Thanks for posting these informative lists. Bus Clarkson should be mentioned, and I think John Murphy may have been his best friend. Toby Harrah had an interesting career too, while Davy Force is an interesting 1870s candidate.

If Devin is doing a series of positional rankings, it's possible he's including Clarkson in the third baseman listing. Clarkson played significant time at 2B, SS and 3B but wasn't renowned defensively at any of them so many of us consider him at third baseman at the MLB level rather than a SS.

Yeah, Clarkson's a 3B. Sorry I've slacked off, I should get 3B up today.

OK, I want to address the issue of the backlog shortstop candidates for the 2010 election systematically. The basic problem is the following: there are five guys (Campaneris, Concepción, Pesky, Rizzuto, and Bancroft) who are splitting the vote. Depending on your weighting of peak vs. career, era concerns, trust in defensive stats, etc., you could come up with virtually any ranking of the five that would be defensible. So this brings up the following questions:

1. Where should the HoM's cutoff line for shortstops be--above this group (as it currently stands), below it, or in the middle?

2. If it is the second option, are we really prepared to have 30 shortstops in the HoM?

3. If it is the third option, can we come up with some sort of a consensus on the ordinal ranking of the five, so we know where to draw the line?

I should make very clear to those who use my WARP that if it "feels" like my system overrates shortstops, there's a very clear reason for it. In his study of Freely Available Talent, the estimable Nate Silver found that shortstop was the only position where replacement players (defined as players over age 27 making less than twice the league minimum salary) were significantly below league average with the glove as well as with the bat (to the tune of 5.5 runs per 162 games), while replacement first basemen were somewhat *above* league average with the glove. This basically gives SS a half-win "head start" on all the other positions.

This is not a universally accepted finding, or sabermetric "settled law" to borrow a term from John Roberts that I'm fond of--Tangotiger, for one, categorically rejects it, and places SS at just 0.5 wins below the 2B/3B/CF level, whereas I have it as more like 0.8 today. So it might be a worthy exercise to subtract, say, 0.3 WARP2 per 1.00 SFrac from all SS, and add 0.1 WARP2 per 1.00 SFrac to all 2B/3B/CF, to see how your ballots would look without this contentious variable. Just to look at Campaneris, for example, it would drop him from $167M in my salary estimator, which is a clear HoM'er, to $151M, which is rrright on the borderline and probably off my ballot. So this is something important for those who rely on my WARP and therefore have 4-5 SS on their ballots to bear in mind.

That said, if we do embrace Nate's finding and think SS should as a result be "overrepresented" in the HoM, we need to determine which ones we want to push first--otherwise, they will keep splitting the vote. I suppose I have a full year to pry into the intricacies of the new fielding and baserunning stats and see what conclusions this data might provide us with. But it would be good to hear from others about how and why they are choosing Rizzuto over Bancroft, or vice versa, etc.

Dan, do you have a link to where Tango rejects this? I'd like to see what he says . . .

Rereading the article (I'd read it before) it seems extremely sound to me. I'm very curious as to where Tango objects.

The most controversial thing I think I see is this:

Obviously, we’d have a selective sampling issue if we gave full credit to Radoslav’s performance, while forgetting about Miroslav entirely--one of the risks when signing an unknown player is that you may waste a significant number of at bats on him before you figure out that he’s not even qualified to carry Mario Mendoza’s jock.

The chosen solution was to “max” everyone’s playing time at the rookie minimum of 130 PA. So, even if the player played a full season, his statistics were weighted as though he’d only had 130 PA. If the player had fewer than 130 PA, then his playing time was taken as is.

Nate's a bright guy, so I tend to trust him on this one, but I can also see that case for not approaching it that way and just crediting everyone based on how much time they actually played. I would have liked to have seen the numbers figured either way at least, so we could make our own call on that one.

I don't remember where exactly--he rejects the entire premise that you can measure replacement level by looking at low-wage veterans, and attempts to construct his model of the defensive spectrum by looking what happens to the UZR of position-switchers. Why don't you just email him? He always responds super-fast. It's quite a critical issue for our evaluation of players throughout history, and until a consensus emerges, I think it would probably good to make voters aware of the range of well-founded positions that are out there.

That said, if we do embrace Nate's finding and think SS should as a result be "overrepresented" in the HoM, we need to determine which ones we want to push first--otherwise, they will keep splitting the vote.

I don't have a sabermetric opinion to add to the discussion here, but there are two points from an electoral perspective.

First, if these guys are all worthy and all nearly equal in value, then voters should put them all up there on the ballot. I had four of the listed five on mine. In building consensus, we need to be careful that we are building a consensus about merit and not agreeing to strategic voting to solve a perceived problem.

