Great point.

Lefty groundball pitchers tend to generate a lot of assist opportunities for the shortstop.

We don't know whether Cleveland's lefties were, in fact, GB pitchers. We also don't know how common platooning was against the Indians of this era. The so-called "lefty pitcher" effect is actually a righty-hitter effect; teams have tended to use more RH hitters against LHP than RHP, especially when platoons were in vouge. Speaker certainly platooned a lot, but did other managers? If they didn't, then Sewell wouldn't necessarily be seeing more opportunities than other SS.

-- MWE

Brent said pretty much everything I wanted to say about Sewell. It seems jim, that you are taking WARP word for it that Sewell was a great fielder and that is the reason why you have Sewell above Rizzuto. Also, Sewell's career rate of 109 was done in fewer games at SS than Rizzuto and without a decline phase at SS. Add in three prime fielding years for Rizzuto and another god year for MiL time if you are so inclined and I think that Rizzuto would winteh career comparison. And Sewell's supposedly superior peak is aso a contruction of how WARP sees his defense, with that great five year run that seems a little hard to believe.

Yes Sewell was better offensively. But I think that Rizzuto was better defensively and had a slightly better peak. Their careers would be about even with Rizzuto war credit. If you think that Sewell was better fine, but they are very comparable players.

Either accept all 1920's and 1930's statistics at face value, and accept the fact that there are going to be twice as many people elected from that era as from, say, the 50's and 60's, or...

Discount all statistics from the 1920's and 1930's to address this problem, so there is not such an inbalance.

I agree that this is a pertinent point, but I would suggest that there is a third option: evaluate the statistics of later periods in light of integration and other expansions of the talent pool, so that there the representation of later periods is not inappropriately smaller than it is for the 1920s and 1930s.

Sunnyday2's advocacy of fairer consideration of "the lost generation" of black ballplayers is in keeping with this approach, and it may well lead to Bus Clarkson making my ballot this year.

It should be kept in mind, though, that studies of competition level do not show that quality of competition in the 1930s esp. was significantly lower than quality of competition in the 1950s and 1960s. The fact that the 1930s will be more heavily represented in the HoM is not an artifact of segretation, although segregation may be contributing in a small way. There were _a lot_ of great players during this era, more than there were at any one time 1876-1930. WWII and competition from other sports took a lasting toll on the MLB talent pool.

My system expects that we should see about as many HoMers from the 1960s as from the 1920s, with some more from the 1930s because there was a talent spike at that time, and fewer from the 1940s and 1950s because fewer players were able to reach a high peak of excellence.

There is circumstantial evidence that some platooning was happening outside of Cleveland in the AL of the 1920s. Left-handed hitters such as Joe Judge sharing significant time with right-handed hitters at a position over a period of years is good but not conclusive evidence of a platoon arrangement.

Of course, it is not only an opposition platoon arrangements which would potentially affect Sewell's defensive ratings. With Cleveland starters, including all those lefties, throwing complete games over half the time, it certainly seems possible that the opposition manager would disproportionately choose right-handed pinch-hitters where required.

All that said, it is very difficult to get a handle on the scale of things. I truly have no idea whether Joe Sewell was a better fielder than Phil Rizzuto. There are probably octogenarian Yankee fans out there who saw them both and have opinions on the question. I would be interested.

karl, I don't think there's any question that the voters as a whole think that Beckley is more meritorious than Cash, or any other eligible first baseman for that matter. He's on twice as many ballots as Cash (20-10) and he has a huge lead in points over the next three first baseman candidates. Beckley sits 12th, while Cash, Cepeda and Taylor are back at 30, 33 and 38 respectively. The other first baseman are even further behind: Chance has two votes, Hodges one, and I don't think Easter picked up a single vote in the last election. So the question isn't whether or not Beckley is more valuable than the other first baseman. The question is whether Beckley is more valuable than players from other positions who are above him on the ballot. Picking on Dave Kingman, who isn't even eligible at this point, isn't going to do much to advance Beckley's cause.

It seems jim, that you are taking WARP word for it that Sewell was a great fielder and that is the reason why you have Sewell above Rizzuto.

No. Win Shares has Sewell as an A- fielder (like Rey Ordonez). He wasn't Vern Stephens out there. Someone want to prove to me that Rizzuto's fielding edge under Win Shares covers the batting gap?

and another god year for MiL time

Sewell was better than Rizzuto when Joe hit the majors. Why are you giving MiL credit to Rizzuto but not Sewell?

Also, Sewell's career rate of 109 was done in fewer games at SS than Rizzuto and without a decline phase at SS

Probably sound like a broken record, but Sewell WAS playing 3B near the end of his career, not DH'ing or playing 1B, and he was playing it VERY WELL, so not that much difference at the end of their careers defensively. Sewell retired at 34, while Rizzuto had his last GOOD season at 35, then had another 235 games at SS over the next 3 years, so not that big of a difference in 'decline' phase.

Look. There's a major problem with the fielding measurements in Win Shares. They don't do a good job of measuring the contributions of the individual players to the team defense. Much earlier, I did a small study of Roy Thomas, which can be found on the Bresnahan thread.

I've done a somewhat similar study using Sewell 1921-1928. He's easy to work with because he played practically every game for Cleveland over those 8 seasons.

First, I did a correlation study between team FRAA and team FWS over the 8 years. They correlate at .92. WARP and Win Shares have substantial agreement about the quality of the Cleveland defense during that time. There are small disagreements, but they both substantially agreed on when it was good and when it was not so good. (And the team was somewhat below average over this time period, -6 FRAA and -1.4 FWS below average per season.)

Then I did a correlation study between their measurements of Joe Sewell's contribution. They correlate at .63. There's agreement there, but a fair amount of disagreement, too. (Not surprising, considering the above debate.)

I then correlated the team ratings with the individual ratings (like I did in the Thomas study, which was Win Shares only). The correlation between Win Shares team ratings and Sewell's individual ratings was .68. Almost half the variance in Sewell's individual ratings comes from the variance in the team ratings (or vice-versa, but that isn't at all likely; a SS isn't that important).

I did the same for WARP. The correlation between WARP's team ratings and Sewell's individual ratings was .20. They co-vary by a small amount, an amount that represents a plausible impact of an individual on a team.

Conclusion:

Under WS, it's very difficult to be a bad fielder on a good defensive team (lots of extra credit), or a good fielder on a bad defensive team (lots of extra blame). It may be close to impossible. It's effect is as if one said, "Joe Dugan played on the 1927 Yankees; it was a great offensive team; therefore Joe Dugan HAD to be good hitter." (79 OPS+ but the big guys carried him and Koenig, team OPS+ of 137.)

I don't trust Win Shares fielding ratings AT ALL unless they are corroborated by other evidence. The team effect on the final results is HUGE. WS says Sewell was A- playing for subpar defensive team; I therefore know that he was better than that, and WARP agrees. WARP says Rizzuto was also excellent, as does contemporary opinion, well, I'm satisfied he was excellent.

Which was better? It doesn't really matter, because the difference can't make up for Joe's offense, particularly when Joe played when fielding was much more important because of the much higher BIP and error rates.

Win Shares has Sewell as an A- fielder (like Rey Ordonez).

What that means is that Sewell made a lot of his team's plays. But like every other non-PBP system, it says very little about how many opportunities he actually had to make plays, and how good he was at converting opportunities into outs. He could be rated highly by WS because he was good at making plays, or because his pitchers gave him a lot more chances to make them. WS isn't sensitive enough to know which one it is.

-- MWE

I'm now amused in the conceit in my Post 15 - sure, those are the most HOMers per league since 1900, but the real records were set in the 1890s.

For I think the first time, the HOMers per year from 1856-1900....

HOMers 1856-70, * is part-time (10+ G, fewer than half in field or 1 IP per G/35 G)
1856 (1) - Pearce
1857 (1) - Pearce
1858 (1) - Pearce
1859 (1) - Pearce
1860 (1) - Pearce, Start
1861 (2) - Pearce, Start
1862 (2) - Pearce, Start
1863 (2) - Pearce, Start
1864 (3) - Pearce, Start, GWright
1865 (2) - Pearce, Start
1866 (2) - Pearce, Start, GWright, Pike
1867 (4) - Pearce, Start, GWright, Pike
1868 (6) - Pearce, Start, GWright, Pike, Barnes, Spalding
1869 (8) - Pearce, Start, GWright, Pike, Barnes, Spalding, White, McVey
1870 (9) - Pearce, Start, GWright, Pike, Barnes, Spalding, White, McVey, Sutton

National Association HOMers, 1871-75
1871 (10) - Pearce, Start, GWright, Pike, Barnes, Spalding, White, McVey, Sutton, Anson
1872 (12) - Pearce, Start, GWright, Pike, Barnes, Spalding, White, McVey, Sutton, Anson, Hines, O'Rourke
1873 (12) - Pearce, Start, GWright, Pike, Barnes, Spalding, White, McVey, Sutton, Anson, Hines, O'Rourke
1874 (12) - Pearce, Start, GWright, Pike, Barnes, Spalding, White, McVey, Sutton, Anson, Hines, O'Rourke
1875 (12) - Pearce, Start, GWright, Pike, Barnes, Spalding, White, McVey, Sutton, Anson, Hines, O'Rourke

