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

Thursday, January 19, 2006

Gil Hodges

Eligible in 1969.

John (You Can Call Me Grandma) Murphy Posted: January 19, 2006 at 03:40 AM | 74 comment(s) Login to Bookmark
  Related News:

Reader Comments and Retorts

Go to end of page

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

   1. John (You Can Call Me Grandma) Murphy Posted: January 19, 2006 at 03:43 AM (#1827892)
What a shortstop he could have been.
   2. KJOK Posted: January 20, 2006 at 05:55 AM (#1829702)
Not much going on here in Hodges land, so I'm going to double post this, since it pertains to Hodges, the argument FOR Hodges centers largely around being the best 1st basemen in his time, and I'm one of the biggest proponents of comparing players to others at their position in their "era":

Mize, McCovey, Dick Allen, Killebrew, Cepeda, and Boog Powell all rank ahead of Hodges in my 'era' list, in addition to "part" of Musial, and also Luke Easter. I can't see voting for Hodges.
   3. sunnyday2 Posted: January 21, 2006 at 03:46 AM (#1831135)
Here's a surprise or maybe not. Seasons OPS+ > 110.

Hodges 142-42-39-37-28-25-23-22-19-12
Ashburn 142-36-22-22-19-16-10-10-10

Career

Hodges 119 in ~8300 PA
Ashburn 111 in ~ 9550 PA

But of course there's that little thing about the A+ CF vs. a B 1B, so WS:

Ashburn 329/29-28-28/137/24.3
Hodges 263/29-26-25/129/20.6

Still I'd have to believe the average baseball fan of the '50s would have taken Hodges over Whitey. I'd guess a more realistic comparison would be Hodges or Mickey Vernon.
   4. OCF Posted: January 21, 2006 at 07:29 AM (#1831262)
I've got to repeat that using not OPS+ but my own favorite RCAA-based tool:

Ashburn 49 48 41 36 34 33 29 28 23 20 20 14 -2 -2 -3
Hodges- 46 46 43 40 33 29 27 27 25 22 5 0 0 -1 -2 -7 -7 -9

So on this scale Ashburn was a tiny bit better offensively than Hodges. A couple of things to point out. The first is that Ashburn was OPB-rich, and thus looks better by RC/league than by OPS+. The other is that Ashburn, as a leadoff hitter, had more opportunities than Hodges (the effect being blunted in RCAA by the fact that it's also more opportunity to make outs.)

Where do I actually have Hodges' offense? About equal to Ed Konetchy. Of course, Konetchy isn't on my ballot. I also have Vernon ahead of Hodges, primariy on longevity grounds - but then I am the best and maybe only FOMV.
   5. Howie Menckel Posted: January 29, 2006 at 08:54 PM (#1842693)
Beckley vs Hodges vs Vernon, OPS+ as regular, 100 OPS+ minimum:

Beckley: 152 44 38 33 31 28 27 27 26 26 26 24 22 12 12 05 02
GHodges: 143 42 41 38 26 26 24 21 13
MVernon: 160 49 37 30 30 25 21 13 13 12 10 04 04

I have Beckley high because he has such a long tail of useful play, and because 1B was a more important defensive position back then. Hodges doesn't go much deeper than Vernon, and fails at a top-2 peak comparison.

Boy, my father could walk to Ebbets Field as a kid, and he used to sit in the bleachers for a nickel.
I won't tell him what this means for ol' Gil on my ballot.
He's got 4 very solid seasons, and 4 or 5 more of some value. But he needs either a higher peak or more longevity.

His widow, Joan, still lives in Brooklyn. A recent story on her said that old-timers recognize her on the street all the time. Sweet lady, but looks like we can't make her day....
   6. John (You Can Call Me Grandma) Murphy Posted: January 29, 2006 at 09:38 PM (#1842727)
Boy, my father could walk to Ebbets Field as a kid, and he used to sit in the bleachers for a nickel.

My great-grandfather Murphy had his own box in the teens, while my father used to ride his bicycle from Borough Park to Flatbush Avenue and met all of the Bums. :-)

The most passionate Dodger fan, however, was my great-grandmother Richardson. She never missed a game on radio or TV and was very knowledgeable about baseball history and statistics in general.
   7. OCF Posted: January 30, 2006 at 06:30 AM (#1843404)
As a response to Howie's post #5: I've posted this before, but in two different places. Consolidating in one place (this is my modified RCAA offensive system)
Hodges  46 46 43 40 33 29 27 27 25 22  5  0  0 -----9
Vernon  63 57 39 32 27 21 20 20 20  
*  * 19 18 16 13 1 0 ----9-24
Beckley 38 36 34 29 29 27 24 20 20 20 19 19 15 15 13 10 8 4 
-8-11 

I have Vernon on my ballot - I think I'm his best friend. I won't have Hodges on my ballot.
   8. Howie Menckel Posted: January 30, 2006 at 02:15 PM (#1843597)
OCF,
Where do you find the biggest differences arise?
Is that adjusted for schedule length?
Does it recognize the mostly unargued greater importance of defense at that time (OPS+ doesn't, but it's neutral, at least)?
   9. Dizzypaco Posted: January 30, 2006 at 03:02 PM (#1843618)
But of course there's that little thing about the A+ CF vs. a B 1B, so WS

We don't know that Hodges was a B first baseman - his reputation was clearly that of an A, and defensive statistics for first basemen are less reliable than for other positions. I'll grant you that an A CF is more important than an A 1B.

I'd take Hodges way before I'd take Vernon. Hodges had a 8 to 10 year peak in which he was a very solid ball player every year. Vernon was extraordinarily inconsistent - you couldn't count on him for anything. Consistency is one of the most underrated factors, IMO.

