The big question for this year isn't who is the MMP (I think), it's whether or not the MVP that the BBWAA elected even makes it into the top 10.

Although OPS is a flawed measure, Mantle's month by month OPS is quite amazing. Maris had one great month, 2 excellent months, and 2 very, very good months. Mantle was out of this world every single month with the possible exception of September, posting a Barry Bonds-esque .508 OBP in July, alongside an .854 SLG.

My contribution to the project: Link to the relevant baseball cards

Didn't want to link the '61 baseball cards?

That's a tricky issue. But the 1962 cards have the 1961 statistics on the back, so I figured they're most helpful. Particularly the league leader cards.

Also, the gum from the 1962 cards is a year younger so it tastes better.

Are we separating our votes by league or are we trying to build a single list that incorporates both leagues? Because if it's the latter, there's an issue you know we need to discuss.

Mantle: 48 WS, 11.9 WAR, 10.8 WARP2

Mays: 36 WS, 9.9 WAR, 7.6 WARP2
Aaron: 37 WS, 9.7 WAR, 9.5 WARP2

But most (all?) of these are measurements against the league - against the AL in Mantle's case, against the NL for Mays and Aaron. We all know that the NL was the stronger league. But how much stronger? In particular, is the league strength difference sufficient to put Mays, Aaron, or both ahead of Mantle?

I think we're ranking all players together.

Can we create (or approximate) a neutral context for the AL/NL in 1961? Bill James used what he called a historically "normal context" in the NHBA of 750 runs per team per 162. If you have the book, it's in the Willie Davis comment on pages 740-1.

"Neutral context" is one thing. How much of that can be done simply by using the stat neutralizer feature of bb-ref? But that by itself wouldn't deal with issue I raised above: what do we make of the difference in league strength? Who gets the #1 spot on your ballot?

HOMers

AL
Mickey Mantle AL CF 48 11.9 10.8
Al Kaline AL RF 29 8.3 5.8
Harmon Killebrew AL 1B 27 5.5 4.8
Whitey Ford AL SP 22 3.9

NL
Willie Mays NL CF 36 9.9 7.6
Hank Aaron NL RF 37 9.7 9.5
Ken Boyer NL 3B 28 8.2 6.9
Frank Robinson NL RF 36 8.0 8.6
Eddie Mathews NL 3B 35 7.4 7.7
Warren Spahn NL SP 26 6.2
Don Drysdale NL SP 20 5.9
Roberto Clemente NL RF 27 5.8 5.9
Sandy Koufax NL SP 21 5.3
Bob Gibson NL SP 19 4.9
Ernie Banks NL SS 20 4.5 5.5

It's one single list.

Looking at the Baseball Prospectus DT cards:

Mantle: 9.4 WARP2, 9.4 WARP3
Cash: 9.1 WARP2, 9.1 WARP3
Aaron: 8.0 WARP2, 8.2 WARP3
Boyer: 7.3 WARP2, 7.4 WARP3
Mays: 7.2 WARP2, 7.3 WARP3

I think the league strength adjustment was in reality a bit larger than what they're using, but that's mostly subjective impression on my part. Maybe Aaron and/or Mays should be above Cash; but not Mantle.

Adding to Howie's list: Lesser HOMers playing regular in 1961

AL
Brooks Robinson
Yogi Berra
Minnie Minoso
Jim Bunning
Billy Pierce
Hoyt Wilhelm

NL
Willie McCovey
Stan Musial
Joe Torre
Ron Santo
Billy Williams
Richie Ashburn
Juan Marichal

As was typical in the 1950s and 1960s:

Those AL lists have 9 white guys... and Minoso
Those NL lists have 8 white guys... and Mays, Aaron, FRobinson, Clemente, Gibson, Banks, McCovey, BWilliams, and Marichal

Let me move a post over from another thread. The background: DanG posted a list of the pitchers with the best FIP given a certain minimum number of innings. I took the exact same list and found RA+ equivalent records for them. I'm listing the leagues separately precisely because league strength is an issue.


AL:
Ford: 19-13
Kralick: 17-10
Lary: 18-13
Bunning: 18-12
Mossi: 16-10
Pascual: 16-12
McBride: 16-11
Barber: 16-12
M'boqtte: 15-11
NL:
Spahn: 18-11
O'Toole: 17-11
Purkey: 16-11
McCormick: 16-11
Koufax: 16-12
Drysdale: 16-11
Cardwell: 16-13
Williams: 14-12
J. Jay: 15-13
Friend: 13-13
Burdette: 15-16

The rank in each league is by equivalent Fibonacci Win Points, based on equivalent W-L, calculated before roundoff. Roundoff can do funny things, so in each league you see at least one 16-12 listed as better than at least one 16-11; and that is the correct order.

A couple of notes: one, this was just not a year for outstanding pitching performances. No one really stands out from the crowd. And man, both Los Angeles teams played in terrific hitters parks that year.

Dan then replied to my post by listing several relief pitchers of possible interest; I haven't had time to respond to that yet, but I intend to.

several relief pitchers of possible interest

I'll bring over an expanded version of that list:

Before I even attempt to do RA-based stuff for relief pitchers, I've got to ask about inherited runners. Is there readily available data on inherited runners, inherited runners scored, and some sort of idea of what that should look like for an average pitcher?

I suspect that inherited runners was more important then than it is now.

OCF, here

Let's see if this shows up correctly...

