.

Five times as one of the ten best players in baseball, five times the best CFer and 8 times as one of the top 2 CFers (if I read your chart right) is pretty similar to Averill's 4,7,10.

It's all within his own league only. But yes, I don't think it's unreasonable to compare that favorably to Averill's record. The difference I see is that Averill competed against Earle Combs and Joe Dimaggio while Wynn competed against Tommie Agee and Willie Davis. It's perhaps a subtle difference, but enough that I personally would rank Averill higher than Wynn.

Sorry, everyone, that I didn't point out the in-own-league thing previously.

Brent,

can I ask you a couple of questions?

While I agree that if the MLE's are good that Oms 15 game per season advantage is probably enough to trump Roush's slight rate advantage

A) How do you reconcile that Oms durability advantage comes from playing every game of a 37-game schedule. (27-28 for example)

B) While I believe MLE's work, when you have two players that are close, how do you put your faith into the MLE rather than the actual hard numbers?

I just can't work around those two items myself, its what held me back on Willard Brown.

Thanks.

Mike,

A) We know that although Negro League players had relatively few official league games, they were also playing many unofficial games - against semipro teams, non-league teams, industrial league teams, etc. Most historians of the Negro Leagues say that they were playing 200+ games per year (during the summer). The same sort of thing went on in Cuba during the winter -- official league games in Havana interspersed with unofficial games at sugar mills in the countryside. So I don't think it would be accurate to characterize Oms as playing a 37-game schedule, even if that's the number of official games.

But I'll certainly agree that the spottiness of the records make it hard to determine which Negro League players missed games due to injuries and which ones were very durable. The small rosters made it hard for them to miss games, but some serious injuries were recorded. However, the fact that these guys often moved from summer team to barnstorming team to winter team can make it hard to figure out whether missing games are a durability issue or just the result of a scheduling conflict. Bottom line - the playing time assumptions used in the MLEs involve some guesswork and are certainly less reliable than the MLE rate statistics. But I'm not aware of any significant injuries or durability problems for Oms.

B) If one always favors the player with the more reliable statistics (i.e., the white player) whenever the data are close, a bias is introduced that favors white players for the HoM. Personally, I refuse to introduce that type of bias. I strongly prefer to live with the extra uncertainty that's associated with poorly recorded Negro League records rather than introducing a bias that works against the Negro League players.

I think that trusting the white player's stats more in the Roush/Oms case could actually benefit Oms. Assuming that you trust our MLE's when it comes to a player's performance, you have to ask if you trust its playing time estimates. If you do not then you cannot tell if Oms missed 10-15 games a year thorugh niggling injuries and/or was the type of player to hold out, or had some trouble agaisnt tough RH/LHers, etc. But we do nkow that Roush happened to have low in season game played totals for someone as good as he was. Therefore you could say that there is still a better chance that Oms wasn't missing games like Roush in season since we nkow that Roush was missing games and we can't tell if Oms was. The chance of Oms missing games is say 50/50 (probably more like 20/80 since most players of this caliber do not si tout 10-15 games a year) but 100% for Roush. Does that make sense?

Of course this is very specious analysis anyway.

Reposted from the Jimmy Wynn Thread:

Dr. Chaleeko Posted: August 23, 2006 at 01:46 PM (#2153981)
Here's an interesting meta list for me. Until very recently my 3/5/10/15/career system was telling me this about MLB-only CFs trhough 2005:

1 Cobb, T
2 Speaker, T
3 Mantle, M
4 Mays, W
5 DiMaggio, J
6 Hines, P
7 Gore, G
8 Snider, D
9 Hamilton, B
10 Duffy, H
11 Doby, L
12 Browning, P
13 Averill, E
14 Griffey, Jr
15 Van Haltren, G
16 Roush, E
17 Ryan, Ji
18 Wynn, J
19 Ashburn, R
20 Carey, M
21 Thomas, R
22 Berger, W
23 Jones, F
24 Murphy, Dl
25 Edmonds, J
26 Williams, Be
27 Wilson, H
28 Pinson, V
29 Seymour, C
30 Puckett, K

A reasonable list based on the comparison criteria.

Now I've been working up another list where I combine Uber stats and keltner quesitons to create a HOM score scaled from 1-90. At 20 you become a candidate, at 30 you're borderlin, at 40 you're getting very hot, 50 and up you're a lock.