That said, I probably wouldn't vote for either of them. Hodges is close, but his peak was a little too short and and not quite good enough.
   10. Dr. Chaleeko Posted: January 30, 2006 at 04:09 PM (#1843680)
FWIW: Win Shares indicates that both Vernon and Hodges were the best 1B in their leagues four times.
   11. Mark Shirk (jsch) Posted: January 30, 2006 at 04:13 PM (#1843682)
Howie,

RCAA I would think would also be nuetral, since it is an offense only metric.

Dizzy,

I must say that I don't put any weight on consistency. When you are thinking about trading for a guy then yes, consistency matters, but when you are looing over his accomplishments, I don't see how it matters if a player had a bad year or two mixed in with his good years so long as those good years as as good as the player who was good consistently for the same length of time.
   12. Dizzypaco Posted: January 30, 2006 at 04:42 PM (#1843733)
I must say that I don't put any weight on consistency. When you are thinking about trading for a guy then yes, consistency matters, but when you are looing over his accomplishments, I don't see how it matters if a player had a bad year or two mixed in with his good years so long as those good years as as good as the player who was good consistently for the same length of time.

J,

It matters a lot if you place any weight on peak performance - how good was this player when he was playing at his highest established level of ability. Even at Vernon's highest established level of ability, you never knew what you get from him. The fact that Norm Cash had an absurd year in 1961, or Brady Anderson hit 50 home runs in 1996, or Luis Gonzalez the year he hit 57 home runs, does not mean that their peaks were among the greatest of all time, IMO; a single great year, or a couple of great years spread out over a period of time does not represent a peak.

When evaluating a player, I use a combination of peak and career, so it matters in my evaluations.
   13. Mark Shirk (jsch) Posted: January 30, 2006 at 05:07 PM (#1843788)
I am a peak/prime voter and I don't put any value on consecutive performance. We are looking at what a player DID, what he ACCOMPLISHED. It doens' matter ot me if his manager didn't kwno if he was going to be an MVP level performer or solid player from year to year so long as he has enough MVP level performances in his career. If we are doing projections, or thinking about trading for a player, consistency would matter to me, but not in HOM debates.
   14. Dizzypaco Posted: January 30, 2006 at 05:35 PM (#1843821)
I am a peak/prime voter and I don't put any value on consecutive performance. We are looking at what a player DID, what he ACCOMPLISHED.

I agree with you. Over Hodges best five year (1951 to 1955) period of his career, he had 129 win shares. Over Vernon's best five year period (also 1951 to 1955), he had 112 win shares. In another words, Hodges accomplished more than Vernon (using win shares - other stats I believe will tell you the same thing). You can change it to 4 years, or six, or whatever, and you would come out with the same thing. To me, that's peak -how much did a player accomplish at his highest level of ability. A player having one great year, and then another one 14 years later, is not establishing a level of ability.

By the way, the fact that I look at what a player did, or accomplished, means that I disagree with almost everyone who gives extra credit for war service - a player by definition has no value if he misses the season. The war merely explains why didn't have any value. But then again, I'm opening up a different can of worms.
   15. Mark Shirk (jsch) Posted: January 30, 2006 at 06:53 PM (#1843942)
That is only if you define 'level of ability' as consecutive five year peak. I agree that a one year peak is insufficient, but why does a peak have to be consecutive? If you take the five best non consecutive years of Vernon and Hodges, how do they measure up (I don't have my database here at work)? To give extra credit for consistency is to value a pattern, not an accomplishment. This is fine in the present tense where you might pay a player a little more or give up more to acquire him because he is guaranteed to give you a certain level of performance. But I don't see how it should affect HOM voting. I think a player accomplishes three MVP seasons whether or not they happen to come one after the other or with a year or two in between. The accomplishment is the same. At the very least consecutive and non consecutive totals should be equal, though I think that is overrating those who were good for a consecutive number of years.

As for war credit, that effects an entire generation of players not just a guy or two. Giving no war credit will lead to the systematic underrating of the war generation.
   16. John (You Can Call Me Grandma) Murphy Posted: January 30, 2006 at 07:05 PM (#1843966)
As for war credit, that effects an entire generation of players not just a guy or two. Giving no war credit will lead to the systematic underrating of the war generation.

Exactly. Besides, if you don't take the war into account, all you will do is wind up electing Hall of Very Good players instead from other generations.

As for consecutive seasons, I could care less. Peak to me is the player's three best seasons, not his three best in a row.
   17. Dizzypaco Posted: January 30, 2006 at 07:07 PM (#1843971)
Giving no war credit will lead to the systematic underrating of the war generation

That's true only if you are measuring ability, rather than what they actually did, or accomplished. If you look at what they actually did, you'll find that the war generation accomplished less than other generations, as a result of the war. I obviously don't blame them for this - its just the way it is. On the other hand, there is no reason why the peak value of a war time player has to be worse than anyone else, so if you consider peak relatively strongly, it will make up for some of the problem. I'd rather do that then pretend like they played those years, which is what you are doing if you give war credit.

As to the consectuive versus non-consecutive argument, I think you are absolutely right when it comes to career value. When measuring someone's career value, it doesn't matter so much when you do it. However, when it comes to measuring their peak value, I think it matters quite a bit when you do it.

Lets say that a player, was great in May and September of his rookie year, July and August of his second year, etc. When evaluating his peak, would you just throw out what the player does in the other months? Of course not - every month counts, just like every season counts. An established level of ability is not what you do over a random group of games, or weeks, or months, or even seasons. Its what you do over an extended period of time - more than a season, and over a consecutive period of time.
   18. DavidFoss Posted: January 30, 2006 at 07:24 PM (#1844004)
Lets say that a player, was great in May and September of his rookie year, July and August of his second year, etc. When evaluating his peak, would you just throw out what the player does in the other months? Of course not - every month counts, just like every season counts. An established level of ability is not what you do over a random group of games, or weeks, or months, or even seasons. Its what you do over an extended period of time - more than a season, and over a consecutive period of time.