This is my preliminary consideration set. I ranked every player for 1961 in 5 uber stats (bb ref's war, baseball gauge's war, win shares above bench (also calculated by baseball gauge), bp's warp1, and my own cobbled together wins above average system (which uses wOBA for offense and Pete Palmer's FR for defense (feel free to shudder now)). I then added their ranks together (total rank column) and reranked based on the sum (final rank column). I would have liked to use even more uber stats (I figure any one system will have its idiosyncrasies, but when you factor them all together, you get a pretty good consensus), but Fangraphs doesn't calculate WAR for pitchers before 1974 and Dan R's warp doesn't include pitchers at all (at least in the files I found on the HOM yahoo site), so these are the only ones I have access to (well... I could have tossed straight win shares into the mix, but using both that and wsab seemed odd). Anyway, here are the top 40:



I say this is a preliminary consideration set, as it will depend on how many we end up voting for. If it's going to be, say, top 15, then I'll probably limit my analysis to the top 25, but if it's going to be top 30 then a 40 player set is reasonable.

Some thoughts based on my table...

Mickey Mantle is #1 and Norm Cash #2 in all 5 systems. Then it gets more murky and there's a group of players (Colavito, Aaron, Mays, Gentile, Robinson) who are pretty close together, before another gap. Then it's spread out pretty well from there.

Not only is Mantle #1 in all systems, but when looking at the underlying numbers it's not all that close between him and Cash. I really don't see anyway he can't be the MMP this year, unless it's the league strength argument OCF raises above. And I'm skeptical it can make a big enough difference for Mays, Robinson or Aaron (the only NL players I see as being even in the discussion for #1) to make up that much ground.

As stated up thread, this was a bad year for pitching performances. Spahn is highest in the list at #13, and the next is... Juan Pizarro. Heh.

I just noticed my system (WAA) hates Roger Maris. Not sure what's up with that off-hand, but I'll have to take a closer look. He'd be just behind the Colavito to Robinson group with a more representative ranking. Definitely be a candidate to move up my list if I think my system is missing something with him.

@Nate the Neptunian:
Awesome list! My thoughts are very much in line with yours regarding the usage of uber stats. Anyway, thanks for taking the time to post the list... Things like this are a huge help to me.

No problem. I'm glad someone else found it useful.

As stated up thread, this was a bad year for pitching performances. Spahn is highest in the list at #13, and the next is... Juan Pizarro. Heh.

Actually, it's Frank Lary, one ahead of Pizarro. I missed him the first time I scanned down the list below Spahn, most likely because I had no idea who he is. Looking him up on BB Ref, I probably should have. He had a good 6-7 year stretch for the Tigers, from the mid-50s to, well, 1961. This was his last good year. I see he finished 3rd in the CY voting in '61 and 7th in MVP. Eh... I can see the CY placement, particularly since the writers' choice (Ford) is a bit suspect, and was most likely based solely on his record (25-4 that year). Lary has a good argument for being the #2 pitcher this year, and an outside shot at #1 (I don't see him as being above Spahn, but at least one of the uber stats does). But 7th in MVP, in a down year for pitchers? Yeah, I don't buy that. Notably, BB Ref sees his 1961 as being significantly less valuable than his '56 and '58, when he didn't do nearly as well in the MVP.

Another thought that occurred to me, is there anyone outside of MLB who deserves attention this year? I know we haven't decided on eligibility issues, but if we used the same rules as the HOM, then anyone except East Asian players and post-Revolution Cuban players are theoretically eligible. By '61 I imagine most of the older NgL players who didn't get a fair shot at MLB (so they ended up playing in the minors) would have retired. But were there any MxL or US MiL players who put up jaw-dropping years? Given the lower strength of the leagues, I can't imagine anyone matching Mantle, but maybe there's someone who deserves to be down-ballot, depending on how many we end up voting for.

Another thought that occurred to me, is there anyone outside of MLB who deserves attention this year? I know we haven't decided on eligibility issues, but if we used the same rules as the HOM, then anyone except East Asian players and post-Revolution Cuban players are theoretically eligible.

That's actually been addressed in DanG's post linked in the rules thread:

3) Criteria – Value to your MLB team(s) in that season only. If part of the season was spent outside MLB, that value should be considered as well, so long as most of his value was accrued in MLB (or top Negro leagues).
4) In other words, no consideration for players spending the bulk of the year in the minor leagues or Japan. There may be gray areas here regarding identification of certain leagues as “minor”: the AL in 1900, certain Negro leagues, the IA in the years around 1880, maybe the PCL for a time, the Mexican league in the 1940’s, maybe others.
5) Negro Leaguers – Yes. We will have to determine at what point Black ball ceased to be MLB caliber. We may have to allow for great players on barnstorming teams. Maybe credit for winter league play, but that’s a gray area, for sure.
6) War Credit – None. Ditto for injury credit. “Value in that season,” no what-ifs. The one exception might be for collusion credit, a la Raines in 1987.

Hi all - as much fun as this exercise would be, I'm unwilling to commit to the time required; life is too full of other good stuff. I look forward to seeing the results as they are published each year.

Tom

2 - Are there any league quality adjustments made to account for the NL being the stronger league?

WAR is league adjusted, DanR's WARP and Win Shares are not.

3 - Doesn't the expansion in the AL increase the standard deviation of performance?

Only DanR's WARP accounts for standard deviation of performance; the other metrics could be adjusted for standard deviations.

Because the expansion in 1961 only occurred in the AL, voters using uberstats other than DanR WARP should downgrade AL performances to account for the lowered league quality in the AL. bbref WAR's league adjustments are calculated over the entire decade of the 1960s, rather than specifically for 1961--so they do not take into account the expansion-induced drop in quality of play in the AL.

Have we determined how and how many are going to be on the 1961 ballot?

Not yet, discussion is ongoing at Most Meritorious Player: Rules for Ballot.