NAME POINTS
----------------------------
1 mantle, mickey 87
2 mays, willie 87
3 cobb, ty 86
4 Speaker, Tris 81
5 dimaggio, joe 74
6 Hamilton, Billy 66
7 hines, paul 62
8 Griffey Jr., Ken 54*
9 Snider, Duke 48
10 gore, george 46
11 roush, edd 43
12 Browning, Pete 43
13 duffy, hugh 42
14 doby, larry 36
15 ryan, jimmyh 36
16 averill, earl 33
17 berger, wally 32
18 wynn, jimmy 31
19 Van Haltren, George 30
20 carey, max 29
21 Wilson, Hack 29
22 puckett, kirby 27
23 Ashburn, Richie 27
24 murphy, dale 26
25 williams, bernie 26*
26 jones, fielder 24
27 Spence, Stan 24
28 Butler, Brett 23
29 t-cedeno, cesar 21
t-thomas, roy 21
t-edmonds, jim 21*
t-lofton, kenny 21*
*=active


The second list above captures the 3/5/10/15/career information as about 1/9th of the total assessment. It also captures MVP type seasons, all-star type seasons, comparability to HOF careers, 3-year best in league and best at position, past-prime performance, likelihood of a player pushing his team to victory. Eight categories, summed, each weighted at 10 points, but with one 20 point category.

This second list has a lot more contextual information than the simple 3/5/10/15/career gague I've been using. An intersting case is Hamilton. He leaps up three slots in the second list because he was really dominant over his leagues and his position. The 3/5/10/15/career tool is not sensitive enough to capture that important information. On the other hand, Stan Spence is propelled too far upward because he had a dominant stretch during the war when competition was thin. But that's why you have to review these things. Look at Edd Roush. He's down in the mid teens in list one, but when you take his various contextual performances into account, he moves up like Hamilton does. Mabye I shouldn't have been so hard on Mike Webber?---see Mike, I'm open minded on this. Who gets dinged? GVH mostly.

Once the NgLers are introduced, it's going to be dicy, of course. And I haven't gotten my re-eval of them underway yet, and I don't know how to begin incorporating them into a system with a great deal of real MLB context in it. It'll be fun trying!

I do expect that this new look will have an effect on my voting in the near future.

Reposted from the Jimmy Wynn thread:
91. sunnyday2 Posted: August 23, 2006 at 02:21 PM (#2154010)
I have what I call the Reputation Monitor. I use it to build consideration sets, not as a final thing, because as the name suggests it's about capturing information about a player's reputation, what some writers might say is "whether he looks like a HoFer or not."

The inputs include Win Shares, both career and peak; TPR (Linear Weights); Hall of Fame Monitor and Standards; Black and Grey Ink; OPS+ or ERA+; defensive WS, counting double IOW, for position players, and hitting WS for pitchers; plus adjustments for competition (AA, US, FL, wartime), for time missed to war, and position (catchers and relievers only). Some of the raw numbers are /2 or -100 or etc. in order to make all of the inputs of relatively equal weight.

A 300 is inner circle; a 200 score is generally a HoF lock; a 175 is a good candidate (i.e. show cause why not); a 150 is a candidate (show cause why...), below 150 probably not, probably a mistake if elected, below 100 is an absolute joke of a candidate much less electee.

I have not evaluated players active since about 2000. Griffey probably comes in right behind the inner circle, probably not in the inner circle.

Again my final ranking is different than this, this is the consideration set basically. (It works better at the very top, it is not as good at differentiating the mid-ballot guys.)

CF

Inner Circle

Cobb 418
Mays 411
Speaker 376
Mantle 368
DiMaggio 328
(Charleston is in here somewhere)

Lock

Snider 249
Puckett 211--aside from the inner circle, TPR has Puckett ahead of everybody but Wynn and Ashburn, and his combined HFM, HFS, BI and GI is pretty good; no real weaknesses in this system

Show Cause Why Not

Dawson 196--a HoMer in my book
Hamilton 195
Browning 187
Hines 186
Roush 184
Carey 176
Wilson 176--e.g. pretty borderline, one-dimensional
Gore 175
(Torriente probably near the top of this group)

Show Cause Why

Doby 173--e.g. added NeL value, that's why
Duffy 170
Murphy 168
Averill 168--e.g. MiL MLE?
Ashburn 162--e.g. extraordinary defense
Berger 155--e.g. great peak
Lynn 154
(I have no idea if CP Bell would come in here or maybe down with Earl Combs... Oms maybe in here somewhere and W Brown, too, though I rate him more with the RFers, and maybe Pete Hill though I rate him with the LFers)

It's a lot easier in my mind to show cause why for these guys than to show cause why not for the 175+.