Taking things a bit too far to the "strawman" extreme there. Peak voters generally go for a single season as being the smallest meaningful quantity -- goes with the 'Pennant is a Pennant'. That's why in-season durability often matters to the peak voter whereas career-length durability does not.

I understand your logic with the consecutivity of peak seasons argument. Its the mythical multi-year contract that you are signing your HOM candidate for in the big all-star game in the sky. There's been much debate about this. Cases like Campy 53-55 and Bench 70-72 are classic examples where a player has a horrible year in between two MVP seasons. The teams fortunes match the players as well. In these two cases, its moot as these guys are shoo-ins but you get the general idea.

If you replace those three seasons with a three year average each season its quite possible a team that went 1-4-1 in the standings could end up 2-3-2 and pennant-less. (Probably not the case for the specific case of the Dodgers and Reds teams of Campy and Bench, but generally its possible).

Anyhow, consecutive vs non-consecutive is a legitimite debate. I suspect some do one, some do the other -- Bill James actually incorporated both.
   19. Dizzypaco Posted: January 30, 2006 at 07:38 PM (#1844046)
Thanks David, I think you summed it up pretty well. My only disagreement is that one season is generally not considered the smallest meaningful quantity - once again, I don't hear too many people naming Brady Anderson and Luis Gonzalez among the list of the greatest players of all time, or the greatest in recent years, even when focusing on peak value as the main criteria.
   20. DanG Posted: January 30, 2006 at 07:47 PM (#1844073)
The argument seems to be saying that there is value in consistency, in predictability of performance. I can see where a team would benefit by having this type of player, knowing that you have a Steve Garvey on hand you can direct your resources to other team needs.

The counter argument would be that it is the Mickey Vernon type who actually has more pennant impact. When he has his superstar year out of the blue, he is more likely to put his team over the top.

To me, it's a wash. I don't worry about consecutive years. Anyway, a major hazard of emphasizing that is it draws you into the quagmire of how many years is a peak? Wherever you subjectively draw the line will unfairly favor a guy who confines his best years to your boundaries.
   21. DavidFoss Posted: January 30, 2006 at 07:58 PM (#1844107)
My only disagreement is that one season is generally not considered the smallest meaningful quantity - once again, I don't hear too many people naming Brady Anderson and Luis Gonzalez among the list of the greatest players of all time, or the greatest in recent years, even when focusing on peak value as the main criteria.

Those flukey seasons will show up on their resumes though. Its not like the year that Mike Bordick hit .352/.365/.682 in April (2000 for those who are curious).

No one can take that 157 OPS+ from Brady and that 176 OPS+ from Luis. That's a real quantifiable performance that -- at least for that year -- helped out their teams immensely. Luis's five year consecutive OPS+ peak (sorted) is 176-142-132-131-128. Sure the 176 is flukey, but he should get credit for it. You can't just throw it out and replace it with another 142. This came up with Hugh Duffy and Mules Suttles before and we'll see it again with Norm Cash.

Yes, peak voters do look at multiple seasons, what I meant was that each individual season acts as the smallest block of performance you can look at within a peak -- at least that's how I look at it.
   22. Mark Donelson Posted: January 30, 2006 at 09:31 PM (#1844270)
I also don't see consistency as a necessary quality in my analysis of peaks (it can be a tiebreaker, I suppose, in the case of otherwise even candidates, but that's it). My logic is simply that the goal of every time is that pennant, a single-season goal. You don't win anything if your team just performs well in April, but if you perform well all year, you do. So a full-year great performance, even if flukey, counts in my book, where as a one-month great performance does not (except to the extent that it helps the full-year performance).

So if Luis Gonzalez manages to post two or three more seasons like the 176 OPS+ one, he gets more serious consideration from me, regardless of when he does so.
   23. Mark Shirk (jsch) Posted: January 30, 2006 at 09:59 PM (#1844319)
Yeah, I would like to state that I don't look at anything smaller than a season.

I think a good example of this is Edd Roush vs. Max Carey. Roush has great per game or per 162 game rates, high enough that you would think he had a pretty nice peak. But for a number of reasons (hold outs, littel injuries) he wasnt' very durable in season. So his peak is very similar to that of Carey's, a player who doesn't have the per game rates of someone like Roush. So I have Carey about Roush (a little extra career value in my opinion as a tiebreaker). It may be one of the reasons why Carey is in and Roush is out, though I have neither in my PHOM.

Of course none of that settles our consistency debate. To me, looking only at five consecutive years (or 7 or 6 or whatever) would miss something like Early Wynn's 1959, which was a very fine season and helped his team win a pennant. That to me should be counted amongst peak. Also, I dont' put much stock in per 162 game rates or other rate stats because they dont' really measure peak.

Five year consecutive is one way to measure peak and it isn't an awful way, I just think that you need to incorporate 5 year non consecutive or 7 year non consecutive, or whatever. O5 course what I do is measure WS acumulated over 15 and WS accumulated over 25 (4.5 and 8 for WARP). This is a littel susceptible to one big year, but I balance that out a little also looking at best 3/5/7 years and by subjectively lowering that player a tead.
   24. Paul Wendt Posted: January 30, 2006 at 11:34 PM (#1844483)
jschmeagol #13 and Dizzypaco #14, indicative of #9-23

I must say that I don't put any weight on consistency. When you are thinking about trading for a guy then yes, consistency matters, but when you are looing over his accomplishments, I don't see how it matters if a player had a bad year or two mixed in with his good years so long as those good years as as good as the player who was good consistently for the same length of time.