National League HOMers, 1876-1900
1876 (12) - Pearce*, Start, GWright, Pike, Barnes, Spalding, White, McVey, Sutton, Anson, Hines, O'Rourke
1877 (11) - Start, GWright, Pike, Barnes*, Spalding, White, McVey, Sutton, Anson, Hines, O'Rourke
1878 (12) - Start, Pike, White, McVey, Sutton, Anson, Hines, O'Rourke, Kelly, Ward, Bennett
1879 (16) - Start, GWright, Barnes, White, McVey, Sutton, Anson, Hines, O'Rourke, Kelly, Ward, Gore, Brouthers, Glasscock, Richardson, Galvin
1880 (17) - Start, White*, Sutton, Anson, Hines, O'Rourke, Kelly, Ward, Bennett, Gore, Glasscock, Richardson, Galvin, Keefe, Ewing*, Connor, Stovey
1881 (20) - Start, Barnes, White, Sutton, Anson, Hines, O'Rourke, Kelly, Ward, Bennett, Gore, Brouthers, Glasscock, Richardson, Galvin, Keefe, Ewing, Connor, Stovey, Radbourn
1882 (20) - Start, GWright, White, Sutton, Anson, Hines, O'Rourke, Kelly, Ward, Bennett, Gore, Brouthers, Glasscock, Richardson, Galvin, Keefe, Ewing, Connor, Stovey, Radbourn
1883 (17) - Start, White, Sutton, Anson, Hines, O'Rourke, Kelly, Ward, Bennett, Gore, Brouthers, Glasscock, Richardson, Galvin, Ewing, Connor, Radbourn
1884 (17.7) - Start, White, Sutton, Anson, Hines, O'Rourke, Kelly, Ward, Bennett, Gore, Brouthers, Glasscock*, Richardson, Galvin, Ewing, Connor, Radbourn, Clarkson
1885 (19.7) - Start, White, Sutton, Anson, Hines, O'Rourke, Kelly, Ward, Bennett, Gore, Brouthers, Glasscock, Richardson, Galvin**, Keefe, Ewing, Connor, Radbourn, Clarkson, Thompson
1886 (19) - Start*, White, Sutton, Anson, Hines, O'Rourke, Kelly, Ward, Bennett, Gore, Brouthers, Glasscock, Richardson, Keefe, Ewing, Connor, Radbourn, Clarkson, Thompson
1887 (19) - White, Sutton, Anson, Hines, O'Rourke, Kelly, Ward, Bennett*, Gore, Brouthers, Glasscock, Richardson, Galvin, Keefe, Ewing, Connor, Radbourn, Clarkson, Thompson
1888 (20) - White, Sutton*, Anson, Hines, O'Rourke, Kelly, Ward, Bennett, Gore*, Brouthers, Glasscock, Richardson*, Galvin, Keefe, Ewing, Connor, Radbourn, Clarkson, Thompson*, Delahanty
1889 (20) - White*, Anson, Hines, O'Rourke, Kelly, Ward, Bennett, Gore, Brouthers, Glasscock, Richardson, Galvin, Keefe, Ewing, Connor, Radbourn, Clarkson, Thompson, Delahanty*, Rusie
1890 (14) - Anson, Hines, Bennett, Glasscock, Clarkson, Thompson, Rusie, Hamilton, Nichols, Burkett, McPhee, GDavis, Young, Caruthers
1891 (25.2) - Anson, O'Rourke, Kelly**, Ward, Bennett, Gore, Glasscock, Galvin, Keefe, Ewing*, Connor, Stovey, Radbourn, Clarkson, Thompson, Delahanty, Rusie, Hamilton, Nichols, Burkett*, McPhee, GDavis, Young, Caruthers, Dahlen, Kelley*
1892 (29) - Anson, O'Rourke, Kelly, Ward, Bennett*, Gore*, Brouthers, Glasscock, Richardson*, Galvin, Keefe, Ewing, Connor, Stovey, Clarkson, Thompson, Delahanty, Rusie, Hamilton, Nichols, Burkett, McPhee, GDavis, Young, Caruthers*, Dahlen, Kelley*, Keeler*, Jennings
1893 (26) - Anson, O'Rourke, Kelly*, Ward, Bennett*, Brouthers, Glasscock, Keefe, Ewing, Connor, Stovey, Clarkson, Thompson, Delahanty, Rusie, Hamilton, Nichols, Burkett, McPhee, GDavis, Young, Caruthers*, Dahlen, Kelley, Keeler, Jennings*
1894 (21) - Anson, Ward, Brouthers, Glasscock, Ewing*, Connor, Clarkson, Thompson, Delahanty, Rusie, Hamilton, Nichols, Burkett, McPhee, GDavis, Young, Dahlen, Kelley, Keeler, Jennings, Clarke
1895 (21) - Anson, Brouthers*, Glasscock*, Ewing, Connor, Thompson, Delahanty, Rusie, Hamilton, Nichols, Burkett, McPhee, GDavis, Young, Dahlen, Kelley, Keeler, Jennings, Clarke, JCollins, Wallace
1896 (20) - Anson, Brouthers*, Ewing, Connor, Thompson, Delahanty, Hamilton, Nichols, Burkett, McPhee, GDavis, Young, Dahlen, Kelley, Keeler, Jennings, Clarke, JCollins, Wallace, Lajoie
1897 (20) - Anson, Connor*, Delahanty, Rusie, Hamilton, Nichols, Burkett, McPhee, GDavis, Young, Dahlen, Kelley, Keeler, Jennings, Clarke, JCollins, Wallace, Lajoie, Wagner*, Sheckard*
1898 (20) - Thompson*, Delahanty, Rusie, Hamilton, Nichols, Burkett, McPhee, GDavis, Young, Dahlen, Kelley, Keeler, Jennings, Clarke, JCollins, Wallace, Lajoie, Wagner, Sheckard, Flick
1899 (20) - Delahanty, Hamilton, Nichols, Burkett, McPhee, GDavis, Young, Dahlen, Kelley, Keeler, Jennings*, Clarke, JCollins, Wallace, Lajoie, Wagner, Sheckard, Flick, Crawford*, McGinnity
1899 (20) - Delahanty, Hamilton, Nichols, Burkett, McPhee, GDavis, Young, Dahlen, Kelley, Keeler, Jennings*, Clarke, JCollins, Wallace, Lajoie, Wagner, Sheckard, Flick, Crawford*, McGinnity
1900 (19) - Delahanty, Hamilton, Nichols, Burkett, GDavis, Young, Dahlen, Kelley, Keeler, Jennings, Clarke, JCollins, Wallace, Lajoie, Wagner, Sheckard, Flick, Crawford, McGinnity

AMERICAN ASSOCIATION, 1882-91
1882 (1) - McPhee
1883 (3) - McPhee, Keefe, Stovey
1884 (4) - McPhee, Keefe, Stovey, Caruthers*
1885 (3.3) - McPhee, Stovey, Caruthers, Galvin
1886 (4) - McPhee, Stovey, Caruthers, Galvin
1887 (3) - McPhee, Stovey, Caruthers
1888 (4) - McPhee, Stovey, Caruthers, Hamilton*
1889 (4) - McPhee, Stovey, Caruthers, Hamilton
1890 (0)
1891 (4.8) - Hines*, Kelly**, Brouthers, Richardson, Jennings

PLAYERS ASSOCIATION, 1890
1890 (14) - White, Gore, O'Rourke, Kelly, Ward, Keefe, Brouthers, Ewing, Connor, Radbourn, Richardson, Delahanty, Galvin, Stovey

NEGRO LEAGUE STARS
1886 (1) - Grant
1887 (1) - Grant
1888 (1) - Grant
1889 (1) - Grant
1890 (1) - Grant
1891 (1) - Grant
1892 (1) - Grant
1893 (1) - Grant
1894 (1) - Grant
1895 (2) - Grant, HR Johnson
1896 (2) - Grant, HR Johnson
1897 (2) - Grant, HR Johnson
1898 (2) - Grant, HR Johnson
1899 (2) - Grant, HR Johnson
1900 (2) - Grant, HR Johnson

jimd makes a GREAT point.

please, you Sewell top-10 voters - if Win Shares is the main reason he's on your ballot (and I think that's true in many cases), I'm very confident you're making a big mistake...

Huh? I thought jimd´s point was in favor of Sewell (WS underrating his fielding, due to him being on bad defensive teams).

And by the way, WARP and offensive value relative to other shortstops are the reasons behind my ballot placing for him.

Sorry, I meant Mike E's message...

Isn't this the "Ashburn and Hamner" problem? When there are two players and one has a very good range factor (Ashburn) and the other's is very poor (Hamner), win shares leans quite strongly toward the assumption that the difference reflects opportunities--that is, the Phillies probably had a flyball staff. WARP leans the other way, assuming that the difference reflects the player's performance in converting opportunities into outs. WS rates Hamner as an A shortstop and Ashburn as an A+ outfielder. I was just getting ready to write that WARP rates Ashburn as a superb centerfielder and Hamner as a poor shortstop (which is what I remember from when we were voting on Ashburn a few months ago). But it looks like WARP must have had a revision to include an Ashburn-Hamner adjustment, because it now rates both of them as a little better than average. Regardless of this particular example, the general principle is that WARP leans more toward assuming that statistical differences reflect differences in performance, while WS leans more toward assuming they reflect differences in opportunities.

Which assumption is correct? Mike is correct in saying that without play-by-play data it's impossible to know. We know from the studies using PBP data that the variations in opportunities can be huge and that WARP is not able to adjust for them adequately. We also know from the PBP data that WS is not adequately picking up actual variations in player performance. Jimd writes that he doesn't trust WS at all--I'll agree that WS can be very unreliable and add that IMO WARP, if anything, is even less reliable than WS. I've just seen too many cases where WARP produces fielding ratings that are wildly at variance with contemporary opinion.

How then to evaluate fielding performance for our players, who are all from the pre-PBP era? Once they become available, I think Gold Glove awards should be considered the most reliable single source of evidence. For earlier periods, I argued above that we ought to be paying more attention to non-quantitative data in the form of contemporary opinion. Unfortunately, that type of information is hard to gather, becomes sketchier the farther back one goes, and can reflect the biases of the writer. But I think we ought to be very cautious about electing someone into the HoM just because WARP (or WS) happens to like their fielding.

Win Shares (no MiL credit either way, WWII credit for Rizzuto)

Sewell 277/29-29-26-24-23-22-21-21-21-17-16-15 (12 years > 10)
Rizzuto 306/35-26-25-25-25-25-23-22-21-21-18-15-12 (13 years > 10)

OPS+

Sewell 109/147-16-16-11-11-9-9-8-2-1 (10 years > 100)
Rizzuto 93/123-4-3-2-2-0 (6 years > 100)

Fielding Runs (Rizzuto adj. for WWII)

Sewell 94/26-22-16-14-9-5-5-4-1-0 (10 years ? avg.)
Rizzuto 165/30-24-24-24-21-19-18-17-11-8-3-3 (12 years > avg.)

Does Rizzuto's fielding edge make up for Sewell's batting edge? WS says yes. (128 fWS to 89, and of course 306 career WS to 277.
That is of course if you credit Rizzuto for time lost in the military, and if so, how much.) And it's not like Sewell was Arky Vaughan with the bat anyway--second best season ever at 116.

Jim,

I don't get your point on the length of career and career rates between the two.

Rizuto played longer and played only SS. Sewell played for not as long and played 3B for about 6 years of that time. So why shouldn't Rizzuto career rating of 106 be seen as more valuable than Sewell's rathing of 109? Sewell never got to decline as a SS (and is most likely castly overrated in WARP anyway) while Rizzuto did. Rizzuto is also missing a few years in his 20's when he would have been an excellent SS. Sewell, then has years at 3B that one can't necessarily assume would make him an excellent SS during the same relative time that Rizzuto have years at SS under his average rate.

Again, in order to believe that Sewell was a better defender (and thus had a better peak and was an obviously better player) than Rizzuto you have to believe WARP's evaluation of him as a fielder. Is everyone ready to believe that? If not, then why Sewell and not Rizzuto?

P.S. I believe that neither is particularly worthy. I guess that I could have my karl moment and start to stump for Charlie Keller but I feel that is better saved for another election.

Jim,

I don't get your point on the length of career and career rates between the two.

Rizuto played longer and played only SS. Sewell played for not as long and played 3B for about 6 years of that time. So why shouldn't Rizzuto career rating of 106 be seen as more valuable than Sewell's rathing of 109? Sewell never got to decline as a SS (and is most likely castly overrated in WARP anyway) while Rizzuto did. Rizzuto is also missing a few years in his 20's when he would have been an excellent SS. Sewell, then has years at 3B that one can't necessarily assume would make him an excellent SS during the same relative time that Rizzuto have years at SS under his average rate.