HoVG

Pinson 148
Seymour 145
Cedeno 144--you can probably show cause why Cedeno and Wynn are as good as the 150-175 but you've used up your causes and you haven't gotten them into the 175+ group (IMO)
E. Davis 139
Beaumont 136
GVH 136
Oliver 134
Wynn 133
Otis 131
D. DiMaggio 128--with WWII credit
Ryan 126
(I have a feeling Poles would be in the bottom reaches of the HoVG)

HoG

Combs 120--ha ha ha

110-119: Chet Lemon, Griffin, Butler, Van Slyke, Roy Thomas, Dykstra, W. Davis (not necessarily an exhaustive list but close)
100-109: Chili Davis (nowhere near an exhaustive list)
<100: Lloyd Waner

Reposted from the Jimmy Wynn thread:
85. andrew siegel Posted: August 23, 2006 at 11:19 AM (#2153812)
CF's:

(1) Mays
(2) Cobb
(3) Mantle
(4) Speaker
(5) Charleston
(6) DiMaggio
[gap]

(7) Hamilton
(8) Snider
(9) Griffey
(10) Torriente
[gap]

(11) Doby
[gap]

(12) Hines
(13) Gore
(14) Averill
(15) Edmonds
(16) Roush
(17) Wynn
(18) Duffy
(19) Ashburn
(20) Van Haltren
(21) Bell
(22) Ryan
(23) Carey
(24) Berger
(25) Puckett
(numbers 12-35 are all pretty close)

I hope Doc, Marc, and Andrew don't mind me reposting their comments here, but hopefully in a couple of election cycles - it is "1984" as I type this - I think Edd being in the top 15 on these lists will be arguements in his favor. Especially since 2 of the three incorporate Negro Leaguers, and one includes active players.

Being the 15th best centerfielder ever is a pretty stong arguement in my book. Still borderline, but on the "in" side of the line.

Don't mind at all. But Charleston, Stearnes, Torriente, the fully MLEed Larry Doby, and Dubya Brown might. ; ) They probably all push Roush backward (Brown the least likely to do so). But even if all five were ahead, Roush would still be ranking at 16 or so on my list. I don't think there's any CF with war-credit issues close enough to give Roush trouble either, so that's a reasonably firm ultimate ranking by this system, I'd guess. Though to be honest, I've not turned the points into final ranks yet.

Dr. Chaleeko Posted: October 12, 2006 at 05:27 PM (#2208675)
Do you consider Edd’s two seasons where he missed significant time while battling for a fair contract similar to the problems Charley Jones faced? If not why? This isn’t something anyone really goes through anymore in MLB, you occasionally see it in the NFL though.

A fabulously interesting question. A lot of this question may have been subsumed by a somewhat offhanded remark that appeared in the BJNHBA (i think) in which Bill says that Edd didn't like spring training or some such notion of mild malingering. Perhaps he's quoting someone at that moment. Anyway, the substance of those comments has perhaps colored our view of that situation, but the question still stands. Isn't it realistically the same thing as Charley Jones? Or Tony Mullane? Or Home Run Baker? I think it likely is.

OK, so the question for Mike is, what years did he hold out, and for how many regular season games did the holdout last? That's the jubber meets the road. Put a number to it for each year, and let's see how many Shares that potentially cost him.

Gee, I’m glad you asked. ;)

I have a whole bio about Edd - but it barely covers his life post 1919. Wish it had more, so I may have to dig around.

Years that Edd could have some additional Credit:

War Years –
1918 played 113 of 129 team games. Missed the final four games of the season when his father fell off a telephone pole, the injuries would eventually kill him. Incidentally Roush missed a batting title by 2 points, and missed the final series of the season in St. Louis, the Cards led the league in hits allowed that year.

1919 played 133 of 140 team games.

Holdouts
1922 – Signed July 23. Played just 49 games. Did not start a game until Aug 10. Played 43 of the team’s final 46.
1930 – Sat out season when he refused to take a pay cut from Giants

Injuries
1928 – Tore a stomach muscle

Unknown to Me, but at least partially injury
1921 – played just 112 games. Missed 1st 15 games, then missed 2 weeks in late Aug early Sept, followed by 10 more games in September. I would be reluctant to give him hold out credit for those first two weeks. I am sure there are many players that had that circumstance.

