Baseball for the Thinking Fan

Login | Register | Feedback

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

Tuesday, September 05, 2006

1985 Ballot Discussion

1985 (September 4)—elect 3
WS W3 Rookie Name-Pos (Died)

348 82.4 1962 Lou Brock-LF
263 79.8 1966 Roy White-LF
224 80.3 1963 Mickey Lolich-P
206 77.1 1970 Thurman Munson-C (1979)
206 74.4 1965 Catfish Hunter-P (1999)
216 63.6 1966 George Scott-1B
204 60.3 1961 Tim McCarver-C
209 58.1 1964 Rico Carty-LF/DH
169 57.4 1968 Andy Mesersmith-P
157 56.3 1966 Ken Holtzman-P
159 42.1 1965 Don Kessinger-SS
129 47.5 1967 Joe Coleman-P
130 40.5 1969 Ted Sizemore-2B
121 35.7 1962 Manny Mota-LF/PH
115 37.8 1968 Dock Ellis-P
132 30.3 1963 Ed Kranepool-1B
100 39.3 1966 Darold Knowles-RP
100 37.0 1969 Jim Rooker-P
109 29.8 1969 Merv Rettenmund-RF/LF
111 28.7 1963 Vic Davalillo-CF

Players Passing Away in 1984
HoMers
Age Elected

94 1938 Stan Coveleski-P
77 1951 Joe Cronin-SS

Candidates
Age Eligible

91 1927 Al Schacht-P/Clown
91 1931 Elmer J. Smith-RF
89 1933 Babe Pinelli-3B/Umpire
89 1938 George Kelly-1B
84 1944 Waite Hoyt-P
83 1939 Glenn Wright-SS
79 1947 Spud Davis-C
78 1951 Gus Mancuso-C
77 1951 Debs Garms-LF/3B
77 1952 Joe Kuhel-1B
76 1950 Ival Goodman-RF
72 1942 Walter Alston-1B/HOF Mgr
63 1966 Jim Hegan-C
58 1968 Billy Goodman-2B/1B
50 1973 Charlie Lau-C/Coach
45 1977 Tommie Aaron-1B/LF

Upcoming Candidate
34 1987 Lynn McGlothen-P

Thanks, Dan!

John (You Can Call Me Grandma) Murphy Posted: September 05, 2006 at 05:30 PM | 280 comment(s) Login to Bookmark
  Related News:

Reader Comments and Retorts

Go to end of page

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

Page 3 of 3 pages  < 1 2 3
   201. OCF Posted: September 14, 2006 at 12:51 AM (#2176239)
Leadoff hitters - what about them? They get more PA, so they have an advantage on most counting stats. Their RBI opportunities are greatly diminished, as is their risk of GIDP. But leverage? RISP is not the only clutch situation; leading off an inning has its own importance.

How many leadoff hitters have we elected? Before about 1910, it may not have been all that unusual for a team to have its best hitter batting leadoff, but since then less so. We have elected both Max Carey and Richie Ashburn (both of whom had defense in CF to claim) - perhaps we should have had this particular discussion when Carey came up.

Brock is going to come out of this year in the mid-backlog; my crystal ball is murky about where he goes from there. Who are the significant leadoff hitters in our future? Rickey, of course, and Rose, both of whom will be easy. Then there's Raines, Molitor, and Bobby Bonds. In all three cases, they batted a lot of leadoff when they were younger, then moved down in the order - mostly to #3 - later. Their best work probably happened as leadoff hitters. (Raines eventually moved back to leadoff.) Brett Butler? Willie Wilson? In our backlog now, ignoring pre-1910: Fox batted 2nd in the order, which also gets him more PA. There are 3 or 4 voters who vote for Aparicio, and there's one who votes for Dom DiMaggio. Anyone else?
   202. Mike Webber Posted: September 14, 2006 at 01:14 AM (#2176256)
Doc & Chris,

Did either of you ever look at Sam Hairston Sr? Not that I'm craving more imaginary candidates, but I just I just read a snippet about him and wondered if either of you did a preliminary workup?
   203. Cblau Posted: September 14, 2006 at 01:15 AM (#2176259)
Brent in #120 wrote:
(Rizzuto is actually a year older than shown in bbref -- see Bill Madden's Pride of October, p. 3.)
I think Madden is confused. During Rizzuto's playing days, and at least as late as the first edition of the Baseball Encyclopedia, Rizzuto's birth year was listed as 1918, because he took a year off his age. But Madden didn't realize that had been corrected in more recent reference books.
   204. jimd Posted: September 14, 2006 at 01:43 AM (#2176301)
I've never tried to tell other voters who they should or shouldn't give minor league credit to.

Neither did I.

But hitting .316 in the American Association didn't get Rizzuto noticed?

If he'd hit those same 1939 minor league KC numbers in a neutral AL park, he'd have had a 97 OPS+. If it did get him noticed, it shouldn't have particularly. It's hardly eye-popping. (No minor league discount was applied here, so his MLE would have been considerably lower.)

But don't Rizzuto's FWS numbers say pretty much the same thing as his WARP fielding rating?

In this unusual case, Yes. As you've noted before, usually, WARP gives more credit to fielding than Win Shares (often much more). So it makes me suspicious (or at least curious) of the Win Share rating when it gives the same amount or more. I suspect that Win Shares is inflating his rating due to the quality teams he was on, but I'm not at all sure of that. It might just be depreciating Sewell due to its insistence that pitching get around 67.5% of all DWS, no matter how many or few K or BB or HR are issued or Errors made.

I can't accept the 1920s AL as superior to the 1950s AL.

I didn't say it was. I said it was superior to the 1920s NL. And that the 1950s AL was inferior to the 1950s NL. So relative to MLB averages for their respective eras, Sewell's numbers were better, and Rizzuto's numbers were weaker.

If you're timelining Sewell's value away, well, then this debate is pointless. If you're going to argue that Sewell's numbers need to be discounted to account for the Negro Leagues, then that's also true for Rizzuto because integration didn't start until mid-career and was far from finished when he was.

leaving the hitting advantage

I see Sewell's hitting advantage mitigated by 4 factors:
- Sewell spent one-third of his career at third base.


Given that 3b-men in the 1920s hit about the same as 1940s SS's, I'm not sure of the relevance of this.

- 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

No doubt. But Rizzuto only surpassed Sewell's career EQA in one season, his flukey 1950.

- With military credit (plus possibly a season of minor league credit), Rizzuto should be credited with the longer career.

Again no doubt. He has only 242 games less before WWII credit. But I don't see the WWII credit giving him a more VALUABLE career.
   205. John (You Can Call Me Grandma) Murphy Posted: September 14, 2006 at 01:50 AM (#2176312)
I didn't say it was. I said it was superior to the 1920s NL. And that the 1950s AL was inferior to the 1950s NL.

Dick Cramer's study from the Seventies agrees with the latter, but not the former. With that said, I do think that the AL had more stars than the senior circuit had.
   206. jimd Posted: September 14, 2006 at 01:55 AM (#2176319)
Wouldn't Loss shares be tautological?

For the most part, Loss Shares would add in the playing time component.

A player who played all 162 G and earned 15 WS should also earn about 25 Loss Shares.

A player who played half the season and earned 15 WS should also earn about 5 Loss Shares.

If I present to you two players, one 15-25 and the other 15-5, you get different impressions. Not tautalogical at all.
   207. jimd Posted: September 14, 2006 at 01:56 AM (#2176324)
Dick Cramer's study from the Seventies agrees with the latter, but not the former.

And contemporaneous opinion from both eras agrees with both assertions.
   208. Mike Webber Posted: September 14, 2006 at 01:57 AM (#2176325)
Chris F wrote in the ballot thread Re Redding:
I know that his reputation took a hit downward after the Hall of Fame vote which leaves me wondering why the voters take so much stock in the negative opinion about Redding but ignore the positive vote for Cooper, Day and Smith


3 pitchers, 3 stories. Day was Newark Eagle teammate of Monte Irvin, and went in when Monte held the keys for Negro Leaguers.

Smith was Buck O'Neil's roommate with the Monarchs, and went in when Buck held the keys for Negro Leaguers.

Cooper is only one that went in on the experts watch. I assume his "uncovered" stats must have pushed him forward.

So really the first two aren't like the last one, and while Mendez selection by the expert panels somewhat validates him, it puts Redding on further shaky ground (with me at least), especially considering they put in a truckload of players and excluded Redding.
   209. sunnyday2 Posted: September 14, 2006 at 01:57 AM (#2176327)
I think Bill James himself said that Rizutto's fielding rating/value is largely the product of a very unusually high number of DPs. I take him at his word on that.
   210. John (You Can Call Me Grandma) Murphy Posted: September 14, 2006 at 02:18 AM (#2176348)
And contemporaneous opinion from both eras agrees with both assertions.

Well, then it should have shown up in Cramer's study for the Twenties. Since it didn't, then the assertion is suspect, IMO.
   211. Daryn Posted: September 14, 2006 at 02:29 AM (#2176361)
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?


Who do you think you are insulting me like that. You don't even know me.

But seriously, the evidence I am relying on is watching literally thousands of games over 30 years. I have seen two hitters rip fastballs they knew were coming because of Rickey or Lofton, I have seen pitchers unravel because these guys, I have seen countless groundballs turn into singles because the middle infielder was running to second to stop them from stealing. That's evidence.

I can't quantify it and it doesn't make Coleman or Podsednik or even Lofton a Hall of Meriter, or for the first two, even close. But for a guy with a super long career, 3000+ hits, a great post season record, an MVP and a ton of runs created, it is an extra not so little factor that pushes him over the line for me.
   212. Mike Emeigh Posted: September 14, 2006 at 02:31 AM (#2176362)
I suspect that Win Shares is inflating his rating due to the quality teams he was on, but I'm not at all sure of that.


As sunnyday2 points out in #209, the larger component of Rizzuto's rating stems from the ungodly number of DPs (relative to runners on 1B) that the Yankees turned. I should point out, however, that from the start of McCarthy's tenure until the end of Stengel's, the Yankees were consistently good at turning the DP, and Rizzuto wasn't there for all of those years.

WS does tend to inflate the performance of defensive players on good defensive teams, but not by more than about a win share at most. Defensive differences at a position just isn't all that important in the WS scheme when DWS are around 17% of total team WS, and then have to be split up among eight defensive positions.