Second, whether or not shortstops should be "overrepresented" in the HoM, right now "glove" infielders ARE "underrepresented." My positional count, which differs slightly from what one would derive from the plaque room, because I count players by half careers, so they can appear at two positions, shows 70 outfielders and only 63 2B-3B-SS in the Hall. This is not a large inequality in representation, but it is an inequality. If we have been slightly undervaluing infielders, then the question becomes, whom should we elect to correct this problem? Given that the greatest athletes play at shortstop, it seems reasonable to suppose that the missing players are most likely shortstops. We have accepted that there is going to be a dearth of third basemen in the HoM, because the set of players who are great enough to be HoMers, good enough defensively to play third base, but not good enough defensively to play shortstop, is smaller than the groups at other positions. If there is going to be a dearth of third basemen, then there should be a corresponding excess of shortstops.

In sum, the current roster of HoMers is not inconsistent with the claim that these five shortstops--Campaneris, Concepcion, Rizzuto, Pesky, and Bancroft--are all among the top candidates and quite probably deserving HoMers who have been overlooked because the electorate has, ever so slightly, undervalued defense relative to offense.

Given that the greatest athletes play at shortstop, it seems reasonable to suppose that the missing players are most likely shortstops.

I really hope we don't go this route regarding the HoM. Yes, each position obviously doesn't need to have equal representation, but the urge to give more spots to a position with more enshrinees than most rubs me the wrong way (especially when Bob Elliott, Bus Clarkson and Pie Traynor are still on the outside looking in). Besides, shortstops may be the best athletes, but they don't do as well historically on the extremely important strength/power meter. IOW, athleticism definitely has its place in baseball, but let's not weight it more than it should be.

Conversely, I think it is wrong to enshrine less valuable players simply because of the position they played. I'm not going to put Pie Traynor, who is nowhere close to being as valuable a player as contemporary Dave Bancroft, on or near my ballot just because he played third base and Bancroft played shortstop. I am going to put Bancroft near my ballot because he was only a little behind Traynor with the stick and he had a lot more defensive value.

Catcher is an exception to this, because the demands of the position itself put a lower ceiling on player's value, because it reduces playing time and weakens hitting.

I am not arguing that we should enshrine anyone on the basis of "athleticism." I am arguing that, because of superior athleticism, we are more likely to find a few additional infielders whose value is HoM-quality because the better athletes tend to be at shortstop. Athleticism in itself should not be a criterion for election, but position should not be a criterion for election in itself, either. Batting and fielding value are generated in the context of position, but once appropriate adjustments for context are made, that should be the end of position's influence on a player's merits.

Edited to read:

we are more likely to find a few additional infielders whose value is HoM-quality at shortstop because the better athletes tend to be at shortstop.

Conversely, I think it is wrong to enshrine less valuable players simply because of the position they played. I'm not going to put Pie Traynor, who is nowhere close to being as valuable a player as contemporary Dave Bancroft, on or near my ballot just because he played third base and Bancroft played shortstop. I am going to put Bancroft near my ballot because he was only a little behind Traynor with the stick and he had a lot more defensive value.

Except the attrition rate for third basemen, at least up to Mathews' time, was higher than for shortstops, so that's why I can place Traynor higher on my ballot than Bancroft.

Except the attrition rate for third basemen, at least up to Mathews' time, was higher than for shortstops, so that's why I can place Traynor higher on my ballot than Bancroft.

But what is the cause of that attrition rate? Is there any evidence that it was due to the demands of the position?

But what is the cause of that attrition rate? Is there any evidence that it was due to the demands of the position?

What else could it be, Chris?

What else could it be, Chris?

It could be that the players just weren't as good, so they didn't last as long.

Do you have a list of attrition rates by position/year or position/era? Or is this educated speculation?

Yes, the null hypothesis of random fluctuation is quite a powerful one...

I suppose I have a full year to pry into the intricacies of the new ... baserunning stats

It appears that the new-and-improved baseball-reference (unveiling in the spring?) is going to have a lot of baserunning stats, presumably back as far as Retrosheet goes. I'm not sure if any of it is new information that hasn't already been calculated by Dan Fox or whoever, or if it's just making that information more accessible to the general public for the first time.

It could be that the players just weren't as good, so they didn't last as long.

An above-average third baseman should be able to play the same amount of career games as an above-average shortstop, whatever the "average" is for each position.

Now, what may hurt third basemen other than injuries is the defensive spectrum. While shortstops can find a home at second or third as they age, pre-Mathews hot corner guys may have been in a quandary as they became older (especially when second basemen equaled and then passed third basemen on the defensive spectrum). Of course, this wouldn't be the fault of the displaced third basemen and should still deserve an adjustment.