So here is the four season where he should get some credits.

WS Credit
Year - Actual – Phantom - Total
1918 – 22 – 4 – 26
1919 – 33 – 3 – 36
1922 - 9 – 16 – 25
1930 – 0 – 15 – 15

Career – 314 – 38 – 352


1922 is a guess. In 1923, at age 30, he had 28 win shares. I won’t argue with you if you think 25 is too many. I won’t argue with you if you say the partial season he did play extrapolates out to 30.

1930 – His sequence starting at age 34 in 1927 - 16, 3,15, 0, 5. The three is the stomach muscle tear season.

The Giants played two RH hitters in CF in 1930, Wally Roettger OPS+ 72, and Ethan Allen OPS+ 90. The Giants missed the pennant by five games, I doubt Edd could have saved them. Cincy had Evar Swanson with an OPS+ of 80. I think a 37 Roush could have had a nice average season.

So what does this all mean?

Well if you are a peak guy Roush top 3 seasons go from 33, 33, 30 to 36, 33, 30. Which pretty much makes him a ringer for Wally Berger’s 36, 33, 31. Is there anyone voting for Berger that isn’t for Roush? I doubt it. But it might help him on someone’s spreadsheet.

So what about career? Roush has 314 Win Shares, tied with Dickey and Pee Wee for 147th all time. No adjustment for season length on that list. 314 is good company, but there are guys around there who aren’t in or won’t be.

Now if you move up the list to 352, well 352 is Duke Snider, number 87 lifetime. One ahead of Max Carey, and Lou Whitaker. I think you could say 350 is the in/out line. There is only one guy above 350 that likely won’t make it, Rusty Staub. Just below 350 are Lou Brock, Tony Perez and Dewey Evans (all 162 game schedule guys with long careers).

the SABR bioproject entry for Roush (which is ample!) confirms what Mike has said. In fact it says that Roush held out four different times, twice resolving the issue by opening day, sitting out through July of 1922 and sitting out all of 1930.

I think 1922 and 1930 are the only appropriate season for which one might consider granting him unfair labor credit. In fact the details of the problematic 1930 contract were very interesting in that the Giants not only wanted to cut his salary, but also included a provision by which Edd's salary would be further cut if the abdominal injury of 1928 recurred. Roush apparently invested his money wisely, so he could get by without and sat out the year in protest, after which he was sold or traded tothe Reds. This is something like Frank Thomas's diminshed-skills clause or like the Tigers' opt-out on Magglio's contract, except that Roush had no leverage, whereas those fellows signed in a relatively free and open market.

I don't think Edd merits any time for his father's death, that's baseball. And I don't really see what evidence there is for the 1919 ex. cred.

So Roush gets a half-year in 1922 and a full year in 1930. Starting with the latter, I think 15 is the upper bound here. Obviously the Giants were concerned enough about his abdomen that they thought he would likely miss time. Also his legs were giving out (per the bio mentioned as well as the one on baseball library). There's evidence that his bat was slipping. His OPS+ slid from the 140s and 130s to two-straight years of 124 at ages 32 and 33. Then it dropped to 96 at age 23 in 1927. The abdomen problem surely caused much further slippage to 75 OPS+ in 1928 (but in only 180 PA). After surgery, at age 36 in 1929, Edd bounced back to a 108 OPS+. After skipping 1930, he could mount only a 78 OPS+ in 400 PA. There's every indication that his 1930 could have been an OPS+ anywhere between 85 and 100. And given the age and injury profile, I think 500 AB around a 95 OPS+ seems about right. Call it 12-15 WS and err on whichever side you think is more appropriate.

Back to 1922. He posts a 131 OPS+ in just 189 PA. His two previous seasons he was 141 and 143. The next two seasons he went 148 and 134. His WS in those surrounding seasons were 33-18 and 28-20. A straightline average would get you 25 for the season. If you weighted out the PAs and stuff, you get 24 and change. I think 25 is fine.

If you adjust for schedule that's 26 for 1922 and 13-16 for 1930.

OK, so I used 26 and 15. It moves Roush ahead of Averill and in a neck-and-neck with Browning. I tilt toward peak, so it's Browning for me. Both are a shade behind Doby (with his various creits).