J,

It matters a lot if you place any weight on peak performance - how good was this player when he was playing at his highest established level of ability.


I agree with Dizzypaco thruout this thread except that I think he concedes too much when he says "I agree with you" or something similar. That means I disagree with a fairly long list of prominent HOMeboys. (Perhaps Dizzypaco is Phillybooster after another name change. I haven't lately read much from Matt in Phila.)

We should have another term in place of 'peak' for what Dazzy Vance "had" or "did in 1924, 1928, and 1930".

This is about the degree of confidence one has in the assessment of a player. Chris Cobb's use of regression in calculation of MLEs reflects a similar interest (in a sophisticated way), as does Clay Davenport's adjustment for season length (in a simplistic and perhaps arbitrary way, with "one size" exponent 2/3 "fits all").

Given a player's statistics in his 46 games played of 52 known, one might guestimate three times that stat line as his achievement expressed in 156-game terms (scheduled for 154, there were two ties) of which he played 138. How much confidence can you have in that assessment? Not much, and regression is a method of moderating the assessment in order to make it more plausible in 156 games. Regress to what? to some average of the player's achievement rates in neighboring seasons.
   25. Mark Shirk (jsch) Posted: January 31, 2006 at 12:26 AM (#1844545)
Paul,

I am somewhat confused about your post. I believe you are saying that we are regressing NeL players to have peaks that look morelike a single mountain than a number of sharp mountains because this is the ideal. While I will concede that the pattern is the most likely pattern, I don't know if it is one that has anything to do with value accomplished over a career or at a peak. I must say that peak performance to me is about the quality of the highest mountain but about a player at his best, so the height of all the mountains of a player's career.

Am I right on my interpretation of your post? If not please respond.
   26. Mark Donelson Posted: January 31, 2006 at 12:37 AM (#1844556)
I agree with Dizzypaco thruout this thread except that I think he concedes too much when he says "I agree with you" or something similar.

Fair enough. Me, I'm not interested in the average of the player's achievement rates. As an extreme peak voter, I'm interested in whether he was dominant for a chunk of his career, consecutive or not. If that's one year, a la Brady Anderson, that's not enough. But three or four years of dominance works for me--even if it's not consecutive, that's three or four years when the player was a great, great help to his team.

So while I understand why regression makes sense when we're trying to take what a player did in a 52-game season and compare it to what other players did in 154-game or 162-games seasons, that to me is a different exercise, and the regression is only used by necessity. That is, comparing the 52-game season just by multiplying would be akin to taking Mike Bordick's April, in the example mentioned above--the sample size is too small. A full season, OTOH, is not, IMO, too small.

Personally, I find it necessary to adjust the regressed numbers in the NeL WS estimates to recreate some kind of peak. It's a highly artificial process, but then again so is smoothing out peaks and valleys. It's all kind of ugly, but there are no clean methods of comparing NeL to ML numbers, as far as I can tell. I'm frankly thrilled to have found any way of doing it at all.

In a way, this becomes a peak vs. career argument, in a sense; it's no coincidence that you have two peak-heavy voters, jschmeagol and yours truly, on this side of things.
   27. sunnyday2 Posted: January 31, 2006 at 02:52 AM (#1844684)
OTOH, Mark, would you agree that all else being equal, 4 consecutive peak years are better than 4 non-consecutive peak years?

I think yes. If I'm the GM and I sign up Joe Ballgame for my team and he is my stud (my future HoMer) and I know that the next 4 years he is going to kick ass (as opposed to kicking ass in years 1, 3, 6 and 8 or some other unknown variation), then I can assemble a roster to really go after a pennant. I mean, assuming I'm not Brian Cashman and I can't bring in fresh meat every year.
   28. Mark Shirk (jsch) Posted: January 31, 2006 at 03:21 AM (#1844710)
Sunny,

I think it is more valuable to a GM but I don't think it changes the value that said player gave his team(s) over his career. So in a HOM context I must say that I disagree. It isn't that putting the yeras consecutively hurts a plyer just that it isn't something that I think should effect his candidacy in any way.
   29. Howie Menckel Posted: January 31, 2006 at 03:41 AM (#1844741)
I agree with Schmeagol.

It reminds me of the Rube Waddell debate. Wow, he's so dominant because he strikes tons of people out.

If I'm a GM, absolutely I prefer the guy who strikes people out, as it's predictive of more success, generally, than a lesser-K guy.

But the HOM is totally different. We're judging what they actually accomplished. The games are all over. Did he earn a spot, or didn't he?

I don't much care if the quality years are consecutive, if we're talking about two candidates who each had four MVP-type seasons and 10 good seasons, for instance.
   30. Mark Donelson Posted: January 31, 2006 at 07:09 AM (#1844906)
I also don't think the postulated GM ever really knows that the guy's gonna do it again next year. I mean, Don Mattingly looked like he was going to keep having great seasons forever until one year he hurt his back and didn't anymore.

That said, I do think the consecutive years are slightly more valuable, HOM-wise, than nonconsecutive ones that are precisely equal. That's what I meant about them being a tiebreaker. But that doesn't come up too often.
   31. Chris Cobb Posted: January 31, 2006 at 03:57 PM (#1845127)
It's good for us to be debating the ways in which we evaluate merit.

I'm curious why voters want to "a career voter" or "an extreme peak voter" or "a consecutive peak voter" when career, peak, consecutive peak, prime rate, and so on all have serious justifications as standards of merit.