-- MWE
   213. Daryn Posted: September 14, 2006 at 02:33 AM (#2176365)
Some of you might have noticed Lou Brock doesn't actually have an MVP, but since I don't recognize Steve Garvey's existence, he has one in my little mind.
   214. John (You Can Call Me Grandma) Murphy Posted: September 14, 2006 at 02:36 AM (#2176370)
Some of you might have noticed Lou Brock doesn't actually have an MVP, but since I don't recognize Steve Garvey's existence, he has one in my little mind.

1) I thought I was going crazy there for a minute, because I didn't remember Brock's MVP.

2) You might not recognize Garvey's existence, Daryn, but it's impossible not to recognize Garvey's forearms!
   215. Brent Posted: September 14, 2006 at 05:23 AM (#2176472)
TomH wrote:

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?


While I'm a proponent of simple analysis, we also need to keep in mind its limitations. Runs plus RBIs usually correlates well with more sophisticated measures such as runs created or batting runs, but we know it underestimates batting performance for leadoff men who get few RBI opportunities. Both RCAA and BRAA agree that Sewell led Rizzuto over those years by about 8 runs per game, rather than 16.

Cblau wrote:

I think Madden is confused. During Rizzuto's playing days, and at least as late as the first edition of the Baseball Encyclopedia, Rizzuto's birth year was listed as 1918, because he took a year off his age. But Madden didn't realize that had been corrected in more recent reference books.

Thanks. Madden clearly states that the 1917 date is a year too young, but I'll take your word that he was confused.

Jimd wrote:

If he'd hit those same 1939 minor league KC numbers in a neutral AL park, he'd have had a 97 OPS+. If it did get him noticed, it shouldn't have particularly. It's hardly eye-popping. (No minor league discount was applied here, so his MLE would have been considerably lower.)

Maybe we have different understandings of the phrase, "get him noticed." I use it to mean demonstrating that a player is capable of playing at the major league level. If we shave 20 points off your calculation of Rizzuto's OPS+, at 77 it isn't eye-popping, but it's still better than 5 of the 8 AL regular shortstops. In combination with his excellent fielding and young age, I think that demonstrates he was ready to move to a major league team if he'd been given the opportunity.

If you're timelining Sewell's value away, well, then this debate is pointless. If you're going to argue that Sewell's numbers need to be discounted to account for the Negro Leagues, then that's also true for Rizzuto because integration didn't start until mid-career and was far from finished when he was.

HoMers were playing in 4 leagues in the 1920s and in 2 leagues in the 1950s (unless we elect someone like Clarkson who was playing in the minors). I think it is quite appropriate to take account of the effects of the contraction of the Negro Leagues in comparing players from these eras (just as I attempt to account for the effects of the 1890s contraction). Obviously, it doesn't have as large an effect on Rizzuto as on others who played entirely post-integration (e.g., Minoso, Fox, Boyer), but I do think Rizzuto faced higher quality competition due to integration. I don't think this adjustment should be equated with timelining.

Given that 3b-men in the 1920s hit about the same as 1940s SS's, I'm not sure of the relevance of this.

Third base had more defensive value in the 1920s and 30s than it has today, but it's never been as important as shortstop. Hitters come and hitters go; I'm not aware of any evidence that the decade-to-decade fluctuations in hitting at a position necessarily have anything to do with fielding contributions.

He [Rizzuto] has only 242 games less before WWII credit. But I don't see the WWII credit giving him a more VALUABLE career.

With win shares, Rizzuto is only 46 WS behind, so with WWII credit he easily surpasses Sewell in career WS. With WARP3, Rizzuto is 28.3 behind, so Rizzuto doesn't quite catch up but could come very close with WWII credit. (Excluding the years 1946 and 48 when Rizzuto was injured, he averaged 8.6 WARP3 per season from 1941-52; if he receives that average as war credit it would leave him within 2.5 career WARP3 of Sewell.) With WARP1, which assumes a lower "replacement level" for 1920s position players, Rizzuto remains well behind Sewell with war credit. In my own analysis, which ignores Sewell inflated (IMO) WARP fielding rating and gives Rizzuto minor league credit for 1940, he comes out well ahead of Sewell in career value.
   216. Brent Posted: September 14, 2006 at 05:50 AM (#2176484)
Sewell led Rizzuto over those years by about 8 runs per season, rather than 16.
   217. Howie Menckel Posted: September 14, 2006 at 12:44 PM (#2176558)
"But seriously, the evidence I am relying on is watching literally thousands of games over 30 years. I have seen two hitters rip fastballs they knew were coming because of Rickey or Lofton, I have seen pitchers unravel because these guys, I have seen countless groundballs turn into singles because the middle infielder was running to second to stop them from stealing. That's evidence."

Well, I have seen guys swing feebly at balls after being distracted by speedsters running to second base, I have seen potential rallies killed by the speedster leading off with a single and then caught stealing, and I have seen countless ground balls up the middle turned into outs (or even DPs) because the middle infielder was running to second to stop them from stealing. That's evidence."

Man, that may sound snarky, for which I apologize in advance.
But is it a fair point?

I mention it because I, too, grew up being told that the speed was contributing to wins. It SEEMS so right.
But if I can't find it evidence of it in relationship to games won by teams, then I am willing to question the baseball belief system I was handed.
   218. John (You Can Call Me Grandma) Murphy Posted: September 14, 2006 at 01:34 PM (#2176601)
Maybe we have different understandings of the phrase, "get him noticed." I use it to mean demonstrating that a player is capable of playing at the major league level.

That's also my definition when doling out MLE credit.
   219. Daryn Posted: September 14, 2006 at 02:04 PM (#2176649)
Well, I have seen guys swing feebly at balls after being distracted by speedsters running to second base, I have seen potential rallies killed by the speedster leading off with a single and then caught stealing, and I have seen countless ground balls up the middle turned into outs (or even DPs) because the middle infielder was running to second to stop them from stealing. That's evidence."

That's exactly right Howie - that is evidence. All I'm saying is that we have to make our own judgments on whether the unquantifiable positives outweigh the unquantifiable negatives and not just ignore them both because we don't have a "magic statistic" that weighs it for us.

Take a look at Rickey -- without any steals in his career he is still a HoMer. But do you think the steals and the threat of the steals significantly add to his value? I'm sure you do, and I'm sure you would also think that if his success rate were slightly lower (as Brock's is). I say slightly lower because the research I've seen seems to say that the SB% break even rate was lower in Brock's time than Rickey's and therefore there is, at least arguably, very little difference in their context adjusted SB%s.

We saw what Rickey did to the game when he was on base. It was, IMO, obviously a large positive for his teams. Those who saw Brock seemed to have seen the same things. For people outside of St. Louis it was mostly in those three World Series (particularly the two he was superlative in), but he left quite an impression.
   220. Daryn Posted: September 14, 2006 at 02:09 PM (#2176659)
To be clear, I'm not blindly accepting a belief system that was handed down to me. My family is British/Irish. I grew up watching baseball in a vacuum and created my own belief systems. Even to the extent that announcers affect those beliefs, I trust myself enough to objectively weigh the evidence on my own.

Was it The Temptations who said don't believe anything you hear and only half of what you see?
   221. TomH Posted: September 14, 2006 at 02:15 PM (#2176669)
The first one to make an argument sounds right, until another one comes to give a different view - Solomon, c. 1000 BCE
Let everything be established by two or three witnesses - St Paul
Seeing, they do not see; though hearing, they do not hear or understand - Isaiah, c. 600 BCE
   222. Dr. Chaleeko Posted: September 14, 2006 at 02:58 PM (#2176739)
Doc & Chris,

Did either of you ever look at Sam Hairston Sr? Not that I'm craving more imaginary candidates, but I just I just read a snippet about him and wondered if either of you did a preliminary workup?


Well, since you asked.... I just ran him through my routine. Hairston is, by the MLEs a long career catcher somewhere between Rick Ferrell and Jason Kendall, assuming average defense.

I've got some issues with the MLEs I've worked up on him, so I'm not terribly comfy with them, and I don't really know anything about him. For instance, in the one year that I have walk data for, Hairston walked 18 times in 198 AB+BB in the PCL. So he might be a good walker and have the Kendall-type OBP. But it could have been a small-sample fluke, and it's the onliest time he played in that league (in a year in which the PCL had pretty normal walk levels relative to other offensive events). Right, so what about power? Not much...except in 1950 when he had tons. He was in his power peak those years, but he never really duplicated that kind of power, and he was mostly playing in the diminunized NAL. Also, he appears to have played a LOT of games for a catcher. Like Trouppe. And being not as good a hitter as QT, I'm not sure whether Hairston could have maintained that in the majors.

My two very preliminary sketches of him show this way, all WS adj to 162

SAM HAIRSTON PRELIM MLES
1944
-1960
AGES 24
-40


                               CAREER BEST BEST  BEST  BEST 
VER
.  AVG  OBP  SLG OPS+    G  SFWS   3 WS 5 WS 10 WS 15 WS 
------------------------------------------------------------
HIGH .278 .367 .351  96  2029   277    87   132   221  273
LOW  .278 .332 .351  85  2020   242    75   114   193  234 


There is one whole-cloth season in there, 1946, where for reasons I don't know, there's no info on him other than his team. In addition, his 1944 rookie season has the same issue. I chose not to whole-cloth the 1944 year. I've done no other shaping than this. The HIGH career line looks awfully goofy, but that's what came out of the usual routines. The LOW line looks more normal and so it appears more comfortable, yet I have no idea which the oral and "final" records support.

Any help is appreciated as always.

In addition, one last thought: If Sam Hairston, then Pancho Herrara?
   223. Mark Shirk (jsch) Posted: September 14, 2006 at 03:09 PM (#2176748)
Jim,

With Loss Shares, I still don't see the point. You mention playing time but if one does a little research they will find that one player played half a season and the other a full season, so if they have the same amount of WS it would seem obvious that player A was better per PA/AB/G than Player B. I still think that Loss Shares, being the inverse of WS is isn't really measuring anything that WS isn't measuring. A loss means a lack of WS to distribute, why make that another column?

As for Rizzuto v. Sewell I dont' think that it is timelining to believe that the 1940's and 1950's AL was of the same quality as the 1920's AL. Also, while Rizzuto played for a number of Pennant Winners I don't think that any of them were of the historically great variety. WS begins to distort only for extreme teams and I don't think the Harris/Early Stengel Yankees were extreme teams, remember their best season (and I believe only team of Rizzuto's to win 100 games, though I could be wrong) is a year in which they didn't win the pennant. So I am not ready to take much off of Rizzuto for winning since his teams weren't historically great and WS doesn't measure World Series victories.

Howie,

Sorry if I sounded too snarky, but you did quote me and then it seemed as if you said that people like this use shallow analysis (I know those weren't your words, but you have to admit it could easily sound like this). It actually never dawned on me that you wouldn't know who you were quoting, so by not putting my name in there is seemed a little backhanded. I apologize.