Because the expansion in 1961 only occurred in the AL, voters using uberstats other than DanR WARP should downgrade AL performances to account for the lowered league quality in the AL. bbref WAR's league adjustments are calculated over the entire decade of the 1960s, rather than specifically for 1961--so they do not take into account the expansion-induced drop in quality of play in the AL.

Also, the AL went to the 162-game schedule that year, but the NL still played 154. So, counting stats (and most WAR/WARP-type stats are essentially counting stats) should probably be adjusted to reflect this as well.

Yes, you have to remember the season length adjustment. The list at the top of the thread is already adjusted.

That's actually been addressed in DanG's post linked in the rules thread:

Sure. But it's still up for discussion in the rules thread, right? Nothing has been carved in stone yet.

The biggest reason I see for excluding the minor leagues is the practical one. It's a lot harder to systematically address minor leaguers than it is major leaguers. The other reason is the modern view that minor league performance "doesn't count", it's for player development rather than for winning games.

And those are decent reasons, but the first one is equally true of the Negro Leagues. And the second hasn't always been true, depending on the time and place. For instance, the MxL league is officially AAA but they're not affiliated with MLB teams and the players are playing to win for their teams. If a player puts up an awesome performance for a team in the MxL league, then I'm sure it counts for the fans of that team. For that matter, were there any independent leagues in Latin America in '61? I see that '60-'61 was the last year of the old professional Cuban Winter League, though it was no longer as independent as it once was (I don't know if it was officially classified as a minor league by this time, but I do know there was an agreement that let some US minor leaguers play on CWL teams). Luis Tiant Jr. was the rookie of the year.

Yes, you have to remember the season length adjustment. The list at the top of the thread is already adjusted.

Yeah, for the record, the systems in my chart haven't been adjusted for season length, unless the system already handles that. (My WAA system scales everyone to 700 PA, a technique I ripped off from Tangotiger, so it's not an issue.) That strikes me as a "what if", so I figure it's up to individual voters to figure out how to handle that (unless the rules end up mandating something).

Sure. But it's still up for discussion in the rules thread, right? Nothing has been carved in stone yet.

Good god, if that's up for discussion, then I think this project is over.

Whitey Ford should be in the discussion in addition to an strong regular season 2 and 0 in the world series with 14 scoreless innings.

I would assume you guys are considering post season.

I would assume you guys are considering post season.

That's one of the questions in DanG's original post, linked in the rules thread:

7) Post Season Play – Should it be included in our evaluations? I’m inclined to say yes, with a caution that it should not be a major factor.

I'm also inclined to say yes.

I think post-season credit should be left up to individual voters. I don't see any reason to mandate it one way or the other. One of the interesting ideas about a project like this is the different ideas people bring to the idea of merit, or value.

I think post-season credit should be left up to individual voters. I don't see any reason to mandate it one way or the other. One of the interesting ideas about a project like this is the different ideas people bring to the idea of merit, or value.

I think that's a discussion to have over in the rules thread.

So an interesting issue with Norm Cash's famous 1961 fluke season.

It looks as if it AND 1962 are BABIP flukes. 370 BABIP in 1961, 215 in 1962, with otherwise very similar peripherals.

If you assume that Cash's true-talent BABIP was somewhere in the 300-310 range during 1961 and 1962, you get two virtually identical seasons - but they're really good seasons, 400/600 type years.

So in the end, I don't think Cash was ever quite as good as indicated by his 1961 seasonal WARP, but had a much higher "true talent" peak than non-HoM types give him credit for. Maybe a smidgen lower than Johnny Mize's peak? And that applies in both 1961 and 1962.

Just as a point of ideology, I think that I won't take BABIP into account. In other words, the fact that a large number of Cash's batted balls fell as hits in 1961 had value and won games - and I won't worry about whether Cash "deserves" the credit for it. In other words, I don't think I'll make MMP a measure of "true talent."

This is clearly a debatable point, and I won't object if another voter treats it differently.

Some players are more variable than others. I don't think representing a player by his mean is tells you much if he shows high variance in performance. It's like reporting the temperature in July in 1961 by smoothing into a 3 year rolling average.

Looking at the consideration list 1961 was not a kind year for the SS and 2B positions. Even 3B is Mathews, Boyer then you are done. And of course C will have a tough time due to playing time issues most years. Overall don't expect many middle infielders and catchers to be getting the MMP; votes yes but not winning the award.

@Nate

I love the uber-uberstat consideration set. One minor thing that leaps out to me is the extreme extent to which your system loathes Maris (95th!). Does Palmer hate his defense THAT much (cursory look at BBRef does show poor range relative to league that year, but 61 is a lot of home runs)?

This leads into a more fundamental struggle I'm having with MMP defense. When it comes to batting (and pitching...I guess...at least for starters...) a season is a significant enough sample size that I'm pretty confident the standard stats capture, if not true talent, at least actual performance. I'm not so sure that's the case for defensive stats. I'm sure I've read Tango and/or MGL arguing that you need 3-ish years of fielding data to get the predictive value of a year of batting stats. Does this suggest one should regress single-year fielding stats to a multi-year average for MMP purpose?

Of course one would have to pay attention to injuries...

One minor thing that leaps out to me is the extreme extent to which your system loathes Maris (95th!). Does Palmer hate his defense THAT much (cursory look at BBRef does show poor range relative to league that year, but 61 is a lot of home runs)?