Because all these sorts of measurements are justified, I include elements of career, non-consecutive peak, and consecutive peak via a peak rate, weighted about equally.
   32. sunnyday2 Posted: January 31, 2006 at 04:12 PM (#1845135)
Believe it or not--I mean, everybody knows I'm a peak/prime voter--but I weight all three equally, too. It's just that when you do that the peak is factored in 3 times. As the peak, as a part of the prime, and as a part of the career. So I'm not actually clear how I would actually weight all 3 equally.
   33. sunnyday2 Posted: January 31, 2006 at 04:15 PM (#1845138)
>It's good for us to be debating the ways in which we evaluate merit.

Yes, there's been a really good discussion the past week or two.
   34. karlmagnus Posted: January 31, 2006 at 04:32 PM (#1845151)
Sunnyday2, it depends on the length of peak, prime and career, since a 2 year peak is 1/9 of an 18 year career. However, if you weight them 1/6 peak, 2/6 prime and 3/6 career, and the peak/prime/career lengths are 2/6/18 then you get

Peak: 1/6 +1/3x2/6+1/9x3/6=1/3

Prime 2/3X2/6+1/3x3/6=1/3

Career 2/3x3/6=1/3

That's probably utterly useless for your purpose, but you see what I'm getting at!
   35. Dr. Chaleeko Posted: January 31, 2006 at 05:11 PM (#1845214)
I measure 3 yr, 5 yr, 10yr, 15yr, and (when appropos) anything else.

But like Sunnyday, I obviously repeat those first three years in each increment. Theoretcially I guess the minimum result of that is to put around 30% of all my decision making in the hands of those first three big peak years, assuming I also end up going out to 20yrs.

At 15 years, much more common, I'm putting about 35% of the decision making in the hands of the 3 biggest years.

Another way to look at it is to take the percentage that those three years play in each of the four/five increments I look at.

So 3yr = 1.00; 5yr .67; 10yr .3; 15 yr .2; 20yr .15

Sum is 2.17 / 4 = 54% peak emphasis at 15 years or 46% at 20.

I'm not sure which is the better way to measure it, but I know that I'm putting 30-50% of my emphasis on peak (depending on the candidate).

I think, however, that the "real" peak emphasis actually varies by candidate and is based on how many years they played and how much their biggest three years contribute to their total value.
   36. Mark Donelson Posted: January 31, 2006 at 05:42 PM (#1845268)
I'm curious why voters want to "a career voter" or "an extreme peak voter" or "a consecutive peak voter" when career, peak, consecutive peak, prime rate, and so on all have serious justifications as standards of merit.

I may be the only person on the peak side who really doesn't take career much into account, so I'll answer this.

I agree that each of those things do have justifications--that is, I understand the arguments for them. I just disagree with those arguments, personally. The largest part of merit, to my mind, is dominance. To me, the best players of all time are the ones who dominated the opposition most. How long they were able to do this is the difference between no-brainers like Wagner and closer calls like Jennings. And guys like Beckley just don't interest me, because IMO they never dominated.

Prime (unless it's really an extended peak) comes into play quite often for me, but largely as a tiebreaking factor between players with similar peaks. Career is more of a last-resort tiebreaker, and rarely enters into my calculations. I've already addressed consecutive peak.

I do make adjustments after that (particularly to even out representation by position, which I feel is important), of course. But that's the general theory.
   37. sunnyday2 Posted: January 31, 2006 at 05:48 PM (#1845282)
karl, I don't disagree though my method is more simple than that. At it's simplest it's nothing more than:

Player A has a higher peak than Player B
Player B has a longer prime with better shoulders
Player A has more total career value

Ergo, by a margin of 2 to 1, Player A ranks ahead of Player B on my ballot.

Now, before anybody goes bashing the gross simplicity (or simple-mindedness) of this method, this is sort of a model of the method, not the entire method itself. I mean, obviously in evaluating any of these elements--peak, prime, career--there are many qualitative or quantitative factors to consider. And there are cases where the differential between two players at their prime is so big and the differentials on peak and career so narrow that I might select Player B. (Eddie Cicotte and Eppa Rixey are Type B's for example.)

But I agree that a perfect 1/3, 1/3, 1/3 weight is not always possible.
   38. KJOK Posted: January 31, 2006 at 07:21 PM (#1845443)
There's actually a pretty good article trying to tackle exactly this subject:

BP - Hall of Fame - Belle, Dawson, etc.
   39. Joey Numbaz (Scruff) Posted: February 01, 2006 at 08:27 AM (#1846233)
That's actually The Hardball Times, not BP :-)

It's an interesting system, but I'm not a big fan of the rank column.

First OBP*SLG*PA is not a great method. It's basically the basic version of runs created.

He's ranking them by annual rate, but he treats a 14 year prime the same as a 9-year prime. He needs to take the next step and multiply rate by years if he's truly looking to measure the prime of the player.

Also, looking at Dawson, who he is entirely too tough on, it's a little like Beckley. All of the little things conspire against him.

*Dawson has 4 extra years where he wasn't prime (should be 5, unless a 106 OPS+ from a CF isn't even a decent season), but was a useful player - that's more than anyone on the list except for Evans and Williams, and a part of his argument for being a Hall of Famer. The prime was close and he had a long career.

*Dawson as 20/26 in a typical prime-season as a base-stealer, which is better than just about everyone listed - but this doesn't make it into his rate stat.

*He lists Dawson as a "good" RF. Dawson won 8 Gold Gloves and played CF for the first 7 years of his career. Yet he's on a list where he's mostly compared to corner OF's and 1B.

Adjust for all of these 'little' things, and he moves way up that list, IMO.

My point is whatever your system is, you need to be aware of players that just skim under (or over) several of your threshholds, and adjust - you have to realize you aren't going to come up with the perfect system. There is no magic bullet, a significant portion of this is art, it's not all science.
   40. Mark Shirk (jsch) Posted: February 01, 2006 at 05:34 PM (#1846515)
Of course Dawson also had a career .322 OBP and never had any big years despite playing in CF, a position that WS loves.