Again, in order to believe that Sewell was a better defender (and thus had a better peak and was an obviously better player) than Rizzuto you have to believe WARP's evaluation of him as a fielder. Is everyone ready to believe that? If not, then why Sewell and not Rizzuto?

P.S. I believe that neither is particularly worthy. I guess that I could have my karl moment and start to stump for Charlie Keller but I feel that is better saved for another election.

Why are you giving MiL credit to Rizzuto but not Sewell?

Rizzuto had two excellent years at Kansas City in the American Association at ages 22 and 23. (Rizzuto is actually a year older than shown in bbref -- see Bill Madden's Pride of October, p. 3.) In his second year with Kansas City (1940) he was named Minor League Player of the Year. I think it's clear that he was being held back by the Yankees because they already had an excellent young shortstop in Crosetti (until his surprisingly early decline) and didn't want to trade away their top prospects.

I don't know much about Sewell's minor league career, but he came to Cleveland at age 21 from New Orleans in the Southern Association, so it seems unlikely that he was playing at the same level as Rizzuto.

Comparing Rizzuto and Sewell's fielding.
I have previously states that I do not use WS or WARP(s).
With fielding, I try to combine, f% and RFg, in comparison with contemporary league averages plus contemporary opinion (if/when possible).
So something that looks really good is when a player has an above-average f% and a much better than average RFg.
Sewell - as a SS, he was somewhat error-prone until his age 25 season, after which his f% was much better than average at position. Final total: Joe Sewell (SS): .951. League: .944. Advantage, Joe: +.007.
The reasons for his move to 3B have been discussed much, and I am willing to buy that at was not because of any fault of Joe's. He moved there, and was great at it from the start. A situation where they preferred their non-Sewell alternatives at SS than at 3B. As I don't blame this on Joe, I will also include the 3B years, and even his odd games at 2B for a final career tally of Joe: .954, League: .946. Advantage, Joe: +.008.
Looks good, but superficial without range to tell me how often he got to the ball.
As a SS: Joe: 5.37. League: 4.84. Advantage, Joe: + 0.53.
As a 3B: Joe: 2.95. League: 2.83. Advantage, Joe: + 0.12.
Total: Joe: 4.53. League: 4.16. Advantage, Joe: + 0.37.
Other numbers: as a SS, Joe had 2591 POs, 3933 As and 665 DPs against 333 Errors in 1216 games.
Conclusion: Joe was excellent with the glove.
Rizzuto:f%. Phil (only as SS): 0.968. League: 0.959. Advantage: Phil, +0.009.
RFg: Phil: 4.79. League: 4.62. Advantage: Phil, +0.017.
Other numbers: Phil had 3219 POs, 4666 As, and 1217 DPs against 263 Errors in 1647 games.

I give both men my highest possible score for defense. Well, second highest. Career (with Phil's military credit) is very similar. I'd even give Phil a half season edge.
Joe ultimately wins due to his bat. Joe was a very good hitter who never struck out. He put the bat on the ball and he got on base.
Phil was a decent hitter for a SS.

That adds up to make the difference in my ballot of Joe: 4, Phil: 37. Joe as best SS, Phil, 4th (also behind Vern Stephens and Dobie Moore).

Brent - Sewell was behind Ray Chapman - who was pretty good til that whole baseball to head, death thing. That's why anyone giving credit to Rizzuto for MiL play should also look at Sewell.
Again, I repeat - the difference between them is not the glove. They were both very good/excellent there. The difference is the bat. Sewell was very good, Rizutto was pretty good for a shortstop.

Did Cleveland have a contract with New Orleans to farm Sewell out? All bbref says is that Cleveland purchased Sewell from New Orleans, which doesn't suggest that they had any previous rights to him.

If someone has Sewell's minor league record, I'd be interested to look at it. Generally, I give minor league credit only after the first year in which a player's record clearly demonstrated that they were ready for the major leagues. In 1939 Rizzuto hit .316, scored 99 runs, stole 33 bases, and led the American Association in double plays (article on 1939 Kansas City Blues). In 1940 he hit .347 with 201 hits, 124 runs scored, and 35 stolen bases (and as I mentioned, was named Minor League Player of the Year). I have little doubt that after his 1939 season Rizzuto was ready for the majors and would have been brought up by most other organizations. Without seeing the statistics, I don't know whether Sewell may have had a similar case, but the difference in age and in league quality makes me doubtful.

<u>Would Electing Sewell Be Premature?</u>

1) Sewell's argument for the Hall of Merit rests mostly on his bat. If one takes a more liberal definition of his prime than I do (I used approximately 120 games played in a season, and not running consecutively), it runs from 1921 until 1931. During this time his BRAA (adjusted for all time) amount to 119. By way of comparison I checked the primes of four other shortstops: George David, Hughie Jennings, Bill Dahlen and (not that I'm advocating voting for him instead) Phil Rizzuto. David got 267 in 16 seasons, Jennings 124/7, Dahlen 116/20, Rizzuto 95/11. However, if one deducts highest and lowest seasons, here's the tally:
Sewell 82 (69%)
David 240 (90%)
Jennings 102 (83%)
Dahlen 105 (90.5%)
Rizzuto 77 (81%)
So much of Sewell's batting value, over one-third, is tied up in a single season, 1923, when he accumulated 44 BRAA, that it gives even a self-confessed peak voter like me pause.

2) He has accumulated a lot of fielding credit, whether your preferred stat is Win Shares or WARP, even if he does not even come close to the heights of Gold Glovers avant la lettre such as Maranville or Bancroft. However, while it's widely accepted with a shrug of the shoulders that the absence of PBP metrics for Sewell's career means we can't get an accurate assessment of his ability to convert opportunities to plays, I'm of the opinion that, "we see as through a glass darkly". The data I provided of the ratio of IF assists to OF putouts shows very clearly that the Cleveland pitching staff generated a higher ratio than the league of the former to the latter, so Sewell may simply be the beneficiary of his pitchers' ability to keep the ball on the ground. (Unless one disregards DIPS theory, and accepts that pitchers can with great accuracy pitch the ball to be hit toward the sure-handed Joe.)

I'm new at all this, but can a single-season performance with the bat carry an individual into the HoM? I'm of the 'do it twice' persuasion, before I'm convinced what one sees is not a fluke. Sewell doesn't pass that test. Without doubt, he's better than average with the bat, but he's not as good as his average BRAA per season.

Sewell. That 1923 season is definitely an outlier for him. But don't punish him (and his consistency) for having one great season with the stick.
In his 13 full(ish) seasons, he OPS+'d above 100 10 times. (46, 16, 16, 12, 10, 10, 9, 8, 2, 1) The 3 seasons below occured in his last 4 active seasons.
He still primes out somewhere in the range of a 110 OPS+. As an excellent shortstop by measures that I can trust.

>the difference between them is not the glove. They were both very good/excellent there. The difference is the bat. Sewell was very good, Rizutto was pretty good for a shortstop.

Well, the obvious difference is one was a SS and the other a SS-3B.

It doesn't matter whether it was Sewell's fault that he was at 3B, he just simply wasn't as valuable defensively. I think the 106 aDSW for Rizzuto vs. 86 DWS for Sewell tells the story.

Then the other obvious difference is the bat, yes. But if you want a bat, why not Stephens? Or Bus Clarkson?

One note per post 15, thanks to David: dismiss the listing of NL 1962-64 with 18 HOMers per year in that post, as Robin Roberts actually was toiling in Baltimore at the time...

Re: Joe Sewell and Minor League Credit.

I do not believe he is due any. He was only in the minors for a few months of his career.
According to Daguerrotypes, 1980, the only time Sewell was in the minors was 1920 when he was with New Orleans. For N.O., he hit .289 (100 H in 346 AB) with 58 runs, 19 2b, 8 3b, and 2 HR. He played SS and fielded .938 based on 147 PO, 261 ASS, and 27 Errors.
According to the Biographical Encyclopedia Baseball, Sewell's father wanted his sons to all get college degrees. Sewell signed with New Orleans after he graduated from Alabama.

Sewell played almost twice as many games as a SS than as a 3B (as opposed to Clarkson, who by my count actually played more games at 3B). And, part of the appeal of Sewell´s candidacy (at least to a fan, like me), is that he has the glove and the bat.

At this point, I´d be more interested on seeing how he was affected by the quality of his contemporaries. And I still think he´s more meritous than Moore, for example.

I have to say I think we're overlooking Stephens, whio I have considerably higher than Sewell and was a much better hitter (and hugely better than Rizutto.) yes a short career compared to Sewell, but doesn't he bear a certain resemblence to Nomar (as a SS) or Jeter, in an era that didn't produce such.

Re: Joe Sewell and Minor League Credit.

I do not believe he is due any. He was only in the minors for a few months of his career.

I agree. Those of us who give minor league credit (and I realize not everyone does) usually ignore the player's first big minor league season, figuring that's just a typical part of player development and establishes that the player is ready to move to the majors. We then reserve credit for any additional seasons played in the minor leagues after the player has clearly demonstrated the ability to play at the major league level.

During the era of independent minor league teams, it wasn't unusual for star players to remain in the minors for several years (e.g., Lefty Grove, Joe DiMaggio, Earl Averill, Gavy Cravath, and several other players we've discussed in the past). Under the National Agreement, all players in organized baseball were bound to their team (major league or minor league) by the reserve clause. These players remained with their minor league teams because the teams weren't willing to to sell or trade them to the the majors. But after farm systems became common in the 1930s, it became rarer to see star players held back in the minors. There were a few exceptions: During the 1950s many teams were slow to integrate and often held back their black players. And the Yankees, with their surplus of talent, were known for holding back top players (including Keller and Rizzuto). But after about 1960 I think we'll find that there were few cases clearly satisfying our criteria for minor league credit.