Yep. My FRAA has him as the worst fielder in the majors that year. He's also hurt by AL RFs hitting very well that year, so he has a large positional adjustment. And while his offense was good, it probably wasn't as good as 61 homers would suggest. Maris's wOBA was .413, which was very good, but was still only 6th in MLB (Cash was the best in MLB with .464.) (I should point out that I calculated these numbers myself, leaving out Reached On Errors, as I don't have easy access to that info, so they're lower than what you'll find on Fangraphs, but since I leave them out for both individual players and league averages, I think it mostly evens out.) AL RFs' wOBA was .363 that year (1B was best in the AL with .366 wOBA) while league wOBA (removing pitchers) was .325.

Here's what my system came up with for the AL players with the best batting performances this year (sraa = stolen (base) runs above average, pr = positional (adjustment) runs... I think the rest of the columns are self-explanatory):



So, you can see it just hates his defense compared to the other sluggers this year. He also has the second biggest adjustment due to a ton of PAs at a very offense heavy position. And while his BRAA is good, he didn't run away from anyone. Those three things combine to hurt him a lot.

Having said all that, pretty much nothing else sees him as that bad of a defender in '61. BB Ref sees him as a very slightly above average fielder (by 1 run). Baseball Gauge's WAR sees his dWAR as -.10249 (215th out of 228 AL position players). Win Shares above Bench sees him as having 0.18909 defensive WSAB (95th out of 228 AL position players). BP has his FRAA1 as -7 and his FRAA2 as -8. Fangraphs' has him as 1 FRAA. None of these have him as a good fielder, and some of them definitely have him as a bad one (particularly BG WAR and BP WARP), but none of them have him as the horrendous fielder my system comes up with.

I certainly wouldn't claim my system is as robust as the others. It's something I pretty much threw together when I started trying to put together a system for this year's HOM ballot, and it's obviously not as tested or as vetted as the publicly available systems. And while I like some of it quite a bit, the fielding aspect is definitely the part I have the least confidence in. So, in the face of the other evidence, when I do my actual rankings I'll probably toss my fielding results for Maris and see where he grades out with average or a little below average defense. I expect he'll move up into the top 10, though I don't think he can pass Aaron, Mays, or Robinson (given the NL's greater strength and that some of the systems aren't accounting for the AL's increased number of games) and probably not Colavito and Gentile. Maybe after that group but before Mathews. I'll dig into it more when we know what the exact ballot length will be.

My FRAA has him as the worst fielder in the majors that year.

Well, that's crazy. He was considered a very good outfielder and he would often replace Mantle in center during his Yankee tenure. (Including a fair amount in 1961, so he shouldn't be straight-up compared to RF.) Obviously, it's possible that in 1961 he didn't play the field as well as he did in other seasons, but I think anyone who watched baseball at the time (I'm not one of them) would find the idea that he was the worst fielder in the majors to be ludicrous.


leaving out Reached On Errors, as I don't have easy access to that info

They're at B-R.

I also think Zop's idea of trying to assess true talent is not what the MMP should be about. It should be performance/value in that season, period. (But that does cause problems with trying to compare Negro Leaguers. If you only have 30 games to judge them by, one of them must have batted .400 and will look like the best player every single season.)

(Again, opinions from a non-participant, so....)

It should be performance/value in that season, period.

But that's what I'm doing - trying to identify who performed the best in 1961. I just believe that good luck does not mean that someone performed better.

The key is that I am NOT trying to smooth out peak seasons - rather, I am trying to identify seasons where the stats don't reflect true talent and adjust accordingly. The best way to do that is to look at component park factors, shoulder seasons, and other factors that help us estimate performance in the year in question. I recognize that I might over or under correct at times, but I submit that it is better to correct imperfectly than to ignore that Norm Cash had a 370 BABIP in 1961.

Simply put: statistics from a single season are not, in isolation, the best measurement of a player's performance within that season.

I have a couple of technical issues with the 1961 stuff posted so far. Neither may be serious, but I thought I should get them out there, because they may affect everyone's analysis:

1) I think Billy Hoeft should be considered a starter in 1961. He was actually used in the "swing man" role that was common at the time, but has all but disappeared nowadays. He did have more relief appearances than starts, but he had 86.1 IP as a starter against only 51.2 in relief. Also, he had 12 starts, while no one else on Dan G's list of relievers (comment #14) has more than two. Note that you could get a LOT of decisions in the 1960s without starting any games. See Luis Arroyo and Stu Miller in comment 14.

2) Does anyone know anywhere I can find a systematic listing of the arenas where the various big systems consistently differ? Take a look at the list at the top of this thread. WAR and WARP2 like Ken Boyer quite a lot, and the list is ranked by WAR, so he shows high. But Win Shares does not "share" this high opinion. Does anyone know why? The same sort of thing happens with Al Kaline, right above Boyer, but this time only WAR out of the three systems has him ranked high, so it's not just a Win Shares thing. It would really be a help if I knew what sorts of differences the three systems have with each other, since I always try to analyze by trying to find a balance between the various systems.

Thanks in advance, Brock Hanke

When WAR is higher than Win Shares it is usually on a defense-first player. This is due to the artificial compression of defense in Win Shares.

DL - Oh, now, THAT's a help. I don't know if you want to consider either Boyer or Kaline to be "defense first" players, but both were good defenders. Before you posted this info, I was worried because both Ken and Al were childhood heroes of mine (I lived in STL, and was a Tigers/Indians/Bosox fan in the AL). Now I have a serious reason, with which I agree, to keep them at their WAR levels instead of dropping to the WS. I also have something to use whenever I run into this kind of discrepancy. THANKS! - Brock

I will be voting mostly using DanR numbers plus a postseason bonus. It would be a pretty boring project if everyone voted like this but I'm pretty sure they won't.