Still, I don't see why beginning with OPS+ is in any way superior to starting with WS, and it isn't like I just said that Brock was better, in fact I said I still thought Beckley was better. I was only saying that different metrics said the exact opposite things and that it would seem reasonable to place them together. Then again you weren't actually responding to me so this paragraph may not matter.
   224. Dr. Chaleeko Posted: September 14, 2006 at 03:21 PM (#2176754)
Was it The Temptations who said don't believe anything you hear and only half of what you see?

The first one to make an argument sounds right, until another one comes to give a different view - Solomon, c. 1000 BCE
Let everything be established by two or three witnesses - St Paul
Seeing, they do not see; though hearing, they do not hear or understand - Isaiah, c. 600 BCE


Don't trust your eyes, they can deceive you. - B. Kenobi, c., uh, a long time ago.
   225. DL from MN Posted: September 14, 2006 at 03:28 PM (#2176768)
I think the overwhelming benefit to a team of a stolen base is having a runner in scoring position and removing the force-out. A pitcher may throw fastballs with a basestealer on and an infielder may change position but the success rate of batted balls isn't going to skyrocket. There are too many negative possibilities (player swings and misses at ball outside of strikezone, player lines to SS stranding the baserunner) to offset the positives.

On Win Shares and Loss Shares - Loss Shares might bring down a player with amazing talent that is frustratingly inconsistent (great range factor, high errors).
   226. andrew siegel Posted: September 14, 2006 at 03:42 PM (#2176782)
On the question of minor league credit, I continue to stand by my position that such credit is not extra credit, but earned credit. We are electing the best baseball players in this country, not the best players in so-called "major leagues." Minor league seasons have historically been ignored when evaluating players for 2 reasons: (1) we lacked the ability to translate those seasons accurately to major league performance and (2) most successful major leaguers passed through the minors at roughly the same rate, so it falls out in the wash. Well, (1) is no longer true and (2) is an argument that we need not worry about giving minor league credit to the ordinary player (who has roughly one major league quality minor league season) but in no way detracts from giving credit to the few players who had more than one such season. Rizutto, Priddy, and Keller all had 2 seasons in which they were at least solid major league players. I'm willing for record keeping purposes to let one of those seasons wash out, but the second is credit they deserve, not credit that some of us are "giving" them. (Incidentally, to be consistent, Robin Yount and Dave Winfield need to lose a little credit off their records from the beginning of their careers, because they don't have the one assumed minor league season. Effectively, they learned the game in the big leagues and got to pad their career numbers while doing so.)
   227. Mike Emeigh Posted: September 14, 2006 at 03:59 PM (#2176807)
WS begins to distort only for extreme teams and I don't think the Harris/Early Stengel Yankees were extreme teams, remember their best season (and I believe only team of Rizzuto's to win 100 games, though I could be wrong) is a year in which they didn't win the pennant.


Extreme teams represent the point at which WS distortions become obvious; that doesn't mean that WS aren't distorted at lower levels of performance. Every single-value rating system distorts "something" - this was a point that James himself made in the past when arguing against doing precisely what he did in developing Win Shares. It's important to understand what the distortions are - or, better yet, to look at multiple ways to represent player value.

-- MWE
   228. Dr. Chaleeko Posted: September 14, 2006 at 04:52 PM (#2176856)
Incidentally, to be consistent, Robin Yount and Dave Winfield need to lose a little credit off their records from the beginning of their careers

There's a very strong argument that in fact the Winfield example is spurious. Winfield was a collegiate all American who developed his game in the college ranks, probably instead of in the minors. Pete Incaviglia was a lesser player who did the same. Maybe Bob Horner too. And as SunnyD points out, there's George Sisler too. Or for that matter, we could say the same of Joe Sewell and his time at Alabama. Or maybe Tom Seaver at USC? Or Eddie Plank at Gettysburg? I dunno. What I do know is that especially in the last four decades in high-level college programs, guys are playing very high-level ball and often succeed very quickly in the pro game.
   229. Chris Fluit Posted: September 14, 2006 at 07:16 PM (#2176994)
208. Mike Webber Posted:
Chris F wrote in the ballot thread Re Redding:

I know that his reputation took a hit downward after the Hall of Fame vote which leaves me wondering why the voters take so much stock in the negative opinion about Redding but ignore the positive vote for Cooper, Day and Smith

3 pitchers, 3 stories. Day was Newark Eagle teammate of Monte Irvin, and went in when Monte held the keys for Negro Leaguers.

Smith was Buck O'Neil's roommate with the Monarchs, and went in when Buck held the keys for Negro Leaguers.

Cooper is only one that went in on the experts watch. I assume his "uncovered" stats must have pushed him forward.

So really the first two aren't like the last one, and while Mendez selection by the expert panels somewhat validates him, it puts Redding on further shaky ground (with me at least), especially considering they put in a truckload of players and excluded Redding.


Okay, Mike. I'm not necessarily saying that we have to induct the Cooper-Day-Smith trio. I do have Hilton Smith fairly close to my ballot but the other two are quite a bit back and it's doubtful that either will ever make it all the way on. What I am wondering is why we're willing to depart from the Hall of Fame concerning their positive positions but not their negative ones.

For example, the Hall of Fame had a positive position regarding Hilton Smith and inducted him. Now, it looks like you're willing to deviate from their position because of the possibility of cronyism. The Hall of Fame had a positive position regarding Leon Day. Again, it looks like you're willing to deviate from their position because of the possibility of cronyism. Fine.

The example that comes the closest is that of Andy Cooper. You posit that the recent experts inducted Cooper because of "uncovered stats." Before the most recent Hall of Fame election, Dick Redding was doing fairly well in the Hall of Merit based on the evidence we had available. Meanwhile, Andy Cooper was not faring well based on the evidence we have available for him. Then the Hall of Fame results were announced. Dick Redding was not inducted into the Hall of Fame and one of the main factors was that the Hall of Fame had different numbers and evidence. Because of that, a number of our voters downgraded Redding with the result that he no longer even cracks our top ten. But the same group that had different evidence for Redding also had different evidence for Cooper. And their evidence led them to a higher opinion of Cooper, high enough to actually induct him. So if the Hall of Fame evidence is so critical to Redding's case, as several voters have mentioned, then why hasn't Cooper seen a dramatic rise to offset Redding's decline? It seems like we're willing to listen to the Hall of Fame experts when they have a lower opinion than we do, but not a higher one.

They had a positive position regarding Cooper, Day and Smith. And we're willing to deviate from their positive position with a (so far) negative one of our own.
They had a negative position regarding Redding. If we're willing to deviate from their positive position with a negative one, we should also be willing to deviate from their negative position with a positive one. Either their opinion counts- but in favor of some and in disfavor of others. Or their opinion doesn't count. But right now, we're being hypocritical if we're downgrading Redding without upgrading the others. The solution I would suggest wouldn't be to upgrade the others. We've deviated from the Hall of Fame when it comes to infielders, inducting Beckwith and potentially Moore instead of Dandridge and Johnson. I think we should go ahead and deviate from them regarding pitchers as well. Our evidence tells us that Redding was a great pitcher. Better than Cooper. Better than Day. Let's listen to our evidence and support this outstanding candidate. But if we're going to ignore our evidence and take another group's opinion as higher than our own, then we should at least be consistent about it, taking their positive opinions as well as their negative ones.
   230. Chris Fluit Posted: September 14, 2006 at 08:06 PM (#2177065)
211. Daryn Posted:
But seriously, the evidence I am relying on is watching literally thousands of games over 30 years. I have seen two hitters rip fastballs they knew were coming because of Rickey or Lofton, I have seen pitchers unravel because these guys, I have seen countless groundballs turn into singles because the middle infielder was running to second to stop them from stealing. That's evidence.

I can't quantify it and it doesn't make Coleman or Podsednik or even Lofton a Hall of Meriter, or for the first two, even close. But for a guy with a super long career, 3000+ hits, a great post season record, an MVP and a ton of runs created, it is an extra not so little factor that pushes him over the line for me.


217. Howie Menckel Posted:
Well, I have seen guys swing feebly at balls after being distracted by speedsters running to second base, I have seen potential rallies killed by the speedster leading off with a single and then caught stealing, and I have seen countless ground balls up the middle turned into outs (or even DPs) because the middle infielder was running to second to stop them from stealing. That's evidence.

Man, that may sound snarky, for which I apologize in advance.
But is it a fair point?

I mention it because I, too, grew up being told that the speed was contributing to wins. It SEEMS so right.
But if I can't find it evidence of it in relationship to games won by teams, then I am willing to question the baseball belief system I was handed.


219. Daryn Posted:
That's exactly right Howie - that is evidence. All I'm saying is that we have to make our own judgments on whether the unquantifiable positives outweigh the unquantifiable negatives and not just ignore them both because we don't have a "magic statistic" that weighs it for us.

Take a look at Rickey -- without any steals in his career he is still a HoMer. But do you think the steals and the threat of the steals significantly add to his value? I'm sure you do, and I'm sure you would also think that if his success rate were slightly lower (as Brock's is). I say slightly lower because the research I've seen seems to say that the SB% break even rate was lower in Brock's time than Rickey's and therefore there is, at least arguably, very little difference in their context adjusted SB%s.

We saw what Rickey did to the game when he was on base. It was, IMO, obviously a large positive for his teams. Those who saw Brock seemed to have seen the same things. For people outside of St. Louis it was mostly in those three World Series (particularly the two he was superlative in), but he left quite an impression.


I’m going to side with Daryn on this one. I’ve been to dozens of games but one of my most vivid memories comes from a game in Milwaukee. The Atlanta Braves were in town playing the Brewers and they had this hot, young rookie named Rafael Furcal. We watched the game from field level, sitting in the second row between the first baseman and the right fielder. And when the game was done, my friends and I asked each other who had been the most impressive player we had seen that day. Every one of us answered, “Furcal.” He had made some pretty decent plays in the field. But more importantly, he had wrecked havoc on the base paths. When he was at first, he changed the game. He changed the way the defense played. He changed the way the pitcher pitched. The Brewers were doing everything they could to stop him from getting to second base but they couldn’t do it. He stole second. He even stole third. I don’t remember the exact score, but I do remember that the Braves won and on that day, their best player was a scrappy little shortstop who stole bases.