a few elections ago I decided to track my pHoM to see how if and how much my pHoM (and thus my system(there is no player in my pHoM that I curently think was a mistake though I was very close to electing some a few years and some players who are not pHoMers that I think should be more then those who aren't elected but they will probobly be elected in the next few years)) is unfair to differant eras
for this list the first numbers are players who had 10 games or 50 IP
for Negroe Leugers I used the HoF when avalible to check for playing time if not I put them first
here's the results up to but not including this election
1866
1. 1.
1867
2. 2.
1868
4. 4.
1869
6. 6.
1870
6. 6.
1871
8. 8.
1872
10. 10.
1873
10. 10.
1874
10. 10.
1875
11. 11.
1876
10. 10.
1877
10. 10.
1878
9. 10.
1879
12. 12.
1880
14. 17.
1881
17. 19.
1882
19. 20.
1883
18. 18.
1884
19. 20.
1885
20. 21.
1886
21. 21.
1887
22. 23.
1888
26. 26.
1889
27. 27.
1890
31. 31.
1891
34. 34.
1892
30. 31.
1893
29. 30.
1894
26. 27.
1895
25. 25.
1896
24. 24.
1897
23. 26.
1898
22. 22.
1899
25. 25.
1900
24. 25.
1901
27. 28.
1902
25. 25.
1903
25. 26.
1904
26. 29.
1905
27. 28.
1906
25. 30.
1907
25. 27.
1908
27. 29.
1909
25. 27.
1910
26. 28.
1911
24. 26.
1912
27. 28.
1913
27. 30.
1914
28. 32.
1915
30. 34.
1916
33. 34.
1917
31. 32.
1918
25. 28.
1919
28. 28.
1920
30. 30.
1921
29. 31.
1922
33. 35.
1923
39. 42.
1924
43. 44.
1925
45. 47.
1926
47. 47.
1927
47. 48.
1928
48. 49.
1929
47. 49.
1930
44. 49.
1931
45. 46.
1932
50. 50.
1933
48. 49.
1934
46. 47.
1935
46. 46.
1936
43. 45.
1937
45. 46.
1938
42. 42.
1939
38. 41.
1940
39. 39.
1941
38. 41.
1942
31. 33.
1943
25. 27.
1944
22. 24.
1945
22. 24.
1946
28. 30.
1947
27. 29.
1948
25. 26.
1949
22. 22.
1950
21. 23.
1951
21. 21.
1952
22. 23.
1953
24. 25.
1954
25. 26.
1955
27. 30.
1956
29. 29.
1957
27. 28.
1958
26. 26.
1959
26. 26.
1960
27. 28.
1961
27. 28.
1962
27. 28.
1963
27. 28.
1964
25. 25.
1965
24. 26.
1966
21. 21.
1967
18. 19.
1968
18. 18.
1969
16. 16.
1970
16. 16.
1971
15. 15.
1972
15. 15.
1973
13. 13.
1974
11. 11.
1975
8. 9.
1976
6. 6.
1977
2. 2.

other then 40's and to and the fifties (where players are doubly hurt because I only give credit as if he had an average season where they could have otherwise been pHoMers though I doubt it would be all of them (Gordon, Doerr, Reece, Pesky ext.) and Negroe Leauge intergration troubles keep players (Brown, Jackie, Irvin, Danridge, Easter, Doby ext.) who might have or not have kept otherwise pHoMer out of my pHoM (Please don't address those points here) )how balnced is my hallcompaired to what it "should" be
are there to many players from the 30's.
to little from the dead ball era
To little from the 50's despite those problems
or any simaler problems

The data I provided of the ratio of IF assists to OF putouts shows very clearly that the Cleveland pitching staff generated a higher ratio than the league of the former to the latter, so Sewell may simply be the beneficiary of his pitchers' ability to keep the ball on the ground. (Unless one disregards DIPS theory, and accepts that pitchers can with great accuracy pitch the ball to be hit toward the sure-handed Joe.)

You don't have to "disregard DIPS theory" to find an alternate explanation. There are two reasons why the ratio of INF assists to OF putouts is not necessarily an accurate indicator of G/F ratio:

1. Bad defensive teams tend to have a higher ratio of INF assists to OF putouts than do good defensive teams, even apart from the G/F ratio. This is because there are more runners on base against bad defensive teams, which gives infield defenders more chances for force plays and assists on relay throws from the outfield.

2. The ratio of INF assists to OF putouts tends to be higher in hitters' parks (which League Park certainly was during Sewell's years) than it is in pitchers' parks, even apart from the G/F ratio. Again, this is due to having more baserunners, and also in part because fly balls tend to become hits more often in hitters' parks because of the park dimensions and/or the carry.

-- MWE

Re 133:

Thanks for that, Mike.

I'm currently trying to get a better handle on IF defense in the 1920s AL by breaking out more teams to see how they differ. But I've only done a few teams from 1921 so far. Presumably I could make an estimate for Men on First and make some kind of adjustments, except Retrosheet doesn't list figures for pitching doubles and triples on the ML totals page.

Prelim Ballot:

1. Ralph Kiner: Tremendous hitter. Seven home run titles! Easy number 1.
2. Jose Mendez: Best pitcher on ballot. Excellent peak.
3. Dobie Moore: I'm convinced. Great hitter at SS.
4. Bill Freehan: Great defensive catcher, not bad at bat, either. He belongs in.
5. Joe Sewell: I still like him a lot, but not as much as Moore.
6. Hugh Duffy: 16.72 RC/27 in his best season. That's freakin awesome. Good glove, made his teams better. I like him a lot.
7. Ken Boyer: Brooks Robinson-lite, but with a peak.
8. Billy Pierce: Excellent peak for a pitcher of his day.
9. Rube Waddell: Awesome peak, good/great prime.
10. Thurman Munson: I might get some heat for this placement, but I think Thurm transcended the numbers. A great leader, he was the unifying force in those "Bronx Zoo" clubhouses. He was a pretty good player, too.
11. Alejandro Oms: I was missing a lot on him for a while. Nice player.
12. Charlie Keller: Poor man's Kiner. Close to Kiner with war credit, but not quite there, and definitely wouldn't have 7 HR titles.
13. Jake Beckley: Took a closer look at him, and moved him here. I wasn't giving him enough credit for the glove. Some sort of a peak, and he'd be top 5.
14. Minnie Minoso: Still don't know what to make of him.
15. Cupid Childs: Pretty good second baseman.
16. GVH
17. Pete Browning
18. Lou Brock: I don't see much to indicate he belons anywhere in the top 15. Did he have a peak? I can't see one. Wasn't great defensively, despite the speed, but for what he was, a leadoff guy, he was pretty darn good.
19. Ben Taylor
20. Norm Cash
21. Frank Howard
22. Roy White: Not much worse than Brock, I don't think
23. Chuck Klein
24. Mickey Lolich: Best new pitcher on the ballot, but not saying too much.
25. Cannonball Dick Redding
26. Addie Joss
27. Nellie Fox
28. Charley Jones
29. Dizzy Dean
30. Gavy Cravath
31. Roger Bresnahan
32. Quincy Trouppe
33. Sam Rice
34. Pie Traynor
35. Vada Pinson
36. Jimmy Wynn
37. Orlando Cepeda
38. Catfish Hunter: I didn't realize how overrated he was. Tremendously streaky, and really not all that special.
39. Bob Johnson
40. John McGraw

fra paolo post 124:

During this time his BRAA (adjusted for all time) amount to 119. By way of comparison I checked the primes of four other shortstops: George David, Hughie Jennings, Bill Dahlen and (not that I'm advocating voting for him instead) Phil Rizzuto. David got 267 in 16 seasons, Jennings 124/7, Dahlen 116/20, Rizzuto 95/11. However, if one deducts highest and lowest seasons, here's the tally:
Sewell 82 (69%)
Rizzuto 77 (81%)

Here is what I have for RCAA (NOT adjusted) for the two players, in high-to-low order, above-zero only:
Sewell. 50 24 24 16 10 .7 .5 .5 .3 .1
Rizzuto 33 10 .8 ..5 ..3

Sewell's offensive advantage is very large, over a win per year. Yes, it would come down some if Rizzuto had a good WWII season or 2.

And who had the one 'career year' with the bat? Both of them.

I don't understand how any stats can show the Scooter in the same league.

I don't understand how any stats can show the Scooter in the same league.


TomH, I would guess it is an effect of adjusting for all time. I'm only quoting figures off the BPro pages. Here's positive BRAA figures by season, as opposed to the totals:

Sewell 44 18 15 15 12 11 10 5 4 2
Rizzuto 19 19 15 15 11 11 9 3

It's a very different profile.

Personally, I don't care for adjusting for all time, but I thought people here would. I'm not particularly interested in promoting Rizzuto's candidacy, either, as I said above. But I'm no longer impressed by Sewell's peak year, which is screaming 'fluke' at me, and plays a very important role in calculations of his batting value.

OK, everybody ignored me on Page 1, so I'll try again and make a specific request. Ken Boyer served in the military during the Korean War, and therefore, might be eligible for credit (although he was in the minors at the time.) Does anybody have his minor legaue stats available, so we can try and see how much time he actually missed, and where his development was at that point?

Thanks.

Well, I did some looking, but couldn´t find his minor league stats

However, he did serve in the military until 1953, and made the majors in 1955, which was his Age-24 season. I don´t think that qualifies for credit (and, in any case, it would be too little to change my stance on him).

From the Professional Baseball Players Database:
The interesting thing is he was a pitcher his first two years in pro ball, 1949 and 1950.
in '49 he was 5-1 with a 3.42 ERA as well as a .455 BA in the North Atlantic League (Class D). In the Class D PONY League the next year, was 6-8 with a 4.39 ERA, and drove in 61 runs with a .342 BA.
Spent 1951 in the Class A Western League, gave up pitching, had 14 HR and 90 RBI. After missing two years in the service, played in AA Texas League, .319/21/116 and then was promoted to the majors in 1955.

Now I see the issue, fra. Your BRAA ##s for Rizzuto are from his F(fielding)RAA column instead. BRAA comparison (adjusted for all-time) would be:

Sewell. 44 18 15 15 12 11 10 5 4 2
Rizzuto 30 .8 ..8 .6 ..5 ..4 ..4 .1 1

Well, on second thought, it looks like Boyer might deserve some mil credit. Once he quit pitching it seems that he developed pretty rapidly. Without missing 2 years, it seems that he could have been ready for the bigs a year or two sooner.

Thanks for the Boyer info. I'd tend to agree with sunnyday - it probably wouldn't add to his peak much (which I think is pretty strong already), but it might give his career a bit of a boost.

Onto something else. I just updated an old study I did involving catcher OPS+, where I averaged the OPS+ of all the starting catchers in the league for each season. Then I added the differences for each player between their OPS+ and the seasonal averages. It's probably just a crude replication of BRAA, but I did all the work, so I'll share it. (The last time I did this, I did something to try and account for Schang & Bresnahan's non-catcher years, but I don't remember what, so I'm skipping it this time.)

The numbers in parentheses are the # of above-average and below-average years. Here's everybody who totals above 200, updated through 1992 (for Carter and Fisk)

Bill Dickey 496 (14,0)
Carlton Fisk 490 (17,1)
Gabby Hartnett 465 (14,0)
Johnny Bench 451 (13,0)
Mickey Cochrane 422 (11,0)
Ted Simmons 420 (13,0)
Ernie Lombardi 402 (13,0)
Yogi Berra 392 (11,0)
Deacon White 385 (8,0)
Wally Schang 334 (10,1)
Buck Ewing 319 (7,1)
Jack Clements 313 (9,3)
Charlie Bennett 305 (10,3)
Joe Torre 300 (8,0)
Gary Carter 278 (11,2)
Deacon McGuire 246 (9,5)
Roger Bresnahan 235 (6,0)
Roy Campanella 222 (7,3)
Mike Grady 218 (4,0)
Bill Freehan 218 (8,3,1 average)
Lance Parrish 216 (10,3)
Chief Meyers 215 (7,0)
Thurman Munson 212 (10,0)
Johnny Kling 209 (8,3)
Cal McVey 208 (4,0)
Darrell Porter 203 (10,1,1 average)

Elston Howard comes in at 127 (5,2), just below Manny Sanguillen and above Dick Dietz.