He will notmake my top 100 when he becomes eligibile, I would rather have Jim Rice (whom I don't like much either).
   41. Mark Donelson Posted: February 01, 2006 at 06:30 PM (#1846588)
My point is whatever your system is, you need to be aware of players that just skim under (or over) several of your threshholds, and adjust - you have to realize you aren't going to come up with the perfect system. There is no magic bullet, a significant portion of this is art, it's not all science.

I agree emphatically--no matter what your system is, an awful lot of this is about adjustments (or what Marc, borrowing from Bill James, likes to call the "BS Dump").
   42. Dr. Chaleeko Posted: February 01, 2006 at 07:02 PM (#1846628)
...what Marc, borrowing from Bill James, likes to call the "BS Dump").

By the by! I had a HOMmark Moment the other day, while driving South on US 1 from Ogunquit, ME to York. There's a business there called Bullshirt, and their access road has a little street sign called Bullshirt Way. I had to chuckle thinking of the many, many times Sunny'd written Bullshirt Dump in one of his posts.
   43. andrew siegel Posted: February 01, 2006 at 07:25 PM (#1846665)
While we're on the subject of Andre Dawson (who I'm not wild about), I make a comaprison between him and Willard Brown on my ballot. DOes anyone think that is either too kind or too harsh to Mr. Brown?
   44. DanG Posted: February 01, 2006 at 07:51 PM (#1846709)
Of course Dawson also had a career .322 OBP and never had any big years

BBRef has .323 OBP, let's not make it worse than it is.

To be fair, Dawson had his peak year in 1981. He earned 25 win shares in two-thirds of a season, which equates to 35 in 162 games (extrapolating from his established level of performance). This gives him a five year peak 1979-83 of 142 WS.
   45. sunnyday2 Posted: February 01, 2006 at 10:42 PM (#1846944)
Bullshirt
   46. Joey Numbaz (Scruff) Posted: February 02, 2006 at 03:43 AM (#1847248)
"Of course Dawson also had a career .322 OBP and never had any big years despite playing in CF, a position that WS loves."

You are joking right? From 1980-83 he was one of the best handful of players in the NL, with two runner-ups in the MVP voting. As Dan says, give him 1.5x on his WS for 1981 for starters. And he had a 15-year extended prime.
   47. Joey Numbaz (Scruff) Posted: February 02, 2006 at 03:51 AM (#1847251)
Also, the OBP is hurt by his playing his entire prime in a park with awful visibility. Olympic Stadium just killed him for some reason, worse than his teammates.

He's a lot like Beckley, only with bigger years (1980-83, 88 and 90), and playing a key defensive position well during his big years. Neither was an OBP fiend, but both were among the best power hitters of their generation. Dawson hit 438 HR back when that was a big deal. He's got the #6 power-speed number of all time.

The OBP is a major negative. Everything else is a major positive.
   48. Joey Numbaz (Scruff) Posted: February 02, 2006 at 03:56 AM (#1847256)
Dawson has grey ink, 164 points. He was top 10 in his league in HR 9 times. He's got 2774 career hits, and 438 HR. Is there anyone 2500 hits, 400 HR and a bunch of gold gloves that isn't in the Hall of Fame?

I can't see summarily dismissing him because of a below average OBP without looking at the other things he did.
   49. sunnyday2 Posted: February 02, 2006 at 05:10 AM (#1847308)
>I can't see summarily dismissing him because of a below average OBP without looking at the other things he did.

Absolutely positively right.
   50. Chris Cobb Posted: February 02, 2006 at 05:13 AM (#1847312)
While we're on the subject of Andre Dawson (who I'm not wild about), I make a comaprison between him and Willard Brown on my ballot. DOes anyone think that is either too kind or too harsh to Mr. Brown?

It's a decent comparison, I think. Brown looks to have been a better power hitter, but his documented plate discipline is even lower than Dawson's. Personalities were totally different, though, and Brown didn't have bad knees. So Dawson doesn't give an image of Brown, but their skill sets seem to have been similar. Talent similar? Maybe.
   51. Joey Numbaz (Scruff) Posted: February 02, 2006 at 05:20 AM (#1847317)
Not to beat a dead horse, but I see Dawson as basically even with Darrell Evans or Dwight Evans, two players who many statheads think should be Hall of Famers, and were jobbed by the BBWAA.

I think they latch on to the other two because the BBWAA dropped them so easily, and knock Dawson because of the OBP. But really all three had similar career value. Dawson and Dwight had a higher peak than Darrell. Dewey was a better hitter than Dawson at their peaks, but Dawson played the tougher defensive position and played it very, very well, so I'd say Dawson 1980-83 is a wash with Dewey 1981-84.

Everyone has their innate biases, statheads aren't immune. I don't see how anyone that champions the Hall of Fame cases of the Evanses cannot support Dawson as well.
   52. sunnyday2 Posted: February 02, 2006 at 05:25 AM (#1847319)
I see Dawson as probably the more valuable player, specifically considering his defense. Both had long careers of good power hitting. Brown hit for a better BA and OBA despite the lack of walks but of course you have to consider the competition. Brown is on my ballot now and I expect Dawson to be there, too. In his prime he was a stud, walks or not.
   53. Mark Shirk (jsch) Posted: February 02, 2006 at 05:31 AM (#1847322)
Wow wasn't aware that you guys were such fans of Dawson, he is a player that I really don't think was anything special.

Top 5 WS

32,29,28,26,24

WS>15...97

WS>25...15

Th closeest OFer I have to this is Kiki Cuyler, though Dawson does have more career. Personally I would rather have Rockey Colavito.
   54. sunnyday2 Posted: February 02, 2006 at 05:53 AM (#1847328)
Colavito in his prime was an all-star...a 2nd team all-star, about the 4th or 5th best OF in the AL. Damn good, underrated by history to be sure.