Does that make Rafael Furcal a Hall of Famer (or a HoMer)? No. It was just one game. But it demonstrated clearly to me that a speed guy can dominate a game as much as a power hitter. Sure, the Braves pitching still had to come through. Furcal’s runs wouldn’t have helped much if the Braves were giving up three-run homers. And yup, Furcal still needed someone behind him in the order to get a hit and drive him in.

But those exact same criticisms can be leveled at players who excel in other parts of the game as well. I’ve seen Shawn Green win a game with a grand slam. But he couldn’t have won the game in that fashion if other players hadn’t gotten on ahead of him. He wouldn’t have been in a position to win the game if Toronto’s bullpen had given up just a few more runs than what they did. Nobody wins a game entirely alone. Not a great pitcher. Not a power hitter. Not a speedster. But each kind of player has the ability to have that kind of impact on a game.

Howie, you brought up some of the negative side effects of the stolen base. Yup, I’ve seen rallies die because the lead-off hitter got thrown out trying to steal second. It happens. I’ve seen all of those negative things that you mentioned. And I’ve seen more. I’ve seen rallies die because power hitters struck out swinging for the fences. I’ve seen slow sluggers stopped at third rather than risk being thrown out at the plate and then being stranded there. There are negatives to the power game as well as to the speed game.

Yet I think some of your negative examples can be countered. Sure, some groundballs up the middle are fielded by second basemen running in to cover the bag. I’ve seen that happen. But as Daryn mentioned, we’ve also seen groundballs sneak through the infield going right through the spot where the shortstop had been before he ran over to second base. They both happen. There’s a bit of luck involved there. But why are you so swayed by the negative? Why are you willing to discount or disregard the positive?

I’m also baffled by the complaint about the effect on the batter. I remember this came up in the Maury Wills thread. Earlier, other members had been complimenting the Dodgers’ number two hitter for being so patient. But when the subject of Maury Wills came up, suddenly people were complaining that the Dodgers’ number two hitter was underrated and would have been even better if he didn’t have to take so many pitches waiting for Wills to steal. Well, which is it? Is it good to be a patient hitter? To take extra pitches? To build up his pitch count? Or is it not good? ‘Cause it seems to me like some people have the attitude that it’s good unless you have a guy who likes to steal bases on ahead of you and then it’s bad. And I’m just not sure that makes sense.

The stolen base is a valuable weapon. It can be overrated, as has often been the case. And it can be underrated, which I think is the case now.

The big complaint about the stolen base is that it doesn’t correlate to winning. Well, looking at this year’s statistics- the Minnesota Twins and the Los Angeles Dodgers are tied for 28th in home runs, yet each is currently in a playoff position. The Detroit Tigers are 27th in On Base Percentage, yet they currently lead their division. St. Louis’ pitchers have the 17th best ERA yet they lead their division. The Phillies have the 23rd best ERA yet they’re in the heart of the race for the Wild Card.

I looked at the list of home runs and stolen bases. Four of the top ten teams in stolen bases are currently in a playoff position. Only three of the top ten teams in home runs are currently in a playoff position. And (surprise, surprise) the teams with the best records in baseball are the teams which make the top ten in both categories: the Yankees and the Mets. Stolen bases do help teams score runs. And they do help teams win.
   231. Chris Fluit Posted: September 14, 2006 at 08:07 PM (#2177066)
211. Daryn Posted:
But seriously, the evidence I am relying on is watching literally thousands of games over 30 years. I have seen two hitters rip fastballs they knew were coming because of Rickey or Lofton, I have seen pitchers unravel because these guys, I have seen countless groundballs turn into singles because the middle infielder was running to second to stop them from stealing. That's evidence.

I can't quantify it and it doesn't make Coleman or Podsednik or even Lofton a Hall of Meriter, or for the first two, even close. But for a guy with a super long career, 3000+ hits, a great post season record, an MVP and a ton of runs created, it is an extra not so little factor that pushes him over the line for me.


217. Howie Menckel Posted:
Well, I have seen guys swing feebly at balls after being distracted by speedsters running to second base, I have seen potential rallies killed by the speedster leading off with a single and then caught stealing, and I have seen countless ground balls up the middle turned into outs (or even DPs) because the middle infielder was running to second to stop them from stealing. That's evidence.

Man, that may sound snarky, for which I apologize in advance.
But is it a fair point?

I mention it because I, too, grew up being told that the speed was contributing to wins. It SEEMS so right.
But if I can't find it evidence of it in relationship to games won by teams, then I am willing to question the baseball belief system I was handed.


219. Daryn Posted:
That's exactly right Howie - that is evidence. All I'm saying is that we have to make our own judgments on whether the unquantifiable positives outweigh the unquantifiable negatives and not just ignore them both because we don't have a "magic statistic" that weighs it for us.

Take a look at Rickey -- without any steals in his career he is still a HoMer. But do you think the steals and the threat of the steals significantly add to his value? I'm sure you do, and I'm sure you would also think that if his success rate were slightly lower (as Brock's is). I say slightly lower because the research I've seen seems to say that the SB% break even rate was lower in Brock's time than Rickey's and therefore there is, at least arguably, very little difference in their context adjusted SB%s.

We saw what Rickey did to the game when he was on base. It was, IMO, obviously a large positive for his teams. Those who saw Brock seemed to have seen the same things. For people outside of St. Louis it was mostly in those three World Series (particularly the two he was superlative in), but he left quite an impression.


I’m going to side with Daryn on this one. I’ve been to dozens of games but one of my most vivid memories comes from a game in Milwaukee. The Atlanta Braves were in town playing the Brewers and they had this hot, young rookie named Rafael Furcal. We watched the game from field level, sitting in the second row between the first baseman and the right fielder. And when the game was done, my friends and I asked each other who had been the most impressive player we had seen that day. Every one of us answered, “Furcal.” He had made some pretty decent plays in the field. But more importantly, he had wrecked havoc on the base paths. When he was at first, he changed the game. He changed the way the defense played. He changed the way the pitcher pitched. The Brewers were doing everything they could to stop him from getting to second base but they couldn’t do it. He stole second. He even stole third. I don’t remember the exact score, but I do remember that the Braves won and on that day, their best player was a scrappy little shortstop who stole bases.

Does that make Rafael Furcal a Hall of Famer (or a HoMer)? No. It was just one game. But it demonstrated clearly to me that a speed guy can dominate a game as much as a power hitter. Sure, the Braves pitching still had to come through. Furcal’s runs wouldn’t have helped much if the Braves were giving up three-run homers. And yup, Furcal still needed someone behind him in the order to get a hit and drive him in.

But those exact same criticisms can be leveled at players who excel in other parts of the game as well. I’ve seen Shawn Green win a game with a grand slam. But he couldn’t have won the game in that fashion if other players hadn’t gotten on ahead of him. He wouldn’t have been in a position to win the game if Toronto’s bullpen had given up just a few more runs than what they did. Nobody wins a game entirely alone. Not a great pitcher. Not a power hitter. Not a speedster. But each kind of player has the ability to have that kind of impact on a game.

Howie, you brought up some of the negative side effects of the stolen base. Yup, I’ve seen rallies die because the lead-off hitter got thrown out trying to steal second. It happens. I’ve seen all of those negative things that you mentioned. And I’ve seen more. I’ve seen rallies die because power hitters struck out swinging for the fences. I’ve seen slow sluggers stopped at third rather than risk being thrown out at the plate and then being stranded there. There are negatives to the power game as well as to the speed game.

Yet I think some of your negative examples can be countered. Sure, some groundballs up the middle are fielded by second basemen running in to cover the bag. I’ve seen that happen. But as Daryn mentioned, we’ve also seen groundballs sneak through the infield going right through the spot where the shortstop had been before he ran over to second base. They both happen. There’s a bit of luck involved there. But why are you so swayed by the negative? Why are you willing to discount or disregard the positive?

I’m also baffled by the complaint about the effect on the batter. I remember this came up in the Maury Wills thread. Earlier, other members had been complimenting the Dodgers’ number two hitter for being so patient. But when the subject of Maury Wills came up, suddenly people were complaining that the Dodgers’ number two hitter was underrated and would have been even better if he didn’t have to take so many pitches waiting for Wills to steal. Well, which is it? Is it good to be a patient hitter? To take extra pitches? To build up his pitch count? Or is it not good? ‘Cause it seems to me like some people have the attitude that it’s good unless you have a guy who likes to steal bases on ahead of you and then it’s bad. And I’m just not sure that makes sense.

The stolen base is a valuable weapon. It can be overrated, as has often been the case. And it can be underrated, which I think is the case now.

The big complaint about the stolen base is that it doesn’t correlate to winning. Well, looking at this year’s statistics- the Minnesota Twins and the Los Angeles Dodgers are tied for 28th in home runs, yet each is currently in a playoff position. The Detroit Tigers are 27th in On Base Percentage, yet they currently lead their division. St. Louis’ pitchers have the 17th best ERA yet they lead their division. The Phillies have the 23rd best ERA yet they’re in the heart of the race for the Wild Card.

I looked at the list of home runs and stolen bases. Four of the top ten teams in stolen bases are currently in a playoff position. Only three of the top ten teams in home runs are currently in a playoff position. And (surprise, surprise) the teams with the best records in baseball are the teams which make the top ten in both categories: the Yankees and the Mets. Stolen bases do help teams score runs. And they do help teams win. There are many ways to win games, and stealing bases happens to be one of them.
   232. Chris Fluit Posted: September 14, 2006 at 08:08 PM (#2177070)
Oops. Sorry about the double post.
   233. rawagman Posted: September 14, 2006 at 08:11 PM (#2177074)
Speaking of Ben Taylor....
   234. Dizzypaco Posted: September 14, 2006 at 08:24 PM (#2177094)
The big complaint about the stolen base is that it doesn’t correlate to winning. Well, looking at this year’s statistics- the Minnesota Twins and the Los Angeles Dodgers are tied for 28th in home runs, yet each is currently in a playoff position. The Detroit Tigers are 27th in On Base Percentage, yet they currently lead their division. St. Louis’ pitchers have the 17th best ERA yet they lead their division. The Phillies have the 23rd best ERA yet they’re in the heart of the race for the Wild Card.

Uh huh. The top five in stolen bases in the AL are the Angels, Yankees, Devil Rays, Orioles, and Mariners. The top six in the NL are Mets, Reds, Dodgers, Padres, Nationals, and Cubs.

In other words, there is no correlation whatsoever between stealing bases and winning. In fact, it might be a negative correlation - the more stolen bases, the worse your team - but I haven't checked.