Why wouldn't it be his fault? If his skills deteriorated to the point that they couldn't find a spot for him on the field, doesn't that make him less valuable and not as good as the player who does have those skills?

You see this with 2B too, after it became more of a fielding position than 3B. It's not like catcher where wear and tear makes it impossible to play after a certain point. The 2b lose their fielding ability, and most cannot hit well enough to play anywhere else, so they wash out early. Is that really something we should adjust for?

Why wouldn't it be his fault? If his skills deteriorated to the point that they couldn't find a spot for him on the field, doesn't that make him less valuable and not as good as the player who does have those skills?

Well, it's not his fault that he didn't have a "no offense really needed" position to fall into like shortstops do.

But that's because he didn't have the skills to play SS. He's not as good of a baseball player.

When Rod Carew didn't have the skills to play 2B any more, he was able to move to 1B, extend his career and get 3000 hits. That's because he was good enough to do that.

When Marty Barrett couldn't play 2B anymore, he washed out of the league. It wasn't because 2B careers were cut short through some physical aspect of the game (like catchers), it was because no one needs an 80 OPS+ hitter that can't play 2B, SS or C.

And if a Barrett-type (I'm just using him as an example, I don't know the real specifics behind him) is good enough to play SS, his hitting may have been enough to still allow him to play 2B after he couldn't play SS anymore. That's because he's a better player, it's not because of something that physically shortens careers of 2B like that of catchers.

One way I've expressed the catcher issue is to talk about Ted Simmons and Bob Boone. Given his offensive quality, the career offensive expectations for Simmons were the lowest at his actual position of catcher. Had he come up as, say, a third baseman, he would likely to have had a longer career (more PA) and better offensive rate stats. That's the wear and tear of the position. On the other hand, the career offensive expectations for Bob Boone were probably greatest at catcher. As a catcher, he had a fabulously long career. Had he come up as, say, a third baseman, he might have played for a few years but it would have been easy for a team to push him aside for a prospect with a bigger bat.

But that's because he didn't have the skills to play SS. He's not as good of a baseball player.

He didn't have the defensive skills, but a third baseman should be better offensively by a considerable amount. Yet, the third baseman will be thrown aside, while the shortstop can just move to third.

This is outright batty. We are rewarding players for having defensive skills so poor they can't move down-spectrum as they age? By this logic, should we give Edgar Martinez 30 years of extra credit, because if he had been able to play shortstop, he could have played until age 70? I don't understand at ALL...

Okay, since I'm batty, I'll stay out of the discussion. Finis.

This is a fascinating debate.

I'm kind of leaning Grandma's way on it (though I have a completely different set of missed-out-upon 3Bs, as a peak voter: Williamson and Rosen). But that's mostly since, if we were to elect all five of DanR's SS candidates above and none of the backlog 3Bers--and none seem especially close to election right now, with McGraw finally in, except for the half-3B Leach--the difference in representation between SS and 3B will be more than "a few additional infielders."

Of course, it's not terribly realistic right now, barring a major shakeup, to say all five SS would ever be elected--Campaneris and Bancroft are pretty far down in the backlog. But right now the SS-to-3B ratio in the HOM is 25 to 19. Add Rizzuto, Concepcion, and Pesky on one side and Leach on the other (assuming he's not more a CF than a 3B?), and you're at 28 to 20.

Then we have Larkin coming up, so that's 29 to 20; I suppose he could be balanced out by Ventura, though I'm a little doubtful. I guess in a world in which Pesky gets in, though...so let's make that a wash: 29-21.

Among active players, we have Chipper and Jeter canceling each other out, and A-Rod split down the middle, though if you have to make a call it'll clearly be 3B by the time he's done (it may be already?). It gets hazy after that, obviously--not sure where guys like Tejada will end up, and it's too early to make calls on the Hanley Ramirezes and David Wrights of the world. But it doesn't look as if the gap will be widening massively there, at least not yet.

As the overall numbers get larger, an eight-player difference starts getting less and less egregious, but it does seem a bit large right now, even if you grant (somewhat) that SS is where the good athletes go and stay. I admit that the dearth of 3B in the HOF itself has always bugged me, just on the basic "you do have to field someone at the position every day" level, and I don't love the idea, personally, of the HOM mirroring it. But that's not a strict value-based argument, of course.