Dawson was a 1st string all-star, about the best or 2nd or maybe 3rd best OF in the NL, a league that was just as good if not better than the AL of the 1960s.

19. Dawson 32-29-28-26-24/132/340 is nuffiin' to sneeze at
26. Colavito 33-32-29/133/273 very comparable except that little thing of about 70 career WS

What is also incredibly misleading about this is they are both listed as RFers. Colavito played 1285 games in RF, and when he wasn't in RF he was in LF--524 games. Zero games in CF. Dawson played 1284 games in RF, 39 in LF...oh, and 1027 games in CF. Dawson a B glove, Colavito a C.

It's true that Colavito's OPS+ is 132 vs. Dawson's 119, and his OBA was .359 vs. Dawson's .323. But somehow or other Dawson manages to slug .482 vs. Colavito at .489 and actually hit 60 more HR, 220 more 2B and 70 more 3B. 350 more XBH. Colavito an XBH every 9.6 ABs, Dawson one every 9.8. Fairly comparable hitters but Dawson has 17 seasons of 100 games, Colavito 12, and Dawson has a huge advantage with the glove. Colavito had outstanding in-season durability, however, 7 seasons of 150 games or more, but Dawson had 6, and Dawson has the edge 10-9 in seasons of 140 games or more.

It's not Dawson's fault they gave him the MVP that year. It shouldn't be held against him.
   55. Mark Shirk (jsch) Posted: February 02, 2006 at 06:15 AM (#1847342)
In theory, WS should have already adjusted for position and defense and if there is a position that it seems to overrate it is CF, not RF or LF. Not sure why listing what position they played is supposed to mean unless you are saying that Dawson should be ranked lower amongst RFers because WS overrates CFers and a good number of his best seasons were in CF.

Best five seasons

Dawson 32,29,28,26,24
Colavito 33,33,30,28,26

3/5/7's

Dawson 89/139/182
Colavito 96/150/193

WS>15
Dawson 97
Colavito 96

WS>25
Dawson 17
Colavito 25

Career
Dawson 340
Colavito 278

Colavito has an edge in everything except career and WS>15 (my version of WSAA), in which they are essentially tied. So do you value 62 career WS or Colavito's peak and prime? I say Colavito.

And why would anyone think I am holding the 1987 MVP against Dawson? I was 6 at the time and a huge fan of the Hawk.
   56. Mark Shirk (jsch) Posted: February 02, 2006 at 06:17 AM (#1847345)
And it isn't like I have Colavito in my top 75 or anything. But the time Dawson becomes eligible he won't be in my top 100. He is kinda like Jake Beckley, though he has a higher peak and a slightly better prime.
   57. Paul Wendt Posted: February 02, 2006 at 07:20 AM (#1847380)
Of course Dawson also had a career .322 OBP and never had any big years despite playing in CF, a position that WS loves.

Dawson's best season was 1981.
Same for Grich, Evans, Schmidt.
and Valenzuela and Fingers.

Peaksters take note.

WS should have already adjusted for position and defense and if there is a position that it seems to overrate it is CF, not RF or LF. Not sure why listing what position they played is supposed to mean unless you are saying that Dawson should be ranked lower amongst RFers because WS overrates CFers and a good number of his best seasons were in CF.

Contrary to or despite that Win Shares seems to overrate CFs: among players with given, strong WS credentials, there is no surplus of clear career CFs above the number of clear LFs or clear RFs. Years ago, I looked at Fielder Jones, who is among the Bill James RFs by mistake. I judged that WS credentials for RF #50 are about equal to WS credentials for CF #40.
   58. Paul Wendt Posted: February 02, 2006 at 07:24 AM (#1847383)
By the way, does anyone here know of Gil Hodges?
;-)
   59. Dr. Chaleeko Posted: February 02, 2006 at 02:50 PM (#1847596)
Agreed with Joe about Dwight Evans. He and Dawson are about the same strata, right next to one another, just outside my personal "tolerable mistake" zone (that is not HOMers and just missing the top 25 as I see them--actually I have Evans at 25, Dawson at 27).

But I would disagree concerning Darrell Evans. He's much higher on my third base totem pole (~15th) than Dawson on the RF totem pole (or CF if you prefer). RF is an absolutely stacked position, and you can't take all 3000 great right fielders.

I think positional balance matters, and in Darell's case, it's not like he's borderliner; he's solidly on my "in" list.

But that's just one voter's opinion.
   60. Joey Numbaz (Scruff) Posted: February 03, 2006 at 05:11 AM (#1848553)
I still consider Dawson a CF, not a RF Dr. Chaleeko. That's where he played during his prime, and he's got over 1000 games there. I think that's where he should be listed, especially since he'd rank higher there than in RF.

I'm curious, do the peak WS numbers above give Dawson (and Evans) credit for their best year being 1981? He needs to get 1.5 credit for that year IMO, in any peak calculation.
   61. Dr. Chaleeko Posted: February 03, 2006 at 02:37 PM (#1848763)
CF, RF, it's arguable. 7 seasons in CF with peaking, 9 in RF with decline. 6 of one, half dozen of the other to me.

As a RF, the Hawk ranks 27th all time in my system. As a CF, he would rank very similarly. Probably 26th, right behind Ashburn, right before Carey. Maybe 25th. This ranking is based on WS blown out for schedule lengths, strikes, wars, everything. It's not substantial enough of a difference for me to think that listing him in CF makes him more HOMworthy than in RF.
   62. sunnyday2 Posted: February 03, 2006 at 03:26 PM (#1848787)
>It's not substantial enough of a difference for me to think that listing him in CF makes him more HOMworthy than in RF.