No statistic is perfectly correlated with winning, but home runs, on base percentage, and ERA usually do pretty well, if you look at a period of several years. Stolen bases do not.
   235. Mike Webber Posted: September 14, 2006 at 09:00 PM (#2177139)
SAM HAIRSTON PRELIM MLES
1944-1960
AGES 24-40


Thanks Doc
   236. sunnyday2 Posted: September 14, 2006 at 10:14 PM (#2177197)
Have we (humankind) actually seen the new HoF data on Redding and Cooper? I forget.
   237. DL from MN Posted: September 14, 2006 at 10:27 PM (#2177208)
Your team will score more runs in a situation with a man on 2nd v. a man on 1st with the same number of outs. Top basestealers are also usually top baserunners, which leads to more runs. That's how basestealers contribute, the rest of the stuff is just noise.

I think the HoF raw data has been more important for Redding than the HoF opinion. People were voting for Redding because he was a "career" value pitcher. The HoF data calls that assumption into question. I actually upgraded Hilton Smith based on his HoF data. The HoF data for Cooper makes him look like an average pitcher with shiny W-L records. I don't think people here are swayed by W-L records if the underlying stats don't support the pitcher as the primary reason for the wins. Cooper won because he played for great teams, not vice versa. I think the HoF committee valued "W-L%" and "lefthanded" much more than this electorate.
   238. Dr. Chaleeko Posted: September 14, 2006 at 10:27 PM (#2177209)
Have we (humankind) actually seen the new HoF data on Redding and Cooper? I forget.

Yes. It's in the .pdf file that the coop posted before the election. It does not give any accounting of the league's ERA or RA, sadly and no other benchmarks for us to work with, so we're left trying to figure it on our own. Which we've attempted to do recently a little bit in one of the NgL threads, but it's very much an estimate at this juncture.
   239. Chris Fluit Posted: September 14, 2006 at 11:42 PM (#2177271)
Are those pdfs still available? I remember looking at them at the time, but wasn't able to find them again when I was looking around a couple of weeks ago.
   240. jimd Posted: September 14, 2006 at 11:43 PM (#2177273)
Extreme teams represent the point at which WS distortions become obvious;

Exactly. The "good team" effect (extra total WS) affects Rizzuto at about .6 WS per season relative to a .500 team (e.g Sewell). The "good defense" effect (extra FWS) affects Rizzuto at some large but unknown fraction of 1 WS per season relative to average fielding teams (Sewell). The "overcompensating park factors" affects Rizzuto at another @1 WS per season relative to a neutral park (Sewell is penalized a similar amount for playing in a good hitting park). Each effect is small in and of itself, but when they all point one way, it causes illusions that could have been corrected for.

In this case, it all adds up to some 1.5 to 3 WS per season in Rizzuto's favor.

Sorry, Brent, I don't have time to dissect the next post in this long series.
It's most likely best that I don't since neither of us accepts as valid the primary evidence that the other uses (WARP1 fielding vs FWS) so it is probably past the point of being useful.
   241. Howie Menckel Posted: September 15, 2006 at 12:06 AM (#2177294)
"People were voting for Redding because he was a "career" value pitcher. The HoF data calls that assumption into question."

This is exactly why Redding fell from near the top of my ballot to off. I prefer long-career pitchers to high-peak ones, so I was a Redding guy. However rough the HOF figures are, I do now believe he didn't pitch that much in the 1920s. That's why he lost out. Mendez has been off my ballot as well and is hoping to get on there this 'year.'


Chris Fluit: I don't care much at all about home runs. What I care about is getting on base and producing a lot of total bases. OPS+, baby, followed by fielding, running, position, etc. adjustments.
I would just ask if you are sure you are considering the huge NEGATIVE value a caught-stealing has.

The new Baseball Prospectus book has the "break-even" mark for SB pct per year, but I don't have it handy. I think the supposition listed in a previous post was correct, though.
   242. Howie Menckel Posted: September 15, 2006 at 12:12 AM (#2177301)
Re 223: jschmeagol, no apology needed...
   243. OCF Posted: September 15, 2006 at 12:34 AM (#2177323)
A few of my thoughts on the SB:

I tend to disbelieve that batting behind a basestealer helps a batter - I think on net it hurts the next batter. Sure, Willie McGee had an outlier batting year (batting champ, undeserved MVP) in 1985 batting behind Vince Coleman, but I think it goes the other way more often. In particular, I think batting behind Brock probably hurt Ted Sizemore.

What kind of batter would you like batting behind a basestealer? Not a power hitter, since that makes the whole less than the sum of the parts - either the SB or the XBH is partially wasted. And not a strikeout-prone hitter, since taking pitches to let the runner steal might put the batter behind in the count and more likely to strike out. How about a high-contact, short-count, high-BA hitter? OK, so I'm describing Tony Gwynn. But Ozzie Smith (while not nearly at Gwynn's level) had some of the same virtues. I see a Gwynn-type hitter as being better able to withstand the damage of batting in that spot, and reaping the advatage of a lower GIDP risk. (And increased RBI chances, if you care about RBI.)

An extremely-skilled basestealer does "change the game," but the effect is really limited to whether that particular player will score a run - it doesn't extend beyond that, and in particular it doesn't have anything to do with big innings. Go look at the summary of the '67 World Series that I posted on the Brock thread. Look in particular at Game 1: A pitchers duel between Bob Gibson and Jose Santiago. It went into the 7th inning tied 1-1. And in the seventh: Brock led off with a single. Stole second. To third of a ground ball. Scored on a ground ball. So somehow the team got a run with only one baserunner in the inning, and that on a single. That mattered because that run mattered - and the Cardinals won 2-1. But that's it: a single run squeaked out in a close, low-scoring game.

The 1985 Cardinals were one of the greatest basestealing teams of all time and they did win the NL pennant. But they were also an outstanding pitching/defense team (and a challenge to those trying to figure out how much of pitching success is actually defense) and they did far and away lead the league in team OBP (by .010.) And at that they needed a little luck to hold off the Mets (just ask Sam M.) The basestealing was certainly part of their identity, but it was far from the only thing they did well.

With all of that, I am counting Brock's basestealing as a big positive for him, and I'm using metrics that count it so. But it only goes so far. It won't make Omar Moreno a good offensive player.
   244. Dr. Chaleeko Posted: September 15, 2006 at 01:08 AM (#2177361)
The Book by Tango, Lichtman, and Dolphin has this to say about the effect of basestealing. They studied pbp data over 2000-2004. They express their findings in woba, which is weighted OBP, which combines the features of SLG and OBP into a single number scaled to the OBP scale (300 bad, 340 average, 400 excellent). In their little summary boxes, ML&D say the following [which I hope they won't mind my quoting at some length. I'm not trying to pass it off as mine own or to sell anything with it, and at about 230 words, it probably falls within the boundaries of fair usage---at least it would at the company I work at, where we consider fair usage from a single book-length sourcework to be 300 words]:

"A runner on first with less than two outs is an enormous disruption on the defense. The batter gains 14 points of wOBA." (they also note that "when we talk about dirsupted defense, we don't necessarily mean a distracted defense.")

"The type of batter most dirsupted by having a runner on first base, with less than two outs, is the young, fast, and/or groundball hitter. Left-handed batters are the most likely to take advantage of the defense with a runner on first and less than two outs."

"The disruptive runner has an enormously negative influence on the batter, enough to almost completely offset the disruption caused to the defense."

"The stolen base attempt reduces the WOBA of the batter by 22 points, compared to the situation if the runner elected not to attempt to steal."

"The break-even point for the stolen base is highly dependent on the inning and the score. The most desireable situations are tied games in the later innings or ones in which the batting team is ahead. The least desirable situations are down by at least two runs in the later innings."

"A manager...cannot call for a steal so infrequently (unless his baserunnres are so incompettent and/or slow) that his opponent never has to pitch out. He must occasionally have his players tstel even with a negative epxectancy in order to force his opponent to occasionally pitch out."

So to summarize here, they seem to be arguing that there's a balance between the offense's need for tactical stealing and the defense's need to prevent teams from running against them at will. In addition, the disruption to the defense, not surprisingly helps lefty hitters, but it also hurts the very kind of hitter we expected it wouldn't hurt as much. Actually, let me restate that, they just don't get the same degree of benefit, in fact, they report it's a push for young, speedy, groundball types, while big slow old dudes gain more. And of course lefties over righties.

Anyway, the implicit message to me is that stealing's not really so great unless you've got a really good reason to do it, but you've got to steal a little bit, otherwise you won't get the benefits associated with the infield defending the steal. All of which says to me that Lou Brock could run as much as he wanted, and unless he stole at 100% it don't amount to much more of a hill of beans....
   245. Daryn Posted: September 15, 2006 at 01:58 AM (#2177394)
With all of that, I am counting Brock's basestealing as a big positive for him, and I'm using metrics that count it so. But it only goes so far.

That's the way I see it, too. It is a significant positive, no more. Particularly in Brock's low scoring era.

It won't make Omar Moreno a good offensive player.


Nor Scott Podsednik.
   246. Chris Fluit Posted: September 15, 2006 at 02:18 AM (#2177410)
Dizzypaco, let’s deal with a couple of your assertions.

One, you posited the possibility of a negative correlation specifically “the more stolen bases, the worse your team.”

Stolen Bases
2003: 3 playoff teams out of the top ten in stolen bases, including both World Series teams
2004: 4 playoff teams out of the top ten in sb, including 1 World Series team
2005: 4 playoff teams out of the top ten in sb, including both World Series teams
2006: 4 current playoff teams out of the top ten in sb, including both teams with the best record in their league

Looking at this chart, it’s pretty clear that the negative correlation does not hold. When nearly half of the playoff teams over the past 3-4 years are in the top ten in stolen bases, it’s clear that you can steal bases and still win games. As best as I can put it, Bill James’ hypothesis was “you can win games without stealing bases.” I’m not necessarily disagreeing with James, but I will say that the opposite (“you can’t win games by stealing bases”) is untrue. You can steal bases and win games. You can steal bases and win championships. You are not a bad team simply because you steal bases.

Now I recognize that a lot of really bad teams steal a lot of bases. Over the past four years, some of the traditional bottom-feeders like Baltimore, Tampa Bay and Kansas City have also routinely placed in the top ten in stolen bases. So, stealing bases does not guarantee winning. But does that mean that these teams are losing because they are stealing bases? I would posit that the inverse is true. Bad teams have trouble scoring runs. And so they are forced to rely more on stealing bases. But they’re not bad teams because they steal bases; they steal bases because they’re bad teams. After all, the earlier chart shows that some teams that steal bases can still win.

Two, you posited that home runs and on-base percentage correlate more closely with winning over a longer period of time.