And, of course, Bill Bergen is at the bottom at -484 (0,8), more than 200 below the runner-up.

I have to say that as a Freehan supporter, the numbers don't look too good. Obviously, he has a lot of defense to add, and this DOES penalize him for bad years, but he's not really in the HoMer section of the list. (Campanella and McVey have much different arguments.) Then again, if this list was really a HoM arbiter, Ernie Lombardi would be doing a lot better in the voting.

Tom,

I don't think anyone is arguing that Rizzuto was a better hitter than Sewell, or at least that is not what I have been arguing. I have been saying that Rizzuto was the better fielder and with War and MiL credit for one year, which I think is reasonable, he career and prime looks very similar to Sewell's. I am not even really arguing for Rizzuto as he isn't in my Top 30 either. I am just saying that they are similar and I don't understand why Sewell gets so much more attention around here.

I guess if you believe WARP evaluation of Sewell's fielding you will have to have Sewell ahead. However, every other metric that I have seen, as well as anecdotal evidence and reputation, has Sewell as good but a decent bit behidn Rizzuto. If this is true they come out pretty much equal.

So support for Sewell and not for Rizzuto has to depend on three things

1. WARP's fielding #'s
2. Voters who dont' give any MiL credit and who give only small amounts of WWII credit
3. Sewell getting a boost by being teh best SS of the 1920 AL then the 1920's AL has a very weak group of SS's.

#1 seems reasonable though I think weak, #3 a little less so. #2 is something we may have a problem with overall (both with MLB and NeL players) and could be a problem with other players.

While I am not a FOJS and my tone may not quite be right is that fair? That FOJS are relying on WARP (or not paying as much attention to defense) and Sewell being the best of the 1920's?

Devin,

I would say the biggest weakness of that list is that it doesn't take into account the # of games caught in each season, which is huge in a Freehan/Lombardi comparison. I guess with that, defense, and not taking credit away for below average years (as a peak voter they are all zero for me anyway), Freehan looks much better. Take heart.

Your BRAA ##s for Rizzuto are from his F(fielding)RAA column instead.

What can I say? Ooops. Thanks, TomH, this is what the table in 124 (1) should look like:

Sewell 82 (69%)
David 240 (90%)
Jennings 102 (83%)
Dahlen 105 (90.5%)
Rizzuto 24 (40%)

It makes Rizzuto look much worse, but I only put him in because I already had figures (albeit the wrong ones) for him.

As I keep saying, I'm not interested in pushing Rizzuto's candidacy. I'm more concerned with raising objections to Sewell, whom I think is being overrated in part for the reasons listed by jschmeagol in 144, points 1 and 3, but also because he had one flukey season with the bat.

It seems a fairly ridiculous proposal to dock a player for having a season in which he was better than normal. Of course, one season of OPS+ of 140+ does not make the player a 140+ guy. But there is no reason to discount his usual 110+-ishness for having had one great season. There is also no reason to discount that one season. It was a good season. He earned it. If he had two great seasons, we would credit him with two great seasons. He did what he did. No more, no less.

fra and jschmeagol, I'm actually probably more a fan of Rizzuto's than many here; I have him in my top 30. I don't wish to demean his candidacy. I merely believe that Sewell's 130 run-ish offensive advantage is larger than Rizzuto's defense + war credit advantage. Yes, I do swallow about 40% of the WARPy fielding medicine (post 144 point 1).

I'm rather amazed that Minnie Minoso is not getting quite the support I would have thought from the knowledgeable people here. Haven't hung around in here much,
but did find his thread from early this year, so grasped the age thing and the MLEs. I dislike basing his candidacy on OPS+ because it undervalues his OBP; for a
solid decade he was consistently in the top 10 for OBP, often the top 5. His Grey Ink probably is his best argument, esp. if you think that would go up some more with
the extra years. Add in the D and listing him 15th (as many seem to be doing) is a crime. [Yes his SB% isn't good] I checked him against Indian Bob and George Burns
and in my mind Minoso has them both clearly beat (unless Johnson was a defensive whiz too). I don't know how you should do any timeline adjustments (I can see arguments
being made either way-yes the competition was worse 100 years ago but sports medicine was also in its infancy), but I think I'll prefer the more modern player, all
things being equal. Too many 19th century candidates being put ahead of guys 50 years+ younger.

Bob Johnson was a good defender also.

Mr. DiFool,

Well, Minoso's peak was godo but doesn't really stand out amongst other corner OFers and his prime, while long wasn't Slaughter/Williams/Kaline long either. To me the best place to start for most players is the uberstats and then adjust for defensive dicrepancies, team quality, other distortions, etc. from there and Minoso doesn't seperate himself from the pack in this way. We also aren't big on timelining around here as most prescribe to the pennant is a pennant is a pennant theory. I like Burns and Veach just as much and I am not really advocating for either of them for the HOM.

Also I should mention that I am a big peak guy so players like Kiner, Keller, and Pete Browning are in my top 10.

However, every other metric that I have seen, as well as anecdotal evidence and reputation, has Sewell as good but a decent bit behidn Rizzuto. If this is true they come out pretty much equal.

What are the other metrics?

Again, there needs to be a very large difference to overcome Sewell's batting gap.
No one has yet shown using Win Shares (which IS overrating Rizzuto due to his being on good teams his whole career) or any other metric that Rizzuto's fielding edge makes up for 16 points of career OPS+ (109 Sewell; 93 Rizzuto). It can't be done. (Unless you can come up with documentation that Sewell was really an OF, or that Rizzuto was much better than Ozzie Smith.)

1. WARP's fielding #'s
2. Voters who dont' give any MiL credit and who give only small amounts of WWII credit
3. Sewell getting a boost by being teh best SS of the 1920 AL then the 1920's AL has a very weak group of SS's.


Yes, I give credence to WARP's fielding numbers (2/3rds) and WS (1/3rd). They're both very good fielders. I don't really care which is better. The hitting is the difference.

I don't give MiL credit to either. I don't think that Rizzuto deserves any. His 1939 season in the link above is not particularly impressive; an empty .316 with few walks and little power. This is still the 1930's. Anyone know the park effect there? Vince DiMaggio hit 49 HRs. And all the Yanks got for him in trade was 40K and two past-prime MiL'ers (Bordagaray and Bongiovanni). OTOH, Priddy looks like he's being held back. (.333 with 24 HRs and 44 doubles.) I'd say this season got Priddy noticed, but not Rizzuto.

Yes, I give Sewell some credit for being the best ML SS in the 1920's. The NL crop is also weak (Bancroft was the best and he had a problem translating his talent to value due to his injury problems.) The NeL'ers are the competition, but they didn't all have great seasons every year.

Sewell had an All-Star quality season (or very close to it) every year from 1921-1929.
That is not common at any position (except for the inner-circle types).

But like every other non-PBP system, it says very little about how many opportunities he actually had to make plays, and how good he was at converting opportunities into outs.

Due to the design of traditional fielding stats, this may remain forever unknowable.

He could be rated highly by WS because he was good at making plays, or because his pitchers gave him a lot more chances to make them. WS isn't sensitive enough to know which one it is.

Win Shares has a L/R adjustment. I don't know how good it actually is. If it works, it should remove any L/R bias due to the pitching staff.

Win Shares also has a flyball/groundball adjustment. Again, I don't know how good it actually is. If it works, it should remove any FB/GB bias due to the pitching staff.

My guess is that the adjustments really don't work like they should. They may push things in the right direction, but don't go far enough to really cancel out the effects.

It seems a fairly ridiculous proposal to dock a player for having a season in which he was better than normal. Of course, one season of OPS+ of 140+ does not make the player a 140+ guy.

I don't even think it's unfairly ridiculous. If a 140 OPS+ in one season gives a guy 40 BRAA, which then puts his average BRAA per season up by 3-4 BRAA a year during an 8-season prime, then one is getting what I'd consider a false reading of his and performance. Better to knock that single season down a little. I don't want to favour one guy ahead of another because of a fluke season when they might otherwise be comparable.

It's all part of my strong preference for a consistently high level of performance, preferably over consecutive years, as opposed to reducing the career to a single sum.

>I don't want to favour one guy ahead of another because of a fluke season when they might otherwise be comparable.

I don't think you mean that. if two guys are comparable for years 2-n, but one guy had a huge best year and the other just a very good one, well, that's a tie-breaker. We're not concerned, or I'm not concerned with what a guy's average value was but what his value was.

Per WS (adj. to 162 games)

Best 3 seasons (no war credit in the next three columns)
Rizzuto 90
Sewell 87

Best 5 seasons
Rizzuto 137
Sewell 136

Best 7 seasons
Rizzuto 181
Sewell 181

WS gained over 15 throughout career
Rizzuto 96 (given 20 WS per season for teh war and no MiL credit)
Sewell 96

WS gained over 25 Throughout career
Rizzuto 15
Sewell 12

Career (with war credit counted)
Rizzuto 302
Sewell 291

Seems pretty even to me, even with Sewell's OPS+ advantage. As for Rizzuto always playing on good teams, it isn't like the Yankees teams he played on were of the .700 variety, in fact the late forties/early 50s' Yankees were not great teams year in and year out. And it isn't like Sewell played for historically bad teams like Bob Johnson. I cant' see this mattering more than 6 WS over their careers, maybe 10. And these two pook pretty even.

Per WARP

WARP over 4.5 throughout career
Rizzuto 34.5 (with 7, 6.5, and 6.5 in war credit, is this too much?)
Sewell 30.6

WARP over 8 throughout career
Rizzuto 4.1
Sewell 6.2

Career WARP
Rizzuto 93.5 (with war credit)
Sewell 87.4

top 3/5/7 (no war credit I believe)
Rizzuto 28.1/43.8/57.6
Sewell 28.3/45.7/60.8

Here it looks like Sewell is slightly ahead in peak and prime, though with war credit Rizzuto is ahead on career. However, I don't believe that Sewell was an all-time great SS but instead merely a very good one. Knocking his fielding numbers down a little woudl again put them roughly even. These numbers may not be up to date, however.

So it seems that WS agrees that Rizzuto and Sewell are about even players, and that Rizzuto's fielding makes up for Sewell's bat (some of this is their OBP/SLG splits which would pull them together a bit). WARP has Sewell ahead but if you dont' believe WARP's estimation of Sewell greatness in the field I think they look pretty even again, with Rizzuto ahead on career (which doesn't matter much to me).

So again I stand by my earlier statements.

Wow, jschmeagol. You really make me question why Sewell is on my ballot.