1) Mickey Mantle
2) Norm Cash
3) Henry Aaron
4) Frank Robinson (small postseason bonus but he doesn't move at all)
5) Jim Gentile
6) Warren Spahn
7) Willie Mays (Why does Dan R think his defense was only average in 1961?)
8) Eddie Mathews
9) Jack Kralick (there's a surprise)
10) Don Drysdale

11) Jim O'Toole (postseason bonus)
12) Vada Pinson
13) Camilo Pascual
14) Juan Pizzaro
15) Mike McCormick
16) Bob Gibson
17) Whitey Ford (postseason bonus)
18) Elston Howard (catcher bonus, postseason bonus)
19) Ken Boyer
20) Roger Maris/Rocky Colavito

I'm going to need to sort out relative pitcher/position player rankings but this is my first swag.

Obviously, it's possible that in 1961 he didn't play the field as well as he did in other seasons, but I think anyone who watched baseball at the time (I'm not one of them) would find the idea that he was the worst fielder in the majors to be ludicrous.

I don't put any stock in defensive reputation. I've seen too many results from the past 10 years where rep seems to badly miss the boat compared to the advanced numbers. Everyone likes to pick on Jeter for this (and it's well deserved), but he's just a drop in the ocean when it comes to examples. To pick another one, Nate McLouth won a GG in CF in 2008 (persumably because he made just 1 error in 150ish games) in spite of advanced numbers showing he was a below average fielder. Defensive reputation favors flashy and/or sure-handed players over ones with better range.

Having said all that, I agree Maris wasn't that bad in 1961, thought I do think he was probably a bit below average. The FR system I use for WAA is basically a linear weight system (using Palmer's formulas), where POs, As, etc are assigned weights, summed together to get FR, then what the average player that year would have accumalated with the same playing time is subtracted to get FRAA. Maris's putouts in the OF this year, for whatever reason (maybe simple lack of oppurtunity), were pretty low compared to the rest of the league, so he suffers in that kind of system. But as I said up thread, given the weight of the other evidence, I don't take that result too seriously. It does look like '61 was a down year for him defensively, but him being that bad is an anomalous result, and it won't effect my voting.

[Reached on Error is] at B-R.

Yeah, unfortunately there's no easy way to download a full dataset from B-R (that I know of), and it's too much work to grab numbers for everyone by hand to incorporate it into my system, especially for a number that I think is mostly luck and evens out. Maybe at some point I'll work on a system to automate grabbing that info, but that's a pretty big project (and I'm a better database programmer than web programmer) so it won't be anytime soon.

I agree that defensive reputation can be biased. The numbers can miss, too. I put some stock in defensive rep even now, and I think it should be done even moreso going years back when the numbers are less accurate. Yes, McLouth won a Gold Glove somehow, but I doubt anyone really considered him a great outfielder. I never heard him called one, and no one will remember him as one. People do remember Maris as one.

Jeter is definitely a weird case. Everyone else that has won that many Gold Gloves is considered an elite fielder. I doubt even managers and writers who love Jeter and think the advanced stats are bullsh!t would list Jeter among the elite defensive shortstops of all time. Ordinarily I would use all the Gold Gloves as evidence he really wasn't as bad as the numbers say. But in his case, I've seen him a lot with mine own eyes, and believe that he was that bad (in some seasons).

What do you think of my other (buried) point, that you shouldn't simply compare Maris's offense to an average RF, since he also played a fair amount of CF?

I will be voting mostly using DanR numbers plus a postseason bonus. It would be a pretty boring project if everyone voted like this but I'm pretty sure they won't.

Out of curiosity, what are you using for pitchers?

I was looking at the underlying numbers for Spahn last night, and was rather unimpressed, particulary with his peripherals. I'm at work now, so I don't have access to them, unfortunately, but I believe he was 7th in IP (a good but not great rank), his SO/BB ratio and FIP were mediocre, and his BABIP was signifcantly below average, suggesting he was somewhat lucky/dependent on his fielders this year. I did have him as the best pitcher in '61, but now I'm not so sure. I need to think about it some more.

Also, that's pretty low for Colavito. Does DanR's rankings hate him that much? All the other systems I looked at had him as a top 10 player (admittedly, some of them, like BR WAR, don't automatically adjust for the different length schedule this year, so are slightly underrating the NL players). Looks like Dan's WARP has him as number #9, if I'm counting right, but with no pitchers. So I guess with 9 (?) pitchers ahead of him, and Howard with a catcher bonus, that makes sense... I see '61 as a down year for pitching performances though, so I'm not sure about that many pitchers ahead of him (or Maris, either, I suppose).

I'm using Dan R's numbers for pitchers. His Salary numbers are higher for pitchers than you would expect given their WAR. This is consistent with GM paying more $$/WAR for pitching.

playerid WARP2
spahnwa01 5.8
kralija01 5.7
drysddo01 5.3
otoolji01 5
pascuca02 4.9
pizarju01 4.9
mccormi03 4.8
gibsobo01 4.7
cardwdo01 4.5
sadecra01 4.5
mcbrike01 4.5
donovdi01 4.4
koufasa01 4.3
fordwh01 4.3
laryfr01 4.2
podrejo01 4.1
staffbi01 4
simmocu01 4
mossido01 4

Yes, McLouth won a Gold Glove somehow, but I doubt anyone really considered him a great outfielder. I never heard him called one, and no one will remember him as one.

You should live here in Pittsburgh. When the Pirates traded him to the Braves, the local columnists tripped over themselves to condemn trading a "gold glove centerfielder" for "nothing".

Saying I don't put any stock in defensive rep was probably a bit of a reach on my part though. Thinking about it some more, I do put some stock in career reputation, particulary for players in the pre-PBP era, if only because there's so little else to go on. Where I don't really put any stock in reputation though, is in yearly results. I think UZR etc. have shown that fielding performance can bounce around a bit, and that even a good/great fielder can have a down year.