And of course, now that McGraw's in, I have two SS and two 3B on my ballot, so I'm not exactly pushing hard against the tide! :)

Nate's study has a selection bias. What happens to SS who become 2B/3B after the age of 27? You can't look at "SS" as if it's a position like QB. SS is more of a "role" in the infield position group of SS/2B/3B. This is the behaviour of players and how teams operate. This issue is even clearer in the OF, with CF and LF/RF. I know it makes life easier for us to treat SS as an immovable position like QB, but that's not reality, and you can't construct a study that presumes so. You can't have a guy who is a lifetime SS, and then moves to 2B at age 26, and suddenly falls out of the pool of SS at age 27 as if he doesn't exist.

***

As for the 2B/SS gap, my comments on my blog are strictly based on contemporaries. When you look at players that play the two positions, whether you look at their "toolsy" numbers, like the Fans' Scouting Report, or their UZR as they move between positions, the gap is around the 0.50 win range. You can make a fair argument that it might be 0.75. But, that's as far as you can take it (for the current crop).

I can't say historically what it should be, since I have yet to study it. But, I will at some point in the coming months, as I get my WOWY better setup for the Retrosheet years.

The more interesting question is the 2B/3B gap (or non-gap).

I'm kind of leaning Grandma's way on it

You shouldn't have said that, Mark. You might have a padded cell with your name on it now. :-)

Tango, well, 0.8 is where I had it for 2005, and given that the gap was shrinking steadily over the 1995-2005 period, I would fully expect it to be in the 0.6-0.7 range for 2008. So we're not far off by any stretch. That said, 0.2-0.3 wins a year makes a significant difference when you are trying to distinguish the 220th-best player of all time from the 250th, which is where we are at in the HoM at this point.

I have 2B and 3B just about equal in the modern game. The big strengthening of 2B took place in the mid-1980's, when it moved from being a clear 2nd to SS to in a clump with 3B and CF, a difference of nearly a full win a year. The difference betwen 1970's and 1990's 2B is just as big as the difference between 3B today and in the deadball era.

You shouldn't have said that, Mark. You might have a padded cell with your name on it now. :-)

Well, the other Marc (sunnyday) should be good company, at least, right? ;)

You shouldn't have said that, Mark. You might have a padded cell with your name on it now. :-)

I think it is worth considering whether voters who do not consider career value as part of their analysis would expect to see a gap between the number of HoM-worthy shortstops in baseball history and the number of HoM-worthy third basemen.

From a career perspective, I agree with the arguments that Joe and Dan have been making in support of the view that third basemen have shorter careers than shortstops not because of the demands of the position but because their lesser fielding ability causes their careers to be shorter, and that they therefore have less merit that players whose greater fielding ability enables them to have longer careers.

However, the peak voter might argue that a player's merit lies in his accomplishments at his best; what happens as he declines, whether it means a shift to a more accommodating fielding position or to retirement oughtn't to concern us. That's a valid position to take, as a normative statement about merit.

So the question: does the peak voter see as many worthy third basemen as shortstops, because it is the longer careers of the shortstops that distinguish the two groups, or doe the advantage of shortstops remain? There tends to be a correlation between the height of a player's peak and the length of his career, but it is not a tremendously strong correlation, and it may hold true at some positions more than others.

Dan, yes it certainly looks like we are close.

***

As for the extra 0.25 wins per season times 16 seasons equals 4 wins, plus/minus, depending how you determine the positional adjustment, that is not really an argument for me.

I'm happy leaving the uncertainty range as +/-0.25 wins per season, since for me, I'm not typically interested in career-wise aggregation. If it comes down to someone being 50 WAR and another guy is either 48 or 52, depending on how the positional adjustment is done, then I don't know that I really want to get into the middle of that, especially if we end up going cross-era.

I think there is enough uncertainty in all the other considerations (wear and tear on catchers, parks not affecting each hitter the same, cross-era, switch hitters making each pitcher be against the platoon more often, the role of running changing, the impact of each event being different in different run environment, and on and on), that I find it easier to just state that up front.

But, if it gets to 0.50 wins each year or more, then I would definitely be interested in that discussion. That uncertainty range is too high for me.

From a career perspective, I agree with the arguments that Joe and Dan have been making in support of the view that third basemen have shorter careers than shortstops not because of the demands of the position but because their lesser fielding ability causes their careers to be shorter,

I'll restart this, because you're always civil, Chris.

Center field is obviously more demanding fielding-wise than left field or right field, yet center fielders tend to have shorter careers. How would you explain that?

Because center fielders don't tend to hit enough to be able to hold down a corner once their defense declines, whereas LF/RF can move to 1B/DH?