Home Runs
2003: 3 playoff teams out of top ten in home runs, none in World Series
2004: 4 playoff teams out of top ten in home runs, including both World Series teams
2005: 4 playoff teams out of top ten in home runs, including World Series champion
2006: 3 current playoff teams out of top ten in home runs, including both teams with the best record in their league

On-Base Percentage
2003: 5 playoff teams out of top ten, including one World Series team
2004: 4 playoff teams out of top ten, including both World Series teams
2005: 5 playoff teams out of top ten, none in World Series
2006: 3 current playoff teams out of top, including the team leading the AL

Looking at this four-year period, home runs and stolen bases have nearly identical correlation to winning over the course of the regular season. However, the stolen base teams have a greater percentage of success in the postseason. On-base percentage does appear to have a higher correlation to winning, as on average one more top ten team makes the postseason than top ten teams in stolen bases and home runs. Yet once again, top ten teams in stolen bases fare better in the short series format of the postseason than top ten teams in on-base.

However, the big thing that jumped out at me as I did this comparison was not the straight comparison of stolen bases vs. home runs or stolen bases vs. on-base percentage. Rather, the most successful teams were the teams that showed up in multiple categories. For example, the World Series of 2004 pitted two teams- Boston and St. Louis- who were top ten in both home runs and on-base. The Cardinals were actually top ten in all three of these categories! The 2005 World Series pitted two teams who were both top ten in steals and the eventual winner was top ten in home runs as well. The best teams do more than one thing well. The Red Sox got on base and hit home runs. The White Sox stole bases and hit home runs. The 2003 Yankees got on base and stole bases. The best teams combine multiple strengths.

So yes, the stolen base can be overrated. You can’t win games or championships by stealing bases alone. But the same is true for hitting home runs or getting on-base. You need to do two of the three, and most of the great teams are doing it all.

On to a few other posts:

241. Howie Menckel Posted:
I would just ask if you are sure you are considering the huge NEGATIVE value a caught-stealing has.

The new Baseball Prospectus book has the "break-even" mark for SB pct per year, but I don't have it handy. I think the supposition listed in a previous post was correct, though.


I’d like to think I have. There is a negative value to stealing and that’s getting caught. I don’t remember what the “break-even” mark was but I know there is one and that it’s usually pretty high like 67%. And so yes, I’m looking at stealing percentage as well as totals.

But I see other voters completely hung up on the negative value of the caught-stealing so much that they appear to have blinders to both the positive effects of the stolen base and the negative effects of other strategies. A lot of big home run hitters also strike out a lot. There’s a negative effect to that strategy. A walk doesn’t move up a guy standing on third. So the on-base strategy sometimes results in more stranded runners but not more runs. The stolen base isn’t the only strategy with risk and reward, yet it seems that many voters here concentrate only the risk part of stolen bases and the reward part of other strategies. I think that we need the whole picture.

243. OCF Posted:
An extremely-skilled basestealer does "change the game," but the effect is really limited to whether that particular player will score a run - it doesn't extend beyond that, and in particular it doesn't have anything to do with big innings. Go look at the summary of the '67 World Series that I posted on the Brock thread. Look in particular at Game 1: A pitchers duel between Bob Gibson and Jose Santiago. It went into the 7th inning tied 1-1. And in the seventh: Brock led off with a single. Stole second. To third of a ground ball. Scored on a ground ball. So somehow the team got a run with only one baserunner in the inning, and that on a single. That mattered because that run mattered - and the Cardinals won 2-1. But that's it: a single run squeaked out in a close, low-scoring game.

I agree with all of that. The stolen base doesn’t help much if the player is stranded at second anyway. But the stolen base does get that player to second, increasing the chances that he will be driven in for a run. It doesn’t happen every time, but it does increase the chances.

And I agree about the value of a stolen base being different depending on a given game situation. A stolen base is more valuable in a 2-1 game than a 10-8 game. I thought I said as much in my earlier example. If Chipper Jones had hit three home runs, or if Tom Glavine had thrown a perfect game, then Rafael Furcal’s stolen bases wouldn’t have had nearly the impact. But those other things didn’t happen, making Furcal’s steals more valuable in that particular game.

I don’t think that stealing bases is the single most important part of the game, which some people assume must be the case as soon as you start arguing in favor of the steal. But I do think that the stolen base has some value. There are some runs over the course of a year that wouldn't have scored if not for a stolen base. There are some games over the course of a year that wouldn't have been won if not for a stolenb base. A successful stolen base does have positive value. The value may be 1/4 th of a home run. It may be 1/8 th as no other runners advance. It may be less than that, when you consider the risk of an out. But despite Dr. Chaleeko’s assertion, the value of a successful stolen base isn’t 0. And despite Dizzypaco’s assertion, the value of a successful stolen base isn’t a negative.

I don't have Brock and Aparicio 1 and 2 on my ballot. But they are on my ballot. They're there because of all of the things that they did, including stealing bases.
   247. sunnyday2 Posted: September 15, 2006 at 02:21 AM (#2177411)
>Looking at this chart, it’s pretty clear that the negative correlation does not hold. When nearly half of the playoff teams over the past 3-4 years are in the top ten in stolen bases, it’s clear that you can steal bases and still win games.

I don't think this is a valid conclusion from the data. Yes, you can steal bases and win. But is that what is being asserted? Do the steals actually help you win? The data doesn't say that. The method cannot say that.

And I think the other assertion is that the value of "base stealing," considering all the other impacts, is near zero even if you steal a bunch. i.e. considering CS, negative impact on the hitter, etc.
   248. Chris Fluit Posted: September 15, 2006 at 02:32 AM (#2177419)
Dizzypaco's negative correlation was "the more stolen bases, the worse your team." When roughly half of the playoff teams every year are in the top ten in stolen bases then I would say that yes, his negative correlation has been refuted.
   249. Chris Fluit Posted: September 15, 2006 at 02:34 AM (#2177420)
Continuing... I didn't with that chart say that "steals actually help you win." I did say that steals don't prevent you from winning. I eliminated the negative though I agree that I didn't establish the positive.
   250. Chris Fluit Posted: September 15, 2006 at 02:38 AM (#2177423)
And that's my final two cents on this subject. I know when I'm tilting at windmills.
   251. Brent Posted: September 15, 2006 at 03:31 AM (#2177448)
240. jimd wrote:

Exactly. The "good team" effect (extra total WS) affects Rizzuto at about .6 WS per season relative to a .500 team (e.g Sewell). The "good defense" effect (extra FWS) affects Rizzuto at some large but unknown fraction of 1 WS per season relative to average fielding teams (Sewell). The "overcompensating park factors" affects Rizzuto at another @1 WS per season relative to a neutral park (Sewell is penalized a similar amount for playing in a good hitting park). Each effect is small in and of itself, but when they all point one way, it causes illusions that could have been corrected for.

In this case, it all adds up to some 1.5 to 3 WS per season in Rizzuto's favor.


While I remember a discussion of the effect of good teams and how WS favors players on good teams because they don't have to play against themselves, my recollection was that we concluded the effects were much smaller, typically less than 0.5 WS per season. I'll admit I don't fully understand what is meant by the "good team" effect, "good defense" effect, and "overcompensating park factors," but I'll try to take a look at some implications of the numbers you've posted.

Over his career, Rizzuto's teams had a winning percentage of .634 (weighting each season's W-L data by Rizzuto's games played), which compares to an average winning percentage of .530 for Sewell's teams. The typical Rizzuto team went 98-56, while the typical Sewell team went 82-72. Because WS are constrained to equal 3 for each team win, that means Rizzuto's teams averaged 48 more WS per season than Sewell's teams.

I'll assume that these alleged biases affected not only Rizzuto but also DiMaggio, Mantle, Ford, and the other Yankee players. Let's take the mid-point of Jim's range, 2.25, and assume that the WS of Rizzuto's teammates were similarly biased. I'll blow up the estimate of bias for Rizzuto to an estimate for the entire team by multiplying 2.25 by 16.5. (I calculated the factor of 16.5 as follows: Rizzuto's teams averaged 98*3= 294 WS per season and Rizzuto averaged 231/13 = 17.8 WS per season; the ratio of 294 to 17.8 is 16.5 -- that is, (1/16.5) represents Rizzuto's average share of his team's WS over his 13 playing seasons.) The implication is that the alleged bias expanded to cover the entire team is 2.25*16.5 = 37.1 WS. However, since the entire difference between Rizzuto's teams and Sewell's teams is 48 WS, the implication is that the difference in teams' actual performance is only 11 WS, with the remaining 37 WS difference representing "good team" bias, "good defense" bias, and "overcompensating park factors."

I'm sorry, but these calculations just don't seem reasonable. I can accept that maybe 10% of the difference in measured WS between the teams may represent various biases, but not 75%. It seems to me that the estimates of these effects clearly need some more work and ought to be much smaller.

By the way, I am not trying to be an apologist for WS -- it certainly has its faults (as does WARP). In this particular comparison, I think the problems with WARP are more pronounced than those with WS, but we need to be critical of both systems.
   252. Brent Posted: September 15, 2006 at 03:45 AM (#2177453)
A comment on the comparison between Brock and Beckley with WS and with WARP -- a factor that wasn't mentioned was pitcher batting. OPS compares batters with a group that excludes pitchers, while WS does not exclude pitchers. Since pitchers hit quite a bit better in the 1890s than they did in the 1960s, there is a sort of weakened DH effect going on in this comparison. It's a small effect, but it contributes to the difference in the two metrics, along with the other factors that have been discussed.
   253. Dr. Chaleeko Posted: September 15, 2006 at 01:32 PM (#2177573)
Regarding team stealing, homering, etc.... Good teams tend to do everything well, and good teams often have a lot of speed in the lineup. Not speed as simply in base stealers, but as in players who possess speed and use it well. The greater effect of fast players is manifest on defense (esp in the outfield), in picking up extra bases, probably in batting average, and occassionally in stealing. So given these two things, it should be obvious that highly successful teams ought to rank highly in many or all categories because they have good players and many, many good players have good speed (and make good use of it---Alex Sanchez need not apply).