What do you think of my other (buried) point, that you shouldn't simply compare Maris's offense to an average RF, since he also played a fair amount of CF?

It's a good point, though my system already accounts for it. He didn't play that much CF in '61 (120 innings, compared to 1282 in RF) but my BRAA are based simply off league average (removing pitchers), and my positional adjustment estimates batting time at the different positions based on innings in the field. I'm not at home currently, so I can't give you the exact numbers (I can double check them later), but of the -27.16 positional adjustment runs I list him at up-thread, I believe 26 or so of them were for RF and 1 and some change for CF.

I'm not sure how exactly all the uber stats handle that though, so it's a valid point that he has more positional worth than someone who solely played a corner outfield spot.

DanR is using a LgAdj of .92 for the 1961 AL and .962 for the NL. Dan R has Elston Howard ahead of Colavito without the catcher bonus. If I rank using just WARP2, no salary considerations then Spahn ends up around Kaline at 14

1) Mickey Mantle
2) Norm Cash
3) Henry Aaron
4) Frank Robinson
5) Jim Gentile
6) Willie Mays
7) Eddie Mathews
8) Vada Pinson
9) Elston Howard
10) Ken Boyer
11) Roger Maris
12) Rocky Colavito
13) Al Kaline
14) Warren Spahn
15) Jack Kralick
16) Roberto Clemente
17) Orlando Cepeda
18) Don Drysdale
19) Jim O'Toole
20) Ernie Banks

Howard's rate stats would rank him as high as 6th with a 7.8 WARP2/YR.

This is an unusual expansion year. Usually there are pitchers that take advantage of expansion but not in 1961. Perhaps the additional games in the schedule made it harder on pitchers than it did on position players. Also, integration was happening faster among position players than pitchers.

I still can't get over how Aaron + Mathews + Spahn = zilch.

I'm using Dan R's numbers for pitchers.

Ah, OK. I had seen in some threads that he had preliminary (?) numbers for pitchers, but his master file on the yahoo site, as well as Tango's site, were never updated with them, and they were't listed in the thread header, so I didn't know anyone was using them.

His Salary numbers are higher for pitchers than you would expect given their WAR. This is consistent with GM paying more $$/WAR for pitching.

Do you think that represents their true worth, or is that a market inefficiency?

Now that we have rules are there any objections to voting 5/30-6/6? I think it would be appropriate to open up a 1962 discussion thread some time next week.

Now that we have rules are there any objections to voting 5/30-6/6?

I suggest you give a shout-out to everyone through the Yahoo site.

Will attempt to connect through Yahoo at home this weekend.

Do you think that represents their true worth, or is that a market inefficiency?

I honestly don't know. It is hard to believe the market has been inefficient for dozens of years.

I will be voting mostly using DanR numbers plus a postseason bonus. It would be a pretty boring project if everyone voted like this but I'm pretty sure they won't.

YES and no they won't.

PS. Congrats and thanks to the guys who got this going against all odds (4 years of inertia).

And finally, how deep is the ballot? 20?

DL from MN--I must admit I don't remember what equation I used for pitchers in that sheet, so I wouldn't put too much stock in it. For pre-1987 seasons I was just fixing the available defensive stats at the time and putting them on the same scale, so if the output seems weird it's because the inputs were. For Retrosheet-available seasons I'd definitely trust a regressed version of TotalZone, DRA, or SFR (preferably all three) over what I have on that sheet. I've always meant to get the right weights for each of those stats myself but have never gotten around to it.

Dan, do you think using TZ alone for the retro-era is superior to using your FWAA?

One minor item I wanted to be sure to mention, which I don't see online anywhere, is the all-star teamed named by The Sporting News after each season since 1925. I believe these are voted on by managers and coaches. In 1961 they decided to name a team for each league for the first time.

AL
1b - Norm Cash
2b - Bobby Richardson
SS - Tony Kubek
3b - Brooks Robinson
OF - Mickey Mantle
OF - Roger Maris
OF - Rocky Colavito
C - Elston Howard
P - Whitey Ford
P - Frank Lary

NL
1b - Orlando Cepeda
2b - Frank Bolling
SS - Maury Wills
3b - Ken Boyer
OF - Willie Mays
OF - Frank Robinson
OF - Roberto Clemente
C - Smoky Burgess
P - Joey Jay
P - Warren Spahn

Probably, since it's using much more granular inputs. But why would you use TZ alone when the other systems are also available?

Attaching bb-ref WAR to that list:

AL
1b - Norm Cash - 10.0
2b - Bobby Richardson - -1.1
SS - Tony Kubek - 3.2
3b - Brooks Robinson - 3.1
OF - Mickey Mantle - 11.9
OF - Roger Maris - 7.2
OF - Rocky Colavito - 7.9
C - Elston Howard - 5.9
P - Whitey Ford - 3.6
P - Frank Lary - 4.1

NL
1b - Orlando Cepeda - 6.1
2b - Frank Bolling - 3.2
SS - Maury Wills - 1.8 (his best year was '62, not '61)
3b - Ken Boyer - 6.4
OF - Willie Mays - 9.4
OF - Frank Robinson - 7.6
OF - Roberto Clemente - 5.3
C - Smoky Burgess - 2.8
P - Joey Jay - 4.5
P - Warren Spahn - 4.8

Some notes:

1. A assume the pitchers were required to be chosen as one RHP, one LHP. That would be traditional. (The tradition never made any sense to me - they play the same position.)

2. All of the corner OF chosen played RF. And if you did something like substitute Aaron for Clemente, you'd still have a RF.