And on the issue of tone, while of course I think we should stay away from ad hominem attacks on other voters, I see no reason why I shouldn't be able to say that a particular argument seems loony to me--regardless of whether it comes from a longtime respected voter like John Murphy or a newbie. I think one of the great strengths of the HoM is precisely how fierce the debates can get, while always remaining focused on the actual Merit-related subject under discussion. If you think my WARP approach is total BS--and you've actually studied the methodology enough to explain why--then I would like nothing more than to see you brutally assault the method on these pages. Just so long as, say, my mother is left out of it. :)

Because center fielders don't tend to hit enough to be able to hold down a corner once their defense declines, whereas LF/RF can move to 1B/DH?

I agree. Like third basemen, center fielder have a hard time finding a place to play once there skills start diminish because of their offense, despite playing one of the most demanding positions (I'm not arguing third base is as tough as CF at the moment, BTW.

I just don't see why this means 3B and CF should get extra credit from voters. If their careers are less valuable, then that sucks for them. And obviously, many 3B can go to first (Brett, Molitor, Allen, etc.)

Maybe this will help you see what we're seeing John . . .

Try looking at it this way. Many 3B/2B and CF only get a job in the majors because of their fielding. If they had to hold down a corner they wouldn't have gotten the job in the first place - or if they did, they would have gotten to the majors later.

They would need to be at say 90% of their peak offensive ability as opposed to 70% if they were a better hitter. And they leave earlier, as once they start to decline, they cannot meet the hitting requirements of the new position, nor the fielding requirements of their old position. I can't fathom how that would deserve extra credit. They just aren't good enough players.

Maybe looking at it that way just confuses the issue more, I don't know. I really can't think of another way to explain it.

To take contemporary examples, Coco Crisp and Jacoby Ellsbury both seem likely to be full time CF on decent teams next year, yet with career OPS+ of 94 and 95 (Ellsbury could improve a bit) they wouldn't hold down a position as corner OF and will be out of the league once they're not average plus CF. At present they may be as good as a sluggish LF/RF with an OPS+ of say 110 but they're unlikely to have careers that are as long. So a peak voter would put them equal with the sluggish OF, a career voter would put them below.

Obviously neither HOM quality, but good examples of what is meant, I think.

They would need to be at say 90% of their peak offensive ability as opposed to 70% if they were a better hitter. And they leave earlier, as once they start to decline, they cannot meet the hitting requirements of the new position, nor the fielding requirements of their old position. I can't fathom how that would deserve extra credit. They just aren't good enough players.

I have understood all of this for decades. This is nothing new to me. However, for HoM purposes, no position should have an obvious advantage over the other ones. I'm not talking about a quota - I have voted for more shortstops over the years than third basemen. Hell, I'm not even that crazy about Pie Traynor, for crying out loud. I also agree that there have been more great shortstops, as I have stated this for the past 5 and a half years. All I'm saying is that we should give each position a fair crack at the HoM, that's it. I don't think that is so controversial.

I just don't see why this means 3B and CF should get extra credit from voters. If their careers are less valuable, then that sucks for them. And obviously, many 3B can go to first (Brett, Molitor, Allen, etc.)

If you look at my original post, I specifically mentioned pre-Mathews third basemen, Dan.

And on the issue of tone, while of course I think we should stay away from ad hominem attacks on other voters, I see no reason why I shouldn't be able to say that a particular argument seems loony to me--regardless of whether it comes from a longtime respected voter like John Murphy or a newbie.

You can debate any issue that you want with me. I have no problem with that in the slightest. Over the years, there have been countles debates here regarding positions that I have made. Many of them I won, some made me change my mind, while others left us at a stalemate.

With that said, I am admittedly thin-skinned when it comes to referring to one of my ideas as loony. I will continue to be that way, since I don't respond that way in kind.

If you think my WARP approach is total BS--and you've actually studied the methodology enough to explain why--then I would like nothing more than to see you brutally assault the method on these pages. Just so long as, say, my mother is left out of it. :)

While I have some (not all of it, BTW) disagreements with your approach, I don't think any of it is wacky. If I did, however, I still wouldn't say it, because that's the way I am.

However, for HoM purposes, no position should have an obvious advantage over the other ones.

But we're not enshrining positions, we're enshrining players. The position that they play is a reflection of how their talents mesh with the demands of the game. Players who can play shortstop at the major league level, and who can hit pretty well have an advantage over players who can play third base but not shortstop and can hit pretty well. It's not the position that gives the player an advantage, it's his ability, which is what enables him to play his position. In 90% or more of cases, players are used at the position that will maximize their value, and so the value that they earn at those positions is a pretty accurate reflection of their merit. When they are no longer valuable enough to a team to play at that position, they either move to a new position where they still have value, or they're done. A player who can succeed at more than one spot on the defensive spectrum has an edge in ability and value than a player who can't.