The big difference between Sewell & Rizzuto is quality of contemporaries. None of the other MLB SS's of the 1920s could hit, while in Rizzuto's time there were MLB-ers Reese, Appling, Boudreau, Stephens, Pesky, Dark and Joost who were all comparable or better hitters.

Sewell's (RCAP-RCAA) is a whopping 222, while Rizzuto's is just 77. Even though the WS/WARP numbers look comparable, its strength versus contemporaries that's the main reason to vote for Joe. Other teams weren't getting that level of value from the SS position.

That said, I'm not sure I like voting for more 1920s guys. And if you think the SS shortage in the 1920s was a fluke (or goes away when taking into account NeL-ers) then that WARP/WS analysis becomes quite compelling. Wow. So much negative pressure on almost everyone on my ballot... and we have 7 non-shoo-in slots in the next three years to induct. I'm running out of guys to vote for. :-)

I don't deduct anything for Sewell's one big season.
I LOVE seasons like Sewell's 1923, and he gets properly credited.
Plus I love Rizzuto's D and give war credit.

Still, neither will make my ballot.

jschmeagol,

Where are you getting your WARP numbers from??

The numbers you post for Sewell don't match the ones I get at all.

Using WARP2, which includes all-time competition and fielding adjustments, but doesn't adjust to 162 games (in other words, the lowest totals for Sewell, I see

Career WARP2 101.4
Top 3/5/7 (consecutive) 32.1/50.8/67.7

For Rizzuto in WARP2 (which is more favorable to him than WARP1)
Career (with war credit but no MiL) 96.0
Top 3/5/7 (con, treating 42 & 46 as con) 25.8/37.2/50.4

As WARP current WARP sees it, Rizzuto is no close to Sewell.
I agree that Win Shares tells a different story.

>I don't want to favour one guy ahead of another because of a fluke season when they might otherwise be comparable.

I don't think you mean that.

Hmm. I don't want to vote for a guy with a high average BR during his prime, when that average is boosted by one year that is 25-30 BR above every other season on his resume. By comparable I mean that their typical year is of similar value in BFW or such like.

I find myself, when assessing players, rapidly moving away from the One Big Number approach that Win Shares, WARP and even linear weighted Player Wins encapsulate, and toward eyeballing individuals' averages by season in the context of their era and position.

I do that myself. My point was, if you have two guys like this:

Win Shares

X 35-25-20-20-20-20-20-20-20
Y 25-25-20-20-20-20-20-20-20

Well, then why not acknowledge and give credit for the one big season?

Now if I were in the business of predicting future performance I'd say they were equivalent. There's a good chance both are going to be worth 20 next year. (It sounds like that's what you're saying, that they're equivalent. I mean, do you throw out the low and high like the skating judges?) But that's not what we're doing. We're assessing past value. And that 35 has more value than the 25 and all else is equal.

And I'm not even a Sewell supporter.

Yeah, for me Player X would be rank higher than player B, even if I were to deduct a little for a fluke season. Of course in the Sewell/Rizzuto comparison They both have one year that was out of step. I think it evens out there.

Chris,

I said the numbers were old and I believe they were WARP3, which adjusts for scedule length. I don't put as much weight on the WARP3 numbers in my system because of the steep timeline, but I do like their league quality adjustments and how it adjusts for schedule length.

Of course I want to restate that my aim isn't to push Phil Rizzuto to the cusp of election. In fact if that happens because of this I will probably become a Rizzuto detractor. I just want to make clear that I don't see much space between them (and BTW WS doesn't see much space between Sewell, Rizzuto, Bancroft, Fregosi, Long, and Stephens) and I am not sure that one should go in with the other out of our top 20. The SS I support for election if Dobie Moore, who at his best was clearly superior to everyone mentioned above.

But then I am wondering if there are just three reconcilable differences that we (EOJS (I guess) v. FOJS) have and it is the three mentioned above. I will restate them here with a little less snark.

1. How good of a fielder was he? Was he an all-time great (WARP) or a very good one (WS, FR, reputation)?
2. When comparing Sewell to Rizzuto how willing to give full credit for WWII and MiL performance (I will say that for a slick fielding SS an empty .316 still looks pretty good)
3. Who is more worthy; a player who was the best of his time against weak competition (which may have made hsi offense more valuable I guess) or a player who was just as good but doesn't stand out from his contemporaries because his contemporaries were very good as well?

I think the best way to answer all three puts Rizzuto and Sewell about even and that we should lower Sewell a little bit in accordance with this. Maybe he is slightly better than Sewell but by this much?

Pretty much that is everything I can say on the subject before we start repeating ourselves and talking past one another.

(It sounds like that's what you're saying, that they're equivalent. I mean, do you throw out the low and high like the skating judges?)

Well, kind of. Sewell has a big number in 1923 (44 BRAA), and lots of smaller numbers (of which the highest is 18 BRAA). If one averages the total for Sewell over his prime, one gets a higher number than the typical season that is boosted by this flukey year. I'm reluctant to give him the full value of that high average for the prime when comparing him with other players' averages, because it doesn’t really reflect his true talent level.

However, I don't want to generalize from this point and do it for everyone at each position. Sewell is a special case which has grown out of a completely new approach I took toward 'fielding positions' after noting the election of Aparicio to the Hall of Fame in 1984.

In researching I became aware that sometimes odd things happen, which I wouldn't necessarily want to give full credit for. You see, I'd dock Aparicio a couple of his gold gloves because he actually had almost no competition for them. In one season, only two other AL players had 120 games at shortstop. There wasn't a lot of choice.

Sewell is definitely a good hitter for a shortstop, but he looks even better thanks to one big year, which he never came close to repeating. I'm basically saying that I don't think the hitting gap between him and the next fellow on my SS backlog hitting analysis (Fregosi, as it happens) is as great as the sum of numbers suggests. It's all about dealing with the comparables. If one votes for Sewell at number one, does Fregosi really need to be at 15? And why is Childs, who has got way more Adjusted Batting Runs over his prime than Sewell, and looks a better fielder, especially after he finally gave in and bought a glove, being ranked below the Cleveland shortstop?

I'm a friend of Childs'. But a better fielder than JS? Not many believe that one!

Just as a curiosity, here is how we collectively rank the SS (1984 position).

5. Joe Sewell--28 ballots with 6 top 3 positions (14.4 points per ballot)
11. Dobie Moore--21 and 7 (14.0)
41. Phil Rizzuto--5 and 0 (11.6)
49T. Rabbit Maranville--3 and 1 (13.7)
51. Vern Stephens--4 and 0 (10.0)
58. Dave Bancroft--3 and 0 (10.0)
59. Luis Aparicio--3 and 0 (9.0)
66T. Artie Wilson--1 and 0 (17.0)
77T. Herman Long--1 and 0 (8.0)
81T. Jim Fregosi--1 and 0 (7.0)
85. Bus Clarkson--1 and 0 (6.0)

Support the best, forget the rest.

Sunny,

That illustrates my point pretty well. Sewell #5 and Rizzuto #41? Bancroft #58? The latter may be about right but I wonder about the former.

And we can both agree that Dobie Moore is much better than any of those players no? ;-)

j,

We agree on Moore for sure, apparently on Sewell, but maybe not on Rizzuto. Here is my top 100.

1. Moore (+10 compared to consensus)
18. Rizzuto (+18)--I'm a full WWII credit guy and trying to honor defense as well
26. Sewell (-21)--hey, I've got him #3 among eligible SSs!
36. Stephens (+15)
42. Clarkson (+43)
44. Lundy (+ infinity)
56. Bancroft (+2)
61. Tinker (+ inf)
69. Aparicio (-10)
76. Fregosi (+5)
81. A. Wilson (-15)
96. Maranville (-47)
100. Pesky (+ inf)

Not Top 100: H. Long (but 15th after Wills)

Honorable Mention: Wills

I don't generally like the really light hitters--Lundy and Bancroft hover around #50, but Aparicio and Maranville are well below reputation, HoF, HoM voting, etc. To me, Rizzuto at OPS+ 93 and 7300 adjAB (WWII credit) is more meritorious than anybody with an OPS+ in the 80s is able to be. I can't justify the gap from Rizzuto to Bancroft (OPS+ 98 in 7200 AB) other than I believe Rizzuto to be a better fielder, though I recognize not everybody believes that. (Tinker is OPS+ 95 in 6500 AB, but he rates as highly as he does in part because I like having Tinker to Evers to Chance in 3 consecutive slots on my ballot, with Tinker the best of the 3.)

Otherwise, through #43 my SSs are all hitters, too. Rizzuto is, IOW, a bit of an anomaly. Very comparable to Nellie Fox who, unlike Rizzuto, is in my PHoM.

Win Shares has a L/R adjustment. I don't know how good it actually is. If it works, it should remove any L/R bias due to the pitching staff.

Win Shares also has a flyball/groundball adjustment. Again, I don't know how good it actually is. If it works, it should remove any FB/GB bias due to the pitching staff.

Neither works.

As I pointed out earlier, what James sees as a left/right pitchers' adjustment is actually an opposing hitters' adjustment - and it only works as well as it does when you assume that there is a significant amount of platooning going on. If other teams aren't platooning that much (and Cleveland's opponents certainly weren't platooning to the extent that Speaker was), then it doesn't work well.

The flyball/groundball adjustment is based on the ratio of infield assists to outfield putouts. The primary problems here are park effects and the quality of the outfielders. If you have a bad defensive outfield, that outfield tends to have fewer putouts - but it also tends to add baserunners, which in turn leads to more infield assists (beyond just double plays, mind you; there are a fair number of 6-4 and 5-4 force plays which would not be outs with no one on base). The same effect shows up in hitters' parks - most hitters' parks are hitters' parks because they yield more fly ball hits. Using INF/OF ratio as a proxy for G/F ratio is a reasonable first-order estimate, no more than that.

When defensive WS are tracked against PBP-based systems, the relationship between DWS and zone *chances* (opportunities) is stronger than the relationship between DWS and zone *rating* (percentage of available chances converted into outs).

-- MWE

Sunny,

We aren't that far apart really. I have Rizzuto #2 and Sewell #3 but they are both about 15 spots lower or so. I also don't have Clarkson that high (I have trouble raising him because of his era though I understand that is all we can really do. He is't Easter, where I can see real HOM talent in the numbers) and Bancroft higher. Bancroft is kinda like Rizzuto Light to me.

I am nto sure of their ranings but here are my top SS's

1. Moore (around 8)
2. Rizzuto (35)
3. Sewell (40)
4. Lundy (43)
5. Bancroft (45)
6. Stephens (48)
7. Fregosi (55)
8. Wills (decent peak actually, 58)
9. Long and Tinker at about 75 or so

A lot of guys bunched between 35 and 50, ultimately I dont' think any of them should get in other than Moore. It isn't like we have a surfeit of SS's in the HOM. I do have a number of 3B (or will in the nera future) and C in my top 25, two positions where we are lacking. I don't think this is me doign quotas so much as me thinking the best players left are not SS's but C's and 3B (and OFers, as I have Keller, Duffy, Kiner, and Browning in my top 10).