That said, there's nothing more lethal to a rally than getting nailed stealing or being pegged out on the bases. Nothing. The typical CS is worth around -.45 runs...more or less depending on your estimator of choice and in a modern run environment. The typical strikeout, as broached by CF, for instance, is worth -.10ish runs (or fewer). But compare to the payoff. We'll assume the payoff in a given situation is a steal for the stealer or a homer for the K artiste. The steal is worth between .15 and .20 runs, while the homer is worth roughly 1.4-1.5 runs. IIRC, this means the breakeven point is about 72-75%. In fact, let's say the SB is worth .18 runs and the CS worth -.45 runs. .45/.18 is 2.5, so a player must steal 2.5 bases for every CS, a little over 71%. The smallest real-life increment of which is going 5 out of 7. What about for strikeouts and homers? If a homer is worth 1.45 runs and a K is worth -.10 runs, then a guy can K 14.5 times per homer and break even. For a 30 homer hitter that's 435 strikeouts worth of homering. But on the other hand, a strikeout is just another out, and in fact it's only about .002 runs worse than making a regular old out. So looking at our 30 homer hitter, he can now strike out 725 times to break even versus making some regular old out (assuming no GIDPs of course). Which brings up why the CS is so deadly...because in every instance, <u>the rally has already started</u>. A homer hitter may always swing for glory, but not every swing takes place in a situation with increased run expectancy. Every time a guy is CSed, he's already in a situation where the run expectancy is heightened.

So let's use this theoretical framework to assess two guys named Reggie and Lou. Reggie always goes for the pump, Lou always goes for the bag.

Reggie hits 563 homers at 1.45 each to make 816 runs (rounded).
Reggie hacks out 2597 times at -.10 each to unmake 260 runs (rounded).
The net is 556.
(Also, Reggie "contributes" -11 net runs on steals attempts.)

Lou steals 938 bags at .18 runs each for 169 runs (rounded).
Lou gets nabbed 307 times for a net of -138 runs (rounded).
The net is a scant 31 runs.
(Also Lou contributes 43 net runs on homers.)

It's not a particularly realistic way to look at these kinds of contributions, but it's fun and it kind of drives home the point.
   254. DL from MN Posted: September 15, 2006 at 02:39 PM (#2177626)
As an aside, is stealing 2nd always more valuable than stealing 3rd? Is a CS at 3rd always more costly than a CS at 2nd? If both are true, your success rate stealing 3rd had better be much higher than your success rate stealing 2nd. I think the risk-reward is even worse stealing home.
   255. Mark Shirk (jsch) Posted: September 15, 2006 at 03:23 PM (#2177667)
DL,

I would think that it would cahnge depending on how many outs. Stealing third with 2 outs can't be as valuable as doing it with 0 or 1 outs. I would think that 1 out is the best situation to steal third in. Stealing 2nd with 2 outs can't be quite as costly as stealing 2nd with 0 or 1 outs correct? And with 2 outs I would think the benefit is greater. SO I guess my answer to your question is that it depends on game context, which is probably where you are as well.
   256. Mark Shirk (jsch) Posted: September 15, 2006 at 03:27 PM (#2177672)
Brent,

I think the park factor overcompensation that jim is talking about wouldn't have the effect of taking WS away from Yankee players but taking them from Yankee position players and giving them to Yankee pitchers (and I guess defenders which would make the effect a little less significant). However, I have never heard that WS overrates park effects before.

However, I agree on the whole that sometimes the effect of these adjustments is overblown. I mean can we really explain half the difference in wins (since WS is based on actual wins) between the Rizzuto Yankees and the Sewell Indians by such things as the Yankees didn't have to play themselves?
   257. DL from MN Posted: September 15, 2006 at 04:04 PM (#2177693)
Actually I'd think that stealing 2nd with 2 outs is potentially more costly because if you get caught your inning is over. The value of removing the force at 2nd (and DP opportunity) is lessened because the defense only needs the force at 1B.

I think we need something like WPA to really figure out the value of Brock's stolen bases.
   258. Mike Emeigh Posted: September 15, 2006 at 04:13 PM (#2177705)
However, I have never heard that WS overrates park effects before.


It does, but the effect is subtle.

Basically, what happens is that G/F ratio is overestimated by the proxy formula INF A/OF PO for a team that plays in a hitters' park, and underestimated for a team that plays in a pitchers' park. If the actual G/F ratio is, say, 1.1, it's likely that the INF A/OF PO ratio will be something like 1.2 in the hitters' park, and something like 1.0 in the pitchers' park. There are two reasons for this:

1. It's easier to get infield assists when there are runners on base, and even though James adjusts for DPs, he doesn't adjust for secondary causes of those extra assists (runners thrown out on an infield relay from the OF, rundown plays, force plays on grounders where the infielder would have had no chance to make a play at first). This is also a secondary effect of playing with a bad defense, as opposed to a good one, so the G/F ratio will often be overstated for a bad defense as well.

2. Park effects tend to be larger for fly balls than ground balls - thus a greater percentage of hits-converted-to-outs (or vice versa) are on fly balls than on ground balls. I also have noted that the effect of good team defense tends to be larger in the outfield than in the infield, as well. It is extremely rare to have good team defense without good outfield defense, but less rare to have good team defense without good infield defense. I know this is counterintuitive to some people, but from what I can tell it appears to be true.

The net effect of both of these is that for someone like Sewell, playing on a poor defensive team in a hitters' park, the G/F factors work against him, making it appear as though the team allowed more GBs than it probably did. For someone like Rizzuto, playing on a good defensive team in Yankee Stadium, the G/F factors work in his favor, making it appear as though the team allowed fewer GBs than it probably did.

-- MWE
   259. TomH Posted: September 15, 2006 at 05:04 PM (#2177746)
according to linear weights, a CS is worth about -.45 runs, but a KO is worth -.28; NOT -.10.

The -.10 comes from measuring fewer runs created (stranding potential runners), but does not account for the extra out.

In the low run 1960s environ, a CS is slightly less costly.

--

Speaking of stealing 2B vs 3B; to analyze steals of 3B correctly, you would have to also find how many were double steals. If Brock steals third and a trailing runner picks up the bag behind him, surely Brock ought to get partial credit for the steal of 2nd, should he not? We can't assume they guy would have pilfered the bag safely if not for Brock's move ahead of him.

The LW gain of a double steal (with 1 out) is about .47. A caught stealing of third, while the trailing runner goes to 2nd, is -.62, for a breakeven % of a mere 57%.

Yes, once in a long while the trailing runner will get caught stealing, which would alter the numbers. However, some times Brock will get "caught stealing" officially while the throw gets dropped and he is safe. Or the throw goes into CF and he gets another base. Like analyzing the sac bunt and needing to account for failed fielders choices, etc., the breakeven point is not so cut and dried.
   260. Dr. Chaleeko Posted: September 15, 2006 at 05:25 PM (#2177759)
DL,

Using Tango's base-out run expectancy tables (at http://www.tangotiger.net/RE9902.html), we could make a guess.

initial    RUN EXP    RUN EXP     difference difference     sb:cs     break even 
BASE
-OUT   RUN EXP   AFTER STEAL AFTER CS   after steal after CS     break even   sb pct
-----------------------------------------------------------------------------------------------
1-0         .953       1.189      .297        .236      -.656       2.78:1       73.5%
1-1         .573        .725      .117        .152      -.456       3.00:1       75.0%
1-2         .251        .344      .000        .093      -.251       2.70:1       73.0%

2-0        1.189       1.482      .297        .293      -.892       3.04:1       75.2%
2-1         .725        .983      .117        .258      -.608       2.36:1       70.2%
2-2         .344        .387      .000        .043      -.344       8.00:1       88.9%

3-0        1.482    1 .555      .297        .073     -1.185      16.22:1       94.2%
3-1         .983    1 .297      .117        .314     .866       2.76:1       73.4%
3-2         .387    1 .117      .000        .730     .387        .53:1       34.6


I may be wrong about the runners at third since the run expectancy tables presumably include steals of home, but those are so rare anymore as to be infintesimal in a table based on 4 or 5 years of games, and you've got to figure that if a guy triples and steals home, say, the run expectancy must be his tally plus whatever expectancy would come after that once the bases are clear again. I don't know if that's accurate analysis or not, so please let me know if not.

It appears that the steal of second generally requires around a 73.5% rate of success to be worth it. While a steal of third requires something closer to 80% to be worth it. In other words, you need to be Raines or Beltran to steal third and gain something from it. Steals of home appear to require a much lower rate of success, however, it seems unlikely that anyone could reasonably expect to match even the modest 34.6% rate required with two outs. Stealing home isn't easy, and to the best of my limited knowledge the lifetime leaders in steals of home are in the teens or twenties. (Someone please correct me!)

So returning to DL's question, isn't it smarter to steal second than third? Yes, I think it must be, though it's reputedly easier to steal third. But oddly enough there's not a gigantic difference between them overall (though there is a difference in certain situations, esp with a man at second and one out or two outs), perhaps as little as five percentage points of success.

But turning to another point of view, what do you get for risking an out? The average MLB team from 1999 to 2002 attempted about 600 steals in the four years (actually closer to 590, but who cares for now?) and about 150 per team a year. Let's say those steals were 75% of second and 25% of third, and further that the out scenarios were 45% for 0 outs and 45% for 1 out and 10% for two outs. The AL stole at around 69% in this period. Assuming that the successes were distributed equally around the base-out situations (a big if, of course), we figure that

initial    RUN EXP    RUN EXP     difference difference     sb:cs     break even 
BASE
-OUT   RUN EXP   AFTER STEAL AFTER CS   after steal after CS     break even   sb pct
-----------------------------------------------------------------------------------------------
1-0         .953       1.189      .297        .236      -.656       2.78:1       73.5%
1-1         .573        .725      .117        .152      -.456       3.00:1       75.0%
1-2         .251        .344      .000        .093      -.251       2.70:1       73.0%

2-0        1.189       1.482      .297        .293      -.892       3.04:1       75.2%
2-1         .725        .983      .117        .258      -.608       2.36:1       70.2%
2-2         .344        .387      .000        .043      -.344       8.00:1       88.9%

           
initial    RUN gained  RUNs lost    NET      NET   PCT RUNS
BASE
-OUT   RUN EXP   AFTER STEAL  AFTER CS   SB RUNS    RUNS  GAINED/LOST
----------------------------------------------------------------------------
1-0          193         +33       -41         -8       185      .96  
1
-1          116         +21       -29         -8       108      .93    
1
-2           11         3       4         -1        10      .91
____________________________________________________________________
FIRST BASE   320         
+57       -74         -17      303      .97   

2
-0           80         +14       -19         -5        75      .94  
2
-1           49         +12       -13         -1        48      .98   
2
-2            5         0.4     2         -1.6      3.4     .68    
____________________________________________________________________
SECOND BASE  134         
+26.4     -34         -7.6     126.4    .94  

=============================================================================
TOTAL        454         +83.4     -108       -24.6     429.4    .95 


Pretty interesting. So the average team (provided my 75/25 and 45/45/10 assumptions are anywhere near valid) are shooting themselves in the foot by 5% over the four years. (And of course, this is probably a bigger issue in the AL than the NL.) This doesn't take leveraged steals into account, because it doesn't calculate the effect of stealing in the late innings versus earlier. Still, teams are stealing around 150 times a year, or 17 times per lineup slot or just under once per game. They may be costing themselves about .04 runs per attempt in so doing. Again, without leveraging information it's hard to know when teams chose to steal, and furthermore, without specific breakdowns of how frequently they stole in each base-out situation it's hard to do more than speculate. It's also hard to say how many are busted hit-and-runs too. I'm also not addressing whether or not it's easier to steal third than second, which could change the success rates I'm using for each and slightly change the numbers we're talking about.