This begins to square the difference in findings between my keltnerized system and my current iteration-based system.

I'm glad people keep up these discussions. I'm inclined to agree with the 1922 credit and adding that in jumps him up 12 slots onto my ballot and into a dead heat with Jimmy Wynn.

John said (I guess it was somewhere else) that everybody could have held out in those days.

But they didn't (well, except for Home Run Baker). They played and they got the Win Shares and the other stats. Roush didn't. So his resume comes up a tad short of what his real ability was.

Some may say, tough luck, it was his choice. Certainly it's not like going off to war. But others may say that, well, it wasn't entirely his choice, the system was stacked against him and he did the only thing he could do. Just because others generally didn't take the same action doesn't change that.

'Course I already have Roush #2 on my ballot without any of the extra credit, so....

Just because others generally didn't take the same action doesn't change that.

As I mentioned on the 1987 thread, most of us happily prorate 1981, 1994, 1995 to full seasons. Well, Roush was simply standing up for fairer compensation and terms of employment. There's an inconsitency in not following the same logic for Roush as you would for the stirkes.

And like DL, I think it's good that this has come up. Whether right or wrong on the subject, it's excellent that we are revisiting it.

I say, who cares about whether he actually put up the stats, I'm more concerned about his true talent level over his career. If he could play at a high level I'm inclined to give him some credit for it.

As I mentioned on the 1987 thread, most of us happily prorate 1981, 1994, 1995 to full seasons. Well, Roush was simply standing up for fairer compensation and terms of employment. There's an inconsitency in not following the same logic for Roush as you would for the stirkes.

As I explained on the '87 discussion thread, I am being consistent about this, Eric.

I think that is an excellent insight, John. I have never given him that credit (of course, he is already #2 on my ballot) but this is a convincing point.

I think that is an excellent insight, John. I have never given him that credit (of course, he is already #2 on my ballot) but this is a convincing point.

What insight was that, Marc? I didn't really state anything in post #57.

BTW, even with credit for those two holdout seasons, Roush still doesn't make my ballot. He was a good one, though.

Doc C Wrote:

I don't think Edd merits any time for his father's death, that's baseball. And I don't really see what evidence there is for the 1919 ex. cred.

I'm just reminding the flock that Edd deserves WW1 shortened season credit.
Which could be a big deal to a peak voter if he is throwing Win Shares into a spreadsheet and multiplying it by something to get his rankings. A 33 * 5.3313 is good, 36 * 5.3313 is better :)

The thing about his fathers death, I just threw it in there because it is interesting, in a Jon Daly kind of way.

Well, John, you got me. What insight? I meant Doc's insight. If you adjust the season length of the strike years to give full credit to players who stood up for their economic rights, then all the more reason to adjust for players who stood up for their economic rights all by their lonesomeness.

Nice to see the discussion about Roush getting credit for his holdouts. I've been giving him this credit for about a decade and a half, and he jumped up my ballot and entered my PHOM years ago.

After seeing some high votes for Roush on the CF ranking thread, I thought it might be worth contributing my viewpoint: I think he is the second-worst post-1893 MLB position player in the Hall of Merit. Here he is in my WARP. I'm relying on BP for the run estimation and conversion factor on his Federal League seasons, using my war credit estimator to fill in the missing part of '22, and just eyeballing his credit for '30 (since he played so few games in '28 and retired after '31, I don't have enough data to feed the equation).




3-year peak: 18.1
7-year prime: 36.0
Career: 53.7
Salary: $138,345,161 (well below the in/out line, in Tim Salmon/Bobby Veach territory)

I don't see what the case here is *at all*. Could anyone explain to me how he managed to get in?

Could anyone explain to me how he managed to get in?

Checking the ballot from the deeply divided backlog year of 1997, I see that sunnyday2, Mike Webber, Andrew Siegel were his best friends. Since all three of those have participated in our ranking exercises, you might get a response to that. Roush appeared on 24 out of 51 ballots.

In the first decade of Roush's eligibility, I had him on the lower half of my ballot, but he was always behind Van Haltren for me and, I eventually decided, behind Duffy as well. He had fallen out of my top 30 by the time he was elected.

1997 had the lowest average consensus score ever, at -20.4. My own score was -24. Of the three elected that year - Dwight Evans, Fox, and Roush, Evans was the only one I had in my top 30, and I listed him 7th.