That's only saying over again what I've already said before, so maybe I should stop saying it at this point, but there it is.

I'll re-ask my earlier question, though: does looking at players from a peak-only perspective show any differences in the number of HoM-worthy third basemen and shortstops?

Well, pre-Mathews third baseman are properly credited in my system, because replacement third basemen were further below average in those days than they were later...no? Unless you're suggesting they should be *double*-credited for a low replacement level *and* having short careers...which is I guess the leap you need to take to get the likes of Traynor on your ballot.

Serious questions:

Suppose you were ranking a SS from the same time period as Traynor, who had Traynor's offensive numbers.

1. If he played average defense, where would you rank him, in the pantheon of SS?
2. If that ranking is comfortably within the realm of HoM consideration, how bad would his defense have to get before you'd take him out of the realm of HoM consideration?

-- MWE

Suppose you were ranking a SS from the same time period as Traynor, who had Traynor's offensive numbers.

What shall we call this hypothetical player: Joe Sewell, perhaps? (I do like Sewell's offensive numbers better than Traynor's. But there's not exactly a ocean of space between them.)

What shall we call this hypothetical player: Joe Sewell, perhaps? (I do like Sewell's offensive numbers better than Traynor's. But there's not exactly a ocean of space between them.)

Not a bad comp. He ranks at the bottom of the HoM list, I see.

Had Traynor come up with any other team, his career progression would have been almost exactly like Sewell's. Traynor was a shortstop, and by the accounts a decent one with the glove (although he didn't play well in the field there in his 1920 cup of coffee). The Pirates didn't need him at shortstop because they had Rabbit Maranville there, and later Glenn Wright, both of whom were outstanding defensively; that's why he played 3B. By the time the Pirates actually needed a shortstop after Wright got hurt, Traynor was entrenched at third and pushing 30; at that stage he wasn't getting moved.

If you treat 3B as an entirely separate position than SS, and compare Traynor *only* to other 3Bs, then you are in effect penalizing him because of circumstances. Treating 3B as a role within the 2B/3B/SS unit, as Tango suggests, makes a lot more sense until about the late 1940s/early 1950s, when the Eddie Mathews model became more of the norm for 3B. Most 3Bs until then could, and sometimes did, play SS; Traynor played in 9 games there as late as 1927.

-- MWE

I prefer the Tango approach to infield positional weights, for probably idiosyncratic reasons. When I studied shortstop defence, I was shocked at how few balls are fielded by shortstops in the hole, and how poor the overall conversion rates are. The great majority of a shortstop's defensive ability is determined by his ability to get to and convert the ball up the middle. The thing is that there are about as many plays of this type for the second baseman and the play is a little harder because of the need to plant and get rid of the ball in a hurry, in addition to getting to it. The DP pivot is, of course, also more challenging for the second baseman. Subjectively, it doesn't add up to that much of a difference. I might add that any system that won't take account of John McDonald in deciding what is replacement level for a shortstop earns a raspberry from me...

As for the ability of a fine defensive third baseman to play short adequately, I believe that this would have been true not only for Traynor, but also for Jimmy Collins, Mike Schmidt (who did play short early in his career), Scott Rolen and Cal Ripken (!). There is a bit of a philosophical conundrum- I suspect that Scott Rolen might have been able to play shortstop as well as Cal Ripken did had he been given the opportunity, but that is, I think, too hypothetical to give credit to.

1. If he played average defense, where would you rank him, in the pantheon of SS?
2. If that ranking is comfortably within the realm of HoM consideration, how bad would his defense have to get before you'd take him out of the realm of HoM consideration?

Using Dan R's WAR, I have run a quick study on this. I took Traynor's BWAA, BRWAA, zeroed out his FWAA, and plugged in SS replacement level for his career, which looks to have generally been around 3 wins/season, and I ran this player through my system.

My system sees 120 points about the HoM in-out line.

Traynor as a very good defensive third baseman scores 106.2. As a career average shortstop, he scores 111.4, a modest improvement, but not enough to turn him into a HoMer. With his career length, hitting and baserunning ability, he would have needed to be somewhat above average defensively as a shortstop to reach the in-out line. Because of the way my system values peak, I'd estimate that he would need to have averaged approximately .3 wins/season above average at shortstop, which would have been 3.6 FWAA for his career. He was 8.3 FWAA at third base.

Dave Bancroft, who is almost perched on the in-out line at 122, is very similar in value to the slightly above average shortstop version of Traynor. He is -6.5 wins to Traynor on hitting and baserunning, but he has 10.3 career FWAA, 6.7 above the 3.6 I posited for the HoMable version of Traynor.