Pretty much that is everything I can say on the subject before we start repeating ourselves and talking past one another.

We are probably at that point now. I started a detailed rebuttal, but decided it was all minutiae.

My summary: If I used Win Shares only, I would probably agree with you that the two were pretty close. Sewell would win on tiebreakers, due to Rizzuto being advantaged by a number of known flaws in Win Shares, but it is close (after the WWII credit is added). On the fielding debate, well, both have 4 WS Gold Gloves; Rizzuto would probably win another one. You find WARP's fielding ratings hard to believe? I have a very similar problem with Rizzuto's FWS numbers.

However, I don't use Win Shares only. WARP sees Sewell as the superior player, and it isn't close. Rizzuto's fielding edge is eliminated, leaving the hitting advantage. In addition, Sewell played in the bettter league, Rizzuto did not. Both were star players for a decade; Sewell's superior hitting (particularly given the emphasis placed on glove-men during that decade) made him much more of an impact player.

Combining it all, I have Sewell well ahead of Rizzuto, after WWII credit.

Neither works.

Thanks, Mike, for the detailed explanation as to why.

jimd wrote:

I don't give MiL credit to either. I don't think that Rizzuto deserves any. His 1939 season in the link above is not particularly impressive; an empty .316 with few walks and little power. This is still the 1930's....I'd say this season got Priddy noticed, but not Rizzuto.

I've never tried to tell other voters who they should or shouldn't give minor league credit to. (For Rizzuto, I don't give credit give credit for 1939 but I do for 1940, when he hit .347 and was named Minor League Player of the Year.) But hitting .316 in the American Association didn't get Rizzuto noticed? C'mon. The median batting average for regular major league shortstops in 1939 was .273. At least a half dozen major league teams had regular shortstops who didn't hit or field as well as Rizzuto did that year.

Anyone know the park effect there?

This was Muehlebach Park--later known as Municipal Stadium. In the 1920s Negro Leagues we saw that it played somewhere between neutral and a pitchers park. In the 1950s AL (after reconstruction) it was slightly hitter-friendly (park factors between 100-105). According to ballparks.com the outfield lines were shorter in the 1950s than in the 1920s, but it isn't clear when they were shortened. All of which is a long-winded way of saying I don't know...but there's no evidence that it was a bandbox.

Vince DiMaggio hit 49 HRs. And all the Yanks got for him in trade was 40K and two past-prime MiL'ers (Bordagaray and Bongiovanni).

This was still the depression, and $40K was a lot of money. Checking retrosheet, 40K apparently was the second largest cash transaction for 1939 and the largest received for a single player. Five years earlier, Vince's brother Joe was sold by the Seals for $25K and five players. Vince never hit 30+ home runs in the majors, but that was probably a combination of playing at pre-Kiner's Korner Forbes Field and the balatta balls that were used during the war.

OTOH, Priddy looks like he's being held back. (.333 with 24 HRs and 44 doubles.)

I agree that Priddy's season was even more impressive than Rizzuto's.

You find WARP's fielding ratings hard to believe? I have a very similar problem with Rizzuto's FWS numbers.

But don't Rizzuto's FWS numbers say pretty much the same thing as his WARP fielding rating? WARP assigns him a "Rate" of 106, which I understand to be interpreted as 6 fielding runs above average for each 100 games. FWS assigns him 7.14 FWS per 1000 innings--since the average shortstop earns about 5 FWS per 1000 innings, Rizzuto's rating is about 2 FWS (or about 6 runs) above average per 111 games. And if you think both measures are inflated, please explain why. My reading of contemporary accounts, MVP and Sporting News all-star selections, etc., all suggest that Rizzuto was regarded as an outstanding defensive shortstop. It's Sewell's WARP rating where I see some dissonance.

In addition, Sewell played in the bettter league, Rizzuto did not.

But the last half of Rizzuto's career was spent in an integrated league -- not yet as integrated as the National League, but there were some outstanding black players such as Doby and Minoso. I can't accept the 1920s AL as superior to the 1950s AL.

leaving the hitting advantage

I see Sewell's hitting advantage mitigated by 4 factors:
- Sewell spent one-third of his career at third base.
- It's likely that Rizzuto's missing WWII seasons at ages 26-28 would have raised his career average.
- Rizzuto's age 37-39 seasons (after Sewell was retired) lowered his career averages
- With military credit (plus possibly a season of minor league credit), Rizzuto should be credited with the longer career.

I'm not arguing that Rizzuto was as valuable a hitter as Sewell; just that the difference is much smaller than it appears from a simple comparison of career OPS or EQA.

Joe Sewell
.... vastly outhit the other shortstops of his day, providing huge value to his teams
BUT
.... he played in a period of unusually poor hitting shortstops
Joe Sewell
.... had a longer career than Phil Rizzuto
BUT
.... Phil missed 3 years to WWII
Joe Sewell
.... played one-third of his career at 3B, lowering his value
BUT
.... on most teams, he would have stayed at SS, since his defense there was very good

Let's be consistent; either play "what-if" and give Phil war credit and wish away Sewell's value because of the poor hitting ability of the 1920s SS, but ALSO admit that Sewell shoulda/coulda been playing SS till almost the end of his career, and not dock him
OR
stick with what actually happened; Sewell's value above the typical SS/3B of his day was enormous.

Tom, does sticking with what actually happens means no WWII credit?

Here's what actually happened (courtesy of jschmeagol, but it's essentially the same analysis that leads me to Rizzuto #18 and Sewell #26).

>Per WS (adj. to 162 games)

Best 3 seasons (no war credit in the next three columns)
Rizzuto 90
Sewell 87

Best 5 seasons
Rizzuto 137
Sewell 136

Best 7 seasons
Rizzuto 181
Sewell 181

WS gained over 15 throughout career
Rizzuto 96 (given 20 WS per season for teh war and no MiL credit)
Sewell 96

WS gained over 25 Throughout career
Rizzuto 15
Sewell 12

Career (with war credit counted)
Rizzuto 302
Sewell 291

So as not to clutter the ballot thread further, I move this observation over here:

I noted that Beckley has a double-digit lead over Brock in adj OPS+ for EVERY season of Brock's career, when listing best to worst.
Since we already know that OPS is a very good though not perfect offensive stat, it tells us quite clearly that Beckley was a better hitter.

BUT, the response is comments like these:
"Actually Beckley doesn't beat Brock until season 12 in schedule Adjusted WS though they do tie a few times along the way."

Now, that clearly tells me one thing - those WS numbers stink, at least in this case and likely many more.
I mean, Beckley clearly played a more important defensive position, and probably was better at it relative to his peers than Brcok was.

I'm baffled as to how anyone could come to any other conclusion, other than a priori assumptions that WS are the holy grail.

Preemptive strikes: the "OBP-heavy" issue is only worth a few pts per year, Brock's CS pct means he didn't really contribute all that much via running, and Beckley may not be getting enough for HIS baserunning. Give Brock a generous 5-pt OPS+ 'bonus' per year, and Beckley still mops the floor with him.

I just see it as far better to take a quality stat like OPS+, and then make all the proper adjustments.
At times it seems like the ultimate WS proponents are just ranking players by some 'magic statistic'.
No, I'm not doing that with OPS+, and I hope you can see from my somewhat eclectic ballot. It's a great starting point, though. In many cases, WS isn't even that, yet it seems to be relied on very heavily.

I'm with you there, Howie. And yes, your ballot is rather eclectic.
So I have to ask you - how does Hugh Duffy stack up to you?

Sorry to say this Howie, but what evidence do you have that I am only ranking players by WS, 'some magic statistic' (I am the one quoted above)? Aren't you only ranking thoseplayers by OPS+ in your post? And why is OPS+ a quality stat and WS not a quality stat? OPS+ doesn't seperate OBP and SLG, it doesn't take fielding and baserunning into account, and it doesn't give credit for playing a full seasons as opposed to 3/4 or 4/5 or 1/2 of one. In fact I pretty much don't use OPS+ because of its many easy to see distortions and would not mind having a post likw yours calling OPS+ a 'magic stat'. Actually, I may have done this in the ballot thread, not sure. If you want to you a rate stat, I think Eqa is VASTLY superior to OPS+ and it still has one of the problems listed above. And who is saying that I don't adjsut WS according?

The reason I post with WS a lot is that a) many times WS aren't posted, b) they give a different view of things, and c) I am most comfortable with them as a metric, more comfortable than WARP (which I will admit can be hard to understand) and or any other complex offensive, defensive, or comprehensive metric. In other words I may know a little spanish but I still speak in English, somethign akin to that.

And my point wasn't that Jake Beckley was a worse player than Lou Brock (reading the entire post that is quoted from above would tell you that I think he was actually better, even though WS disagrees), but that OPS+ may distort things a little and to look at another stat it shows pretty much the exact opposite. The Truth is probably somewhere in between. Rusty's ranking of Brock at #1 and Beckley at #3 seems perfectly acceptable to me in this light.

Well except where Brock and Beckley are in his top 50...;-)

Howie wrote:

I noted that Beckley has a double-digit lead over Brock in adj OPS+ for EVERY season of Brock's career, when listing best to worst. .... I'm baffled as to how anyone could come to any other conclusion, other than a priori assumptions that WS are the holy grail.

Is it a durability thing? I can look at Brock's record and see that he only missed about 7 games a year from 1963 to 1974, I can't do that with Beckley eyeballing it. I do see that from ages 28-30 he didn't play as frequently - of course he played forever and he was durable, but Brock might have just been playing a higer % of his team's games.

Really though once we decide who is meritous, Brock or Beckley, could we move on to the bigger question of which is blacker the pot or the kettle?

sunny, I guess it shows that Rizzuto is slightly better than Sewell..if you believe WS. Again, WS has been known to upgrade performance on great teams, and Sewell would have gotten more WS if he hadn't been moved to 3B prematurely.

How about this for simplistic analysis:

Rizzuto played for NYY temas that scored about 5 r/g over his prime years (1941-53). He averaged himself scoring 82 runs a year and driving in 53.

Sewell played for Clev teams that scored about 5 r/g over his prime years (1921-33). He averaged himself scoring 87 runs a year and driving in 80. Using fewer outs.

Was Rizzuto 16 runs a year better with the glove?

If Jake and Lou had a love child, they could call it Brockley.

>Sewell would have gotten more WS if he hadn't been moved to 3B prematurely.

Woulda/coulda/shoulda...but didn't.

I hear you on WS but I personally dislike WARP even more. All in all, they're 8 slots apart on my ballot, er, off my ballot, so....

Dobie Moore uber alles!

If Jake and Lou had a love child, they could call it Brockley.

If Jake and Lou had a love child, they could call it a miracle.

Is it a durability thing? I can look at Brock's record and see that he only missed about 7 games a year from 1963 to 1974, I can't do that with Beckley eyeballing it.