3. Bobby Richardson sort of sticks out like a sore thumb there. Being a Yankee got him a lot of attention, but actually thinking he was good? Sheesh. And it isn't just this vote: he won the Gold Glove and got some down-ballot MVP support, finishing 24th.

The arguments for Richardson:
- He was a .300 hitter. OK, that was 1959, and he actually batted .261 in 1961, but that "wasn't too bad for a MI". (.261 in a league .258 context).
- He never struck out. 23 K's in >700 PA, so he had "bat control."
- He wasn't error-prone (18 errors, .978 F%)

The part about getting 700 PA in the front of a lineup that had Mantle and Maris in the middle, and scoring 80 runs - hey, who notices things like that? No walks, not much power, .261/.294/.316, OPS+ 67.

Who are the reasonable candidates for best 2B and best SS in each league?

I did have [Spahn] as the best pitcher in '61, but now I'm not so sure. I need to think about it some more.

Replying to myself, I'm not sure I do need to think about it all that much more, because I'm not sure any pitchers are going to make my ballot. Spahn was the best one in terms of results, if not ability/performance, and he's still outside of my preliminary top 10.

And finally, how deep is the ballot? 20?

I believe 10 is what was settled on over in the rules thread (expanding in years where the number of teams is greater than 20).

I honestly don't know. It is hard to believe the market has been inefficient for dozens of years.

Yeah, I'm not sure either. It is hard to believe that GMs have so consistently gotten it wrong... but then, how long was OBP undervalued compared to BA?

Here's a repost of my consideration chart from up-thread, now limited to 25 players, and with season length and league difficulty adjustments, if anyone finds that helpful. I increased the stats for NL players by 6.2% to get up to 162 games (except WAA as that's scaled to 700 PA), and then increased the new numbers (including WAA this time) by 4.4% for the higher difficultly of the NL (gotten from DanR's numbers). I'm not 100% sure that's the correct approach, but I'm tossing it out there for people to take a look at.

Who are the reasonable candidates for best 2B and best SS in each league?

Frank Bolling is a solid pick by the Sporting News as the best second baseman in the NL. His 93 OPS+ actually leads the position. And he was a suitable defender- fielding runs has him at +14 and defensive WAR has him +1.4.

At shortstop, the pick should have been Ernie Banks. Shortstop was still his primary position (over 100 games, most of the rest in LF), he hit for a 122 OPS+ and he was still slightly above average defensively (+2 fielding runs, +0.7 defensive WAR). It was a down year for Ernie but his 4.3 WAR demolishes Maury Wills.

In the AL, I'd go with Jerry Lumpe of the Kansas City Athletics at second. Once again, his 97 OPS+ led the position. And he was above average with the glove (+2 fielding runs, +0.1 dWAR). I haven't seen any second baseman with a better WAR than Lumpe's 2.8, though it's possible I missed somebody who had a great defensive year (but not Maz, his WAR for '61 was only 1.4).

AL shortstop was a bit of a wasteland. Kubek may very well have been the best of the bunch. I don't think any shortstops beat his 90 OPS+ and he was good enough with the glove (+14 fielding runs, +1.4 dWAR) that nobody could close the gap that way.

Who are the reasonable candidates for best 2B and best SS in each league?

Here's how my system (using the adjustments described in the last post) sees the top 5 of each (the rankings are in terms of the total player base):





Mr. Cub had a decent year, but the rest of these guys are pretty unimpressive.

Looks like my system pretty much agrees with you, Chris Fluit, except it has Held over Kubek, and that's a pretty minor difference.

how deep is the ballot? 20?

From the Rules thread [#130]:

"Ballot Length: The length of the ballot will be proportional to the number of major league teams with a minimum ballot length of 10. The ballot length shall be N/2 where N is the number of major league teams for seasons where there are more than 20 major league teams [1969-present]."

I would encourage voters who are able to rate more than 10 to post a larger ballot.

Dan, which other systems are freely available, and how can I find their results? I'd certainly like to create a fielding metric for the retro-era using all available systems as an input (I'm thinking of weighting each system equally, and adjusting for the different metrics' different standard deviations).

Gonna try this again, since I screwed up the season length percentage in #69, and it's too late to edit it. It should only be an increase of 4.94% to go from 154 games to 162. Clearly, math is hard.

Looks like my system pretty much agrees with you, Chris Fluit, except it has Held over Kubek, and that's a pretty minor difference.

I'll add that Held had a very good year with the bat. 121 OPS+, .342 wOBA (Lg Avg .314), in 590 PA as a SS. Banks was pretty even with him, 122 OPS+, .338 wOBA (Lg Avg .312) in 573 PA, but Banks also played 23 games in LF, while Held played all of his games at SS. And none of the AL shortstops were close (well, Pumpsie Green and Bob Johnson were sorta close, but neither broke 300 PA). Unfortunately for Held's MMP chances, he was probably average, at best, defensively this year (though BP's WARP1 think he's better than that, which is how he did so well in their rankings). That's how Kubek closes the gap in total worth by so much, in spite of hitting a lot worse.

If Held had played defense as well as Kubek, he would have had an outside shot at the top 10, though a few more PA would have helped as well. I think it's going to be tough for many MI to crack the top 10 most years though.

I looked through the MiL batting leaders on BR. It's not complete data, as there's nothing for individual players in the MxL, but it's better than nothing. The stats are pretty complete for the other AAA/AA leagues, but don't include HBPs for AAA, or SF/SH at all, so I'm just ignoring them (depressing OBPs for guys who got plunked a lot, I suppose).