And Dan, who is raising the question, did vote in the 1997 election. Of course he didn't vote for Roush, or for Fox, either, His "elect me" votes went to McGraw, Evans, and Reggie Smith. While I was phrasing the question as "Why Roush and not Van Haltren?" perhaps Dan would prefer to phrase it as "Why Roush and not Smith?"

Someone has to be the second-worst post-1893 MLB position player in the Hall of Merit.

> that year - Dwight Evans, Fox, and Roush

Scraped the bottom of that barrel in 1997...

It takes two to make the lowest consensus score, the voters and the candidates.

As I recall, Mike Webber was Edd's champion and there is great disagreement about his fielding prowess, including great disagreement among sabrmetricians. As for Duffy and Averill. But I may be wrong.

Bill James gives Duffy and Averill A+; Roush A-. "Great disagreement" with A+ is easier than with A-.

Clay Davenport rates him up and down with two swings incredible in magnitude that do not closely match his ups and downs at bat. On the contrary. (CF 'Rate' at baseballprospectus, minimum 30 games at CF, actually 39)

--; -- 104; 108 89 94 110 112 99 98 92 92 99 88 92 111 109 -- 86 : career Rate 99

Semicolon demarcates the Federal League. Bold marks his 1917-1920, playing "every day" and posting his four best EqA. Underline marks his two short seasons, 180 and 189 PA, 43 and 39 games fielding CF.

Those two were his very short mid-career seasons. In the latter 1928 Davenport shows him suddenly playing great CF again, at the same time he plummets to career-low and miserable .236 at bat. Bill James might ridicule that without looking further.

His four best EqA 1917-1920 decline slightly, .337 .329 .321 .318. His next four best were 1916 and 1921-23, unfortunately only 418 games played. Remarkably, from his rookie FL season 1914 to the depths of 1928, 15 years, EqA is monotonic increasing, then decreasing --.278 up to .337 down to .236-- except for two blips three and five points up.

There is also a generous reading of his batting, twelve very good years 1915-1926. A true deadball batsman. The fresh clean baseballs worked in favor of others like Zack Wheat who were true sluggers.

This is what I wrote in 1997:

Higher on the All-Time list at his position than anyone except the third basemen. A star in his own time--a decent fielding CF always near the league lead in OPS. While his record is superficially similar to Van Haltren, his performances produced leaderboard offense while GVH did not, plus there were not a bunch of almost-as-good guys like Ryan, Griffin, etc. clipping at his heels. I took seriously all the challenges to him, but in the end see him as someone who hit like Reggie Smith or Bob Johnson and played solid CF. On a tight ballot that is enough for the top spot.

On reconsideration, I want to back track a bit. I am a sophisticated subjective voter and Dan is an objective voter. Usually, we end up with very similar conclusions but sometimes you need an objective system to figure out how all the plusses and minuses fit together.

The basic case for Roush is that he was a top 10, often top 5, a couple of times top2 OPS+ guy for about a deacde while playing a decent CF. That case remains true.

The problem is that every little thing you add to the equation pushes him down. His OPS+ is slugging-heavy. He was quite as durable as some others. His league was weak. There were often a few guys right behind him on leaders lists. The "decade" I spoke of was really nine years. While he was an "adequate" CF, he was on the low end of that description.

Put those things together and I can see how Dan has him a mistake. I still would have him PHoM, but I'm dropping him from about 14th to about 21st.

Yeah, basically the problems with the "prime OPS+ at position" argument in favor of Roush is the following:

1. CF from 1915-25 was NOT the same as the position we think of today. CF, the outfield corners, and 1B were all more or less at the same level of scarcity until after WWII. The gap in defensive value between CF and LF/RF in those days was around 0.2 wins/0.7 Win Shares per season. If Win Shares is giving prewar CF the same positional credit it gives to postwar CF, that's a clear error.

2. He had serious durability problems--he only had two years with over 90% of a full-timer's PA from 1916-25.

3. His best years offensively (1917 and 18) were his worst years defensively, gobbling up his peak.

4. He played the first half of his career in by far the weaker league (although I'm actually not docking him for that).

5. He played the second half of his career in a period of extremely high standard deviations, meaning that his OPS+ didn't "buy" as many pennants as one would expect.

6. His career is on the short side.

After all of these deductions, and no offsetting credits, he just falls well short of the HoM's standards.