That's a small part of it, though it can't account for the whole difference.

In Brock's 11 best seasons by OWP, he played in 1705 adj. games (I adjust 1972 to 162 games).
In Beckley's 11 best seasons by OWP, he played in 1670 adj. games (all seasons adjusted to 162 games).

I'm note sure that Beckley's 11 best seasons by OWP, as listed on the Brock thread, are the same as his 11 best win-share seasons, but, if so, then

Beckley in 1670 adj. games over his best 11 seasons, topped Brock in OPS+, EQA, and OWP, and earned 212.2 adj. BWS
Brock in 1705 adj. games over his best 11 seasons, trailed Beckley in OPS+, EQA, and OWP, and earned 242.2 adj. BWS

Explanations for this discrepancy would seem to be
1) Brock, hitting leadoff, gets a higher share team PA and thus of team WS than Beckley, who bats 4-6
2) Brock, playing in a lower-scoring environment, gets more wins above avg. for his RCAA
3) Win shares doesn't show us loss shares, of which Brock would also have more than Beckley

I don't think that win shares is just "screwing up" in its evaluation of either player's offensive contribution, but, once we recognize the ways in which batting win-share totlas reflect certain contextual factors, voters might decide that batting win share totals are not, by themselves, an adequate measure of offensive merit in judging between these two players. Some may prefer the contextual measure of win shares, which brings out Brock's greater value, while others may prefer measures with more normalizations built in that show Beckley's greater relative offensive excellence, which did not translate into value as readily as Brock's did because of his role in his team's offense and his overall offensive context.

I think it's good to know that Brock honed offensive skills that were partiuclarly valuable to the context in which he played and that win shares both registers that fact and overrepresents Brock's contribution because it doesn't register his loss shares.

I am far from being ready to submit my 1985 ballot -- Brock and Munson are both borderline candidates for whom an accurate initial placement is crucial and the backlog is in need of thorough review -- but at present I prefer Beckley to Brock. The overstatement of Brock's offensive value in win shares is less serious, I think, than the understatement of Beckley's defensive value in win shares.

My analysis goes like this:

1) Brock and Beckley are similar in that they are long-career, low-peak candidates, but Beckley's career was longer.
2) Offensively, they are close in value. Beckley was, percentage wise, the better offensive player, but Brock makes up ground by his developing skills suited to his context, his durability, and the relative scarcity of lead-off talent.
3) Defensively, Beckley is more valuable. He was an above-average offensive player at his position; Brock was below average. First base 1890-1910 was probably a more important defensive position than left field 1960-1980, and they are at least equivalent, with right field definitely being a less-important fielding position than first base in Beckley's era and first base being a less-important position than left field in Brock's era.

Overall, advantage Beckley.

As this backlog year gets underway, I sense that individual bitternesses are strengthening. Some people are realizing that in the teddy-bear war, only some will get in, and they may not be yours.
It baffles me that Hugh Duffy may not make it. But Jimmy Wynn might.
Some are incensed that Joe Sewell is looking strong, but Phil Rizzuto is as far as ever.
Some cannot understand the love for Jake Beckley and others the hate.
Folks...this is war.
I live in Israel - I know of these things.

Brock's CS pct means he didn't really contribute all that much via running

Statistical analysis has many flaws and misses many difficult to quantify issues. To me, one of the most glaring examples is in regard to the truly elite baserunners. I'm talking about the top 6 or so all-time as well as the half dozen or so others who managed 250 steals in any five year period. These guys were game changers. When they were on base the game changed -- the defence changed, pitch selection changed, the defence's approach to the inning changed.

Brock's post season accomplishments, in season durability and his baserunning are the three reasons I have him ahead of Beckley (by one spot). I consider the two even defensively having read everything on this site pro and con regarding Beckley's defence.

I don't feel terribly strongly about who gets in for these elections, actually, because all the candidates have significant flaws.

That said, I still think we can make better or worse choices, and I will argue for the ones that I think are better.

As you should. As should we all.

1) Brock and Beckley are similar in that they are long-career, low-peak candidates, but Beckley's career was longer.

Beckley also "dominated" his position, unlike Brock. On a career basis, Eagle Eye was clearly the best of his era among first basemen, while one couldn't say the same thing for Brock compared to left fielders from his generation.

One decent question to ask is should we be giving credit to Brock for his having a lot of PA's by hitting leadoff? One the one hand, he doesn't decide this and it isn't fair to those who were jsut as good but hit 7th or 8th (as many speedy, low slugging types do later in their careers). On the other hand, those PA's did have real value assuming he played well in those PA's and for most of his career, he did.

I think the second reason that Chris states plays a part in that say, 20 RCAA, is more valuable in the 1960's than the 1890's.

I also don't really see the need for the loss shares since a loss means fewer WS for the tream to split up. I guess you could say that certain players add more losses to a team than others, but wouldn't those always be the players who don't get many WS per PA/G/AB/etc.? Can you at once add a lot of wins AND a lot of losses (a lot being above average per PA, of course if you get 700 PA and suck you may get 15 WS and a number of losses). I just wonder because it is a zero sum game, Wins and Losses. Wouldn't Loss shares be tautological?

I also don't really see the need for the loss shares since a loss means fewer WS for the tream to split up. I guess you could say that certain players add more losses to a team than others, but wouldn't those always be the players who don't get many WS per PA/G/AB/etc.?

Exactly, Mark. That's why I haven't jumped on the bandwagon for loss shares myself.

Can you at once add a lot of wins AND a lot of losses (a lot being above average per PA, of course if you get 700 PA and suck you may get 15 WS and a number of losses). I just wonder because it is a zero sum game, Wins and Losses. Wouldn't Loss shares be tautological?

The "if you get 700 PA and suck," or "if you get 700 PA and are only a little above average" is the case where loss shares matter a great deal.

Having loss shares available would make the win shares system more informative, in that we wouldn't have to turn to a separate metric, such as offensive winning percentage, to learn that Beckley was actually a more effective offensive player than Brock ona per/PA basis. We could see that, although he had fewer win shares, he also had fewer loss shares in comparable playing time.

I don't dabble in win shares myself, but correct me if I'm wrong - wouldn't having less win shares and less loss shares mean that the player simply did not have as much of an impact on the game?

Sunny Day wrote:

>Sewell would have gotten more WS if he hadn't been moved to 3B prematurely.

Woulda/coulda/shoulda...but didn't.

For a guy advocating Negro Leaguers from the WW2 era, woulda/coulda/shoulda is a interesting comment.

But what I really wanted to ask you Marc, I know you have been advocating for electing Negro Leaguers from the WW2/post-integration era to kind of even things out. Do you have the same feelings about the white players where WW2 clouds their candidacy? The primary two are (unfortunately) both Yankees Rizzuto and Keller.

In my opinion (supporting facts to follow), almost every system - WS, WARP - overrrates leadoff hitters. They get the advantage (as they should) of getting more at bats. But the leadoff hitters' ABs are less leveraged than those guys who hit 3-4-5, since at least once a game they hit with no one on. Studies of avg-number-of-people-on-base show they are worth about 8% less per instance than others; probably 15% less than the clean-up man's ABs.

Which I way I downgrade guys like Brock a bit when using WS/WARP.

I do agree that if you place a decent weight on post-season performance, and if SBs and speed in general are worth a little more than the typical stats formulae say, then Mr. Brock has a decent case. He had a pretty unique career.

Chris,

COuldy ou have more WS and more LS in a comparable number of PA's of G? Is that even possible? And coudln't you jsut look at WS per 500PA if you wanted to see hwo Brock and Beckley, for instance compare? Actually I am not sure why no one has done this yet. Except for the reason I am not doing it, laziness. It would probably go a ways toward figuring out this riddle.

I don't feel nearly the "bitterness" that we had back in the day.

(Signed)
Dickey Pearce

And coudln't you jsut look at WS per 500PA if you wanted to see hwo Brock and Beckley, for instance compare?

A large portion of Beckley's career is in a very high scoring context. Each game in a higher scoring era will have more PA's, yet still just 3 WS. WS/PA won't work because of that.

If Brock cloned himself was batting both 1st and 8th, then 1-Brock would have more WS & LS than 8-Brock -- with presumably the same WS_WPCT. To tell you the truth, I hadn't thought much about this before this thread. How many WS per season is batting leadoff worth?

"How many WS per season is batting leadoff worth?" I mean that from an is-it-really-worth-that-much perspective.

Well apparently each spot in the order gets 18-19 more Pa (162/9) than the spot after it (In a perfect, controlled environemnt of course where every spot in the order fnishes the same number of games). Therefore going from 1st to 8th would mean 126-133 (is this right?) PA over the course of a 162 game season. That is a lot and could have a big effect. but I guess you aso have to ask yourself, shouldn't the player who plays well at leadoff get that much more value since he is actually adding value?

jschmeagol,
I don't understand your comments relating to my post 175:
- you are not mentioned by name;
- I did not claim that ANYONE, named or unnamed, was using only WS (note the comment, "At times it seems like the ultimate WS proponents are just ranking players by some 'magic statistic'", which does not match your comment. 'At times it seems like the Mets will never lose another game', but that doesn't mean I'm wrong if they lose tonight).
- Ironically, in response you DO say of me by name, 'Aren't you only ranking those players by OPS+ in your post?' That doesn't make much sense in light of the same post you referenced, but that's a quote.
- My very post referenced all but one of your gripes about OPS+ (playing time, which I reference constantly, especially noting pluses for the Billy Williamses and minuses for the Norm Cashes whenever I speak of them).
- I have no problem with EQA.

I didn't intend to refer to you at all, if someone asked me who I grabbed the quote from, I would have no idea.

My point, again, is that OPS+ clearly shows Beckley to be a better OFFENSIVE player. Everyone understands you have to add in defense, baserunning, etc.
The danger of WS, imo, is that it can be tempting for a voter - no, I don't mean you, or anyone in particular, I wish that already was clear to you - to figure, "Hey, more WS, so better player."

But it's awfully hard to look at that dominance by Beckley in OPS, give Brock his minor edge in baserunning and playing time, and not realize that Beckley was still relatively better given that it's quite a stretch to give Brock a defensive-value edge.

rawgman, I like my OFs to be dominant offensively, and for the most part Duffy was not. I'm intrigued by the defensive history stuff written lately. It bolsters his case, but not enough to come close to my ballot.

Finally there is this quote (again, nothing personal, which is why I don't bother naming the person):
"These guys were game changers. When they were on base the game changed -- the defence changed, pitch selection changed, the defence's approach to the inning changed."

That sounds so reasonable - yet as far as I know, there is no evidence to show any significant effect. Have you ever seen any?
I do know that stolen bases have far less correlation with winning than virtually any offensive categories (even less than triples, I think, if I remember a mid-1980s Bill James Abstract correctly).
We're at the statistical point where if it's unnoticable in the win-loss category, I'm not sure I'd get too excited about it....