Having looked at this hypothetical example and the shift it produces, I would generalize to say that

1) if Dan's system's replacement level is underrating 3B (for some era or another) by the amount that shows up in the Traynor hypothetical, candidates who might switch from off-ballot to on-ballot would be Tommy Leach and Ron Cey;
2) if Dan's system's replacement level is overrating shortstops (for some era or another) by the amount that shows up in the Traynor hypothetical, then Bancroft and Pesky of the five shorstops under debate would drop away from my ballot. Campaneris, Concepcion, and Rizzuto would probably remain on ballot, but they would move from the top third to the bottom half.

Now we are just getting into the great value vs. ability conundrum, which is simply a question of voter preference. My WARP make no attempt to assess ability; they only determine how many wins a player actually contributed to his team, above the number that a replacement player at the same position would have generated in the same playing time. If you want to credit Traynor, or any other 3B, for what you think is their hypothetical capacity to play SS, be my guest. But it's not what actually happened on the field, so I'm staying out of it.

Mike how many other 3B were 'blocked' in that time. 3B was closer to SS at that time anyway. Dan's system gives quite a bit of credit to 3B just for playing the position. And you would think an average SS would be a great defensive 3B, and Traynor does get 'credit' for that already.

But I have to think something close to the 16 best SS's were actually playing SS. If Traynor was going to be a great defensive SS, and lose that much of his value as a 3B, they would have traded him, for equivalent value they could use.

I can see Chris's idea of bumping him very slightly, almost giving him a tie-breaker or something. But I really can't see going much further. I'm open to being convinced, but not really seeing it.

But we're not enshrining positions, we're enshrining players.

Well, maybe I should say it this way: I feel we're missing something when it comes to pre-Mathews third basemen. If you read accounts of third basemen of that time, they weren't considered second-tier players. That doesn't mean that Traynor is close to being the greatest third baseman, of course (for the record, I have Pie as borderline for the HoM, so I'm not really advocating his enshrinement), but I think we don't know everything about pre-1950 third base like we think we do. Of course, the managers of that era could have been maroons, but...

One disadvantage to 3B in relation to CF is that you can throw lefthanded and play CF. Look over our enshrined 3B - lots of LH bats for a RH throwing position. One reason I have sympathy for Ken Boyer - he was platoon disadvantaged and still made it over the borderline.

above the number that a replacement player at the same position would have generated in the same playing time.

And that's the rub.

Look, many of you give minor league credit to guys like Charlie Keller who were "blocked" at the major league level, right? This is another side of the same coin. It is no less being "blocked" when a player comes up and plays a different position because his team already has his primary position filled than it is being "blocked" to stay in the minor leagues. If you're going to do the one, you really ought to consider doing the other.

Mike how many other 3B were 'blocked' in that time.

Brooklyn had Jimmy Johnston, who initially played OF because the Dodgers had Ivy Olson at SS and Mike Mowrey at 3B, and who later moved to 3B while Olson was still there; he gathered significant PT at SS after Olson left. Joe Dugan came up as a shortstop with the As, but moved to 3B when the As added Chick Galloway (who had one good year), played at both positions with the Red Sox, and then went to 3B permanently when he was traded to the Yankees, who had Everett Scott.

I have to think something close to the 16 best SS's were actually playing SS.

Don't be so sure. You should see some of the other SS of that era (and 3Bs, for that matter).

If Traynor was going to be a great defensive SS, and lose that much of his value as a 3B, they would have traded him, for equivalent value they could use.

First of all, I don't think Traynor was going to be a great defensive SS; he'd have been adequate at best, IMO.

Second, the argument I'm making here is that 2B, 3B and SS were all considered to be equivalent positions in the 20s - and given the amount of position movement that was going on I think it's pretty clear. When Glenn Wright came up in 1924 the Pirates didn't trade Maranville; they shifted him to second, where Johnny Rawlings (another converted SS) had played a year earlier. When Maranville finally was traded, a year later, Rawlings went back to second, but eventually Eddie Moore (yet another guy who was originally a SS) took over the position.

There were no mechanisms until well into the 60s to ensure a well-distributed flow of talent among teams. Players were "blocked" in a number of different ways - and not just in the minor leagues. What I see happening here is that some of you are looking at the pre-WWII era through the lens of today's highly specialized game - you're applying contemporary standards to positions that simply weren't applicable to the game as it was played then - and you do some players an injustice by doing so and unfairly boost some others like Bancroft. My suggestion is that you ought to at least consider the "possibility" of positional credit for players who were playing what you "think" is a lesser position - but which at the time really wasn't.