But the big elephant, is that the net change in runs over 600 steals over four years, a team's net runs on steals, is utterly miniscule versus its total runs expected in a situation, let alone it's total runs scored. The steals come out to less than a 25 run swing on 450 expected runs in the situation and about 3200 total runs over four seasons.

Which further underscores why even the greatest, most extreme basestealer you can ever imagine (say, 130 steals, no CS) would net less than 25 runs (2.5 wins) on his legs alone in one single season.

Well, i've lost track of DL's original point, but in the continuing service of making sure that everyone in the whole wide world knows that the steal is nearly worthless at face and only takes on value in moments of extreme leverage, I'm happy to make big charts.

I just hope i interpreted them correctly.
   261. Dr. Chaleeko Posted: September 15, 2006 at 05:40 PM (#2177776)
but less rare to have good team defense without good infield defense. I know this is counterintuitive to some people, but from what I can tell it appears to be true.

This does make sense in a way. If your outfield defense has awful range, a LOT of hits will drop in. If your infield defense has awful range, it's not as big a problem because range in the infield probably only means a step or three, depending on position. So fewer hits allowed, and, naturally, fewer extra base hits, and also, likely, fewer runners advancing 1st-to-3rd or maybe even 2nd-to-home on the infield hits versus the gappers and bloops falling into the lumbersome outfield.

according to linear weights, a CS is worth about -.45 runs, but a KO is worth -.28; NOT -.10.

TomH, I was using Furtado's xR formula as the basis of my comments. Why the diff between the two?
   262. DL from MN Posted: September 15, 2006 at 05:52 PM (#2177790)
The other circular part of this is with the higher success rate required, only top basestealers will attempt to steal 3rd. Of course these players probably have a higher rate of scoring from 2nd than your average player also.

The charts above are fabulous, but they're also the "mean" that doesn't take into account the actual game situation and the abilities of the players. A player may decide to run on a good catcher with a lefty on the mound if his team's pitcher is at the plate and they will probably pinch hit for the guy if he leads off the next inning. In that situation the risk is pretty much nil and the reward doesn't have to be much to justify the attempt. By the same account you might not steal 2nd if your slugger is up at the plate with a 2 ball count and 1 out and they'll just walk him to put the force back on. The leverage aspect really does matter with stolen bases, perhaps more than any other offensive statistic.
   263. TomH Posted: September 15, 2006 at 07:03 PM (#2177880)
Furtado's xR formula I think predicts total runs scored, like James' RC.

Using RC (I am more familiar with it), if a team has a game where 11 men reach base in 37 PAs and they get 13 total bases, the RC formula says they should score 11*13/37 = 3.86 runs. Add another out and it is 11*13/38 = 3.76. Minus .11 runs. But this "out" also cost them 1 of their 27 outs; in the first example, they still had one out to go in the 9th inning. 1/27th of 3.86 runs is .14, so the total cost of the out is .11+.14=.25. Or something like that.
   264. Mike Emeigh Posted: September 15, 2006 at 07:45 PM (#2177939)
The leverage aspect really does matter with stolen bases, perhaps more than any other offensive statistic.


Basestealing typically "isn't" indiscriminate, either; it tends to be used in very specific situations, even by a Wills/Brock/Rickey.

At least in recent years, basestealing has tended to be an early-inning strategy; in high-leverage situations later in the game, teams tend to sacrifice instead of trying to steal. I haven't yet looked at all of the Retro-years to see how basestealing strategies have evolved from the sixties going forward.

-- MWE
   265. KJOK Posted: September 15, 2006 at 09:24 PM (#2178040)
according to linear weights, a CS is worth about -.45 runs, but a KO is worth -.28; NOT -.10.

Very important point - if your baseline is AVERAGE RUNS, an OUT value is around -.10, and a CS is around -.31. If you're using a baseline of "ZERO RUNS", an out is around -.27, and a CS is around -.49.
   266. Jeff M Posted: September 16, 2006 at 04:16 PM (#2178783)
How can Kiner be top 10, and Oms #30 in the final tally? Has Oms been forgotten?

I've never really campaigned, and I'm not trying to convince anyone to put Oms #1 on the ballot. However, I do think he deserves a look from some who may have forgotten him, and others who never have looked. Here's some data:

Kiner has 242 WS and Oms has about 340 in Chris Cobb's MLEs. Oms played CF, one of the toughest defensive positions, and Kiner played LF, one of the easiest. We believe Oms was a good, but not incredible centerfielder. Kiner was a weak outfielder.

It can't be the peak argument, can it? Because (1) Oms has a slightly better seven year WS peak (196.7) than Kiner (192.7) and (2) although Oms lags Kiner in 3-year and 5-year peak, Chris Cobb has reminded us frequently that the way the MLEs are done means that we cannot accurately evaluate peaks for Negro Leaguers, particularly over only a couple of seasons. See post #121 here: Alejandro Oms MLEs

To quote Chris Cobb (the MLE architect): To sum up, the MLEs become more directly comparable to mlb win shares the larger group of seasons you consider. I don't think a three-year, non-consecutive peak measure offers a fair comparison between players with MLE records and players with mlb records. Five-year consecutive, or ten-year consecutive will yield a truer comparison." Kiner does win the five-year consecutive peak measure, but if go any further and Kiner falls behind.

I can't think of any other situation where player A (1) has an approximately equivalent peak to player B (depending on how you measure peak), (2) exceeds player B by 100 career WS, (3) played a more difficult position than player B and (4) was better at playing that tougher position than player B...

...yet player A lags player B in the final vote tally by 24 spots and 246 votes! I think Kiner was a better hitter, but not by the margin reflected in the final vote. And Kiner played five fewer years at a weaker defensive spot.

See post #107 for the MLEs and post #108 for the WS estimates here: Alejandro Oms MLEs
   267. Jeff M Posted: September 16, 2006 at 04:20 PM (#2178787)
...but if go any further and Kiner falls behind.

Huh? I must need to eat lunch. Try "...but if <u>you</u> go any further<u>,</u> <strike>and</strike> Kiner falls behind."
   268. Jeff M Posted: September 16, 2006 at 04:47 PM (#2178803)
If you're using a baseline of "ZERO RUNS", an out is around -.27, and a CS is around -.49.

Outs are around -.31 since 1994 and were as low as -.22 in the deadball era (in the AL anyway).
   269. Mark Donelson Posted: September 16, 2006 at 05:02 PM (#2178811)
I need to look at Oms again, but to take on this point:

It can't be the peak argument, can it? Because (1) Oms has a slightly better seven year WS peak (196.7) than Kiner (192.7)

To me, seven years is not a peak, unless you're talking about the Mays/Mantle/Wagner/Williams/Cobb types. You're comparing primes in my book.
   270. Mark Shirk (jsch) Posted: September 17, 2006 at 01:11 AM (#2179174)
I agree, Kiner's peak is a decent bit better than Oms. However, I do have Oms in my top 25 (maybe as high as #20 this year I am not sure) and I feel he was better than the already elected Willard Brown. We also have to realize that the MLE's are nothing more than an educated guess at what Oms value during his career was.

However, neither was as good as Charlie Keller...;-)
   271. Dr. Vaux Posted: September 17, 2006 at 02:04 AM (#2179214)
I haven't been able to find time to re-evaluate my ballot this week, so you'll have to have the election this "year" without me. I hope to be back next "year" with a more balanced voting-method.
   272. mulder & scully Posted: September 17, 2006 at 04:17 AM (#2179245)
We'll miss your vote.

I look forward to seeing your ballot again.
   273. Devin has a deep burning passion for fuzzy socks Posted: September 18, 2006 at 01:53 AM (#2180097)
The stealing analysis brings to mind an odd situation I saw in a minor league game this year. If you have 2 outs, a fast runner on third base, and the pitcher batting, does it make sense to try and steal home?* The pitcher is likely to end the inning, so a CS isn't as damaging as usual (although it does mean the pitcher will lead off the next inning). I would think the break-even point would be substantially lower than a normal stealing decision, although obviously steals of home are much more difficult.

(*It's an odd situation because lineup construction tendencies don't have fast guys batting 5-7, which is who would normally be on base at this point. For some reason the Binghamton Mets were batting Carlos Gomez 5th.)
   274. OCF Posted: September 18, 2006 at 01:59 AM (#2180113)
It's an odd situation because lineup construction tendencies don't have fast guys batting 5-7

And why not? Tradition, and all of that, of course. But doesn't it make sense to send up your base stealer after your Pujols/Manny/Ryan Howard bats and not before?

(This thought not original with me, of course.)
   275. Dr. Vaux Posted: September 19, 2006 at 02:10 AM (#2181445)
Thank you, mulder & scully. I should be back soon.
   276. DL from MN Posted: September 19, 2006 at 01:32 PM (#2181816)
> If you have 2 outs, a fast runner on third base, and the pitcher batting, does it
> make sense to try and steal home?

If you have a lefthanded pitcher who throws a lot of nasty breaking pitches, go for it.
   277. Paul Wendt Posted: June 01, 2007 at 10:20 PM (#2387686)
This page is not linked to the directory under Important Links
   278. John (You Can Call Me Grandma) Murphy Posted: June 01, 2007 at 11:34 PM (#2387813)
Again, I don't know what happened to the link, Paul. It even looked right. I retyped it and it's working now.
   279. Paul Wendt Posted: June 02, 2007 at 04:15 PM (#2388930)
Odd. '1985' was bold black and I reloaded a couple of times.

If I am archive-diving again, I'll try again next day before writing. "Take two aspirin and call me in the morning."
   280. John (You Can Call Me Grandma) Murphy Posted: June 02, 2007 at 06:31 PM (#2389099)
If I am archive-diving again, I'll try again next day before writing. "Take two aspirin and call me in the morning."


:-)
Page 3 of 3 pages  < 1 2 3

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
Guts
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Syndicate

Page rendered in 0.8751 seconds
67 querie(s) executed