Strictly in terms of value I'm sure this is true. I mean, his defensive contributions in CF are in the numbers.

But let's be honest, we all vote also in part based on a mental image of a player and simply thinking "Dawson, RF" is a bit narrow as a mental image of the Hawk.

It's like somebody once argued, Pete Browning rates ahead of Dawson because Browning was a CF and Dawson a RF. That would be a somewhat delusional argument. (It was not Doc who made that argument, it is just a e.g.)
   63. Dr. Chaleeko Posted: February 03, 2006 at 04:04 PM (#1848841)
Right and my point is more that the difference between Dawson's relative rankings in CF and RF don't suggest to me that there's compelling evidence to revise my opinion of him upward.

Or to look at it another way, CF is almost as stacked as RF, but it's with different kinds of players. The nice thing about WS is that it allows us to compare genres of players on equal footing, which is why Dawson's 438 HR come in a little behind Ashburn's walks and flycatching for me.
   64. DavidFoss Posted: February 03, 2006 at 04:50 PM (#1848905)
Or to look at it another way, CF is almost as stacked as RF

I don't know. Since the retirement of Willie Mays, CF has become much less of a 'hitter' position. Other than Griffey, who else is there? Those hitting-by-position spectrum charts that are posted periodically show the same thing. This should give a higher boost to guys like BeWilliams, Puckett, Dawson & Dale Murphy than was given to borderline candidates from a previous era like Berger & HWilson.
   65. Dr. Chaleeko Posted: February 03, 2006 at 04:57 PM (#1848917)
David, that would be true if RF had been a big-value position all along. But how many great 19th century RFs are there? King Kelly is like half a RF and Keeler played 1/2-1/2 in the 90s and 00s. Likewise Flick and Crawford played more in the 00s than 90s. There's Tiernan of course, but he's not up in the range of the All-Time Greats.

And there's plenty of CFs from the 1800s. Gore, Hines, Duffy, Ryan, Hamilton. GVH too. Griffin, Hoy, Pike as well.
   66. Dr. Chaleeko Posted: February 03, 2006 at 05:05 PM (#1848927)
Just a thought...

In a way it makes sense why CF would become less of a hitter's position. In our now ten-year era where everyone hits homers, there's more flyballs and line drives in the air since everyone is gunning for the fences.

If ballparks are smaller, which they are said to be, the opportunity for a good defensive CF to cut down the number of line-drive hits may be higher, by dint of his closer positioning, than during a time of bigger parks. Playing a ballhawk who can turn more singles and doubles into outs would help reduce players on base that can score by the ominpresent home run.

Just a postulate, though no evidence to support it.
   67. Mark Shirk (jsch) Posted: February 03, 2006 at 05:41 PM (#1848967)
My major problem with Dawson is that even when he was at his fielding and hitting peak (early 80's/late 70's) he still doesn't seem to have been as dominant as many remember him as. He just didn't have a really nice peak, especially when compared to other CFers AND RFers. It looks like Dawson was a player who looked better than he actually was. Kinda like Robert Clemente lite (Clemente was obviously better than Dawson but he wasn't Robinson/Aaron/Mays like many claimed).

To me it doesn't really matter whether he is classified as a CFer or a RFer (I have him in my CF database). The positions are close enough that there aren't major WS or WARP swings between them. I am beginning to think that the CF advantage that WS has is limited to the 19th and early 20th centuries when CF maybe wasn't as important defensviely as it is today, so those guys were getting an artifical boost in defense.
   68. DavidFoss Posted: February 03, 2006 at 06:15 PM (#1849016)
I am beginning to think that the CF advantage that WS has is limited to the 19th and early 20th centuries when CF maybe wasn't as important defensviely as it is today, so those guys were getting an artifical boost in defense.

Makes sense. Except that it appears that the 1980s guys *should* be getting that boost. Hardly seems fair to penalize the 1980s because of some unfairness in the rankings of the 1890s guys.

This underscores my dislike for cross-positional comparisons using WS & WARP. I try not use those tools for that purpose (of course, that makes it tough for me to analyze multi-positional guys).
   69. Mark Shirk (jsch) Posted: February 03, 2006 at 06:40 PM (#1849075)
Well you could give the 1980's a boost or simply give some demerits (or whatever you want to call it) to the guys who did get a boost (Jones, Duffy, GVH, Thomas, maybe Browning). It seems to me that this would be far simpler.
   70. Chris Cobb Posted: February 03, 2006 at 08:00 PM (#1849197)
But let's be honest, we all vote also in part based on a mental image of a player and simply thinking "Dawson, RF" is a bit narrow as a mental image of the Hawk.

In order to treat early and modern players consistently, it might be better not to vote this way?
   71. sunnyday2 Posted: February 03, 2006 at 09:00 PM (#1849284)
>In order to treat early and modern players consistently, it might be better not to vote this way?

Probably.
   72. DL from MN Posted: February 23, 2006 at 05:41 PM (#1872500)
Reading Lords of the Realm I noticed the anecdote about Hodges paying Jackie to take the discretionary popups because he was afraid of dropping one.
   73. Howie Menckel Posted: February 26, 2006 at 03:56 PM (#1875697)
I just bought my dad the book "Praying for Gil Hodges," the memoir of the 1955 World Series written by Tom Oliphant, for his 88th birthday.
I won't tell him that Gil struck out in our HOM voting.....
   74. DL from MN Posted: February 26, 2006 at 11:24 PM (#1875965)
I've got Gil hovering around #30 on my list.

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
Harveys Wallbangers
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Syndicate

Page rendered in 0.8816 seconds
49 querie(s) executed