-- MWE

John,

It is a tough one how much weight to attach to the accounts of the time concerning Traynor. I remember growing up in the late 60s reading repeatedly how Pie Traynor was the greatest third baseman ever and never reading that Eddie Mathews was. The fact that Traynor had a career batting average of .320, I think, may have influenced the perception of how good and important he was defensively, as it does now for someone like Jeter.

Well, to a certain extent, this is reflected in the defensive spectrum of the time. In 1906, I have 2B at 1.5 wins below average per season, 3B at 2.0, and SS at 2.5; in 1922, those figures are 2.0, 2.1, and 2.9. SS is still clearly the most valuable, but that's nothing like, say, 1971, when I have 2B at 2.2 wins below average, 3B at 1.2, and SS at 3.9 (which is, of course, why I vote for Campaneris).

Dan's 356 hits the nail on the head. He adjusts for Mike's issues #354 pretty well I think, which is one of the reasons I like his system so much.

John,

It is a tough one how much weight to attach to the accounts of the time concerning Traynor. I remember growing up in the late 60s reading repeatedly how Pie Traynor was the greatest third baseman ever and never reading that Eddie Mathews was. The fact that Traynor had a career batting average of .320, I think, may have influenced the perception of how good and important he was defensively, as it does now for someone like Jeter.

I grew up in the '70s thinking Pie was the greatest, too. It wasn't until the '80s when Pete Palmer introduced to me the concept of context regarding baseball stats that he fell off my hot corner pedestal.

With that said, Traynor was perceived as a great player among his own peers during the '20s and '30s. How great is the question.

What about the Hall of Merit members from behind the color line, in the so-called Negro Leagues?

Hall of Merit fielding position
LF - Irvin(?)
CF - Hill, Charleston, Torriente, Oms, Stearnes, Bell, Brown (and Doby?)
RF - none

Hall of Merit fielding position
3B - Wilson, Beckwith
SS - Johnson, Lloyd, Moore, Lundy, Wells
2B - Grant

1. Have I missed anyone here?

2. Note the distribution by fielding position, so many center outfielders and so many center infielders. Does it seem "distorted"?

3. Please take this into account re the total numbers of HOM members by fielding position.

4. Irvin and Doby played LF and CF in the major leagues, mainly 3B-SS and 2B in the Negro Leagues. Martin Dihigo, where did he mainly play in the Negro Leagues? He is a pitcher in the Hall of Merit, which may pertain to the broad theme of this note.

350. David Concepcion de la Desviacion Estandar (Dan R) Posted: December 19, 2008 at 07:50 AM (#3033756)
Now we are just getting into the great value vs. ability conundrum, which is simply a question of voter preference. My WARP make no attempt to assess ability; they only determine how many wins a player actually contributed to his team, above the number that a replacement player at the same position would have generated in the same playing time. If you want to credit Traynor, or any other 3B, for what you think is their hypothetical capacity to play SS, be my guest. But it's not what actually happened on the field, so I'm staying out of it.

--
There is a reply inchoate in Tom Tango's comment on Nate Silver's work.
329. Tango Posted: December 18, 2008 at 02:04 PM (#3032755)
Nate's study has a selection bias. What happens to SS who become 2B/3B after the age of 27? You can't look at "SS" as if it's a position like QB. SS is more of a "role" in the infield position group of SS/2B/3B. This is the behaviour of players and how teams operate. This issue is even clearer in the OF, with CF and LF/RF. I know it makes life easier for us to treat SS as an immovable position like QB, but that's not reality, and you can't construct a study that presumes so. You can't have a guy who is a lifetime SS, and then moves to 2B at age 26, and suddenly falls out of the pool of SS at age 27 as if he doesn't exist.

***

As for the 2B/SS gap, my comments on my blog are strictly based on contemporaries. When you look at players that play the two positions, whether you look at their "toolsy" numbers, like the Fans' Scouting Report, or their UZR as they move between positions, the gap is around the 0.50 win range. You can make a fair argument that it might be 0.75. But, that's as far as you can take it (for the current crop).

I can't say historically what it should be, since I have yet to study it. But, I will at some point in the coming months, as I get my WOWY better setup for the Retrosheet years.

I haven't read Silver or Tango so that's all for now.

--
In another forum, I have urged that others consider SS "center infield" between 3b and 2b, by analogy to CF "center outfield" between lf and rf. If the 3b and 2b all appear to be too weak for you to honor, vote for more shortstops.
--another forum, another time period, but same context in another sense whether to fill a shadow Hall of Fame with equal numbers by fielding position.