I didn't see anyone who's a slam dunk case. Most of the best performances came in D or D- leagues. I have no clue what the correct conversion rates would be for those leagues, but I suspect it would be pretty low. Class D was 6 levels below the majors, D- was 7. That would make them the apparent equivalent of the modern Rookie Leagues and the DSL/VSL respectively. The MLE calculator on minorleaguesplits.com doesn't go down that far, the lowest league it covers is the A leagues. But in '61 Tony Olivia put up .410/???/.639 for a D- team in 249 AB. Putting that performance on a neutral team in the modern SAL (A level league) would give an MLE of .295/???/.426. And I suspect a D- league in '61 was significantly farther below the majors than a modern A league is, so his performance would be even worse. Given all that, I chose to ignore anyone below AA level.

So, assuming modern conversion levels and a neutral team in the IL for AAA players, a neutral team in the EL for AA players, regardless of what specific team/league they played for (a crude measure, definitely, but I'm just trying to identify candidates), these would be the top hitters...
Don Leppert C AAA 127 PA .386/.449/.596 MLE .332/.386/.497 .883 OPS
Gene Oliver 1B/C AAA 503 PA .302/.421/.612 MLE .255/.356/.492 .848 OPS
Dick McAuliffe SS AAA 290 PA .353/.418/.565 MLE .304/.359/.471 .830 OPS
Carlos Bernier OF AAA 588 PA .343/.463/.538 MLE .289/.391/.439 .830 OPS
Boog Powell 1B/OF AAA 539 PA .321/.388/.593 MLE .277/.333/.490 .823 OPS
Bobby Prescott OF AAA 518 PA .301/.413/.582 MLE .255/.350/.471 .821 OPS
Cliff Cook 3B AAA 583 PA .311/.391/.588 MLE .267/.334/.484 .818 OPS
Bob Farley 1B AAA 452 PA .307/.436/.538 MLE .258/.367/.435 .802 OPS
(skipping some other corner hitters who don't have as good a case as the guys above for other interesting candidates)
Bob Johnson 3B/SS AAA 384 PA .332/.338/.548 MLE .288/.334/.458 .792 OPS
Joe Torre C AAA 117 PA .342/.376/.532 MLE .298/.327/.451 .778 OPS
Bob Uecker C AAA 373 PA .309/.418/.510 MLE .262/.354/.418 .772 OPS (in reality Uecker outhit Torre, as the rounding on Torre's small totals is giving him better slash stats than he really should have, I think)
Johnny Lewis LF/RF AA 516 PA .293/.428/.533 MLE .230/.334/.394 .728 OPS (top AA candidate, keeping in mind that I have no data for the MxL, which was considered AA during this time period)

I put everyone in the same context for their level, rather than using their actual leagues and teams (which still exist in some cases), because while I know the PCL is more of a hitters league than the IL nowadays, I wasn't sure if it was in '61. And some of these guys played in the AA, which doesn't even exist anymore, and thus isn't an option on the MLE calculator, so it just seemed fairer this way. For the IL, a neutral team seems to be using a conversion rate of .88 for BA, and .78 for OBP/SLG (the PCL looks more like .85/.75). For the EL it looks to be about .84/.71.

Anyway, league average OPS in '61 (removing pitchers) was AL .750 and NL .757. To make the top 25 in OPS (min. 100 PA) a player would have needed to put up a .909 (which would have tied Al Kaline). As I said above, this is a crude measure, but I don't see any of these guys making a case for a vote solely on their offense.

The guy who comes out looking best here might be "Not An Indian" Bob Johnson, who spent half the year in the majors at SS, and hit .295/.350/.442 and 115 OPS+ in 248 PA. Combining his major league stats with his MLEs gives stats of 633 PA .290/.340/.453 OPS .792 and an OPS+ of 116 (using the Senators' BPF for his MLEs). Unfortunately for his MMP case, everything I looked at had him as an average defender at SS in his MLB time this year, and I won't even attempt to analyize his MiL fielding performance. So he has a case as the 2nd best AL SS this year, or maybe even the best if his MiL fielding was better than MLB average, but that's about it. Woodie Held is pretty much the upper limit of his possible value.

Don Leppert also played in MLB this year, and if had hit anywhere close to his MLEs he might have had a case for 2nd best catcher after Howard, but he didn't (97 OPS+ in 62 PA), so it's moot.

Alex King, the other systems are SFR (which should be up on BP if you search for Dan Fox) and DRA (just released in Michael Humphreys's new book). An equal weighting is surely wrong--you want to multiple regress the three systems against an average of current PBP statistics (e.g. UZR) to see what combination is most accurate. That's what I did for my 1987-2005 numbers; I just never got around to doing it for earlier seasons.

Dan, the problem with regressing against UZR is that you are then assuming that UZR is accurate, and UZR as well as other batted-ball metrics have been called into question recently, in all the discussions about range bias at the Book Blog. So I don't necessarily think that we should be using the non-batted ball metrics to predict the batted-ball metrics when those batted-ball metrics are of questionable accuracy.

Also, Dan Fox's SFR appears to be only available for infielders in most of the Retro-era. Thanks for pointing out the locations of DRA and SFR!

Yeah, but it's really f'ing good for infielders, at least for the period where it somehow has some more granular data than the others (Humphreys once explained this to me). I took the "true" rating to be an average of the STATS database systems (old UZR and Dial's RSpt) and BIS systems (DRS and then PMR).

Is there any doubt that actual play-by-play metrics are superior to the efforts to mimic them with mere Retrosheet data??

Last year, I would have agreed that the batted-ball metrics were far superior to the PBP metrics. Now I'm not so sure, because of several discussions at the Book blog about the possibility of "range bias":

http://www.insidethebook.com/ee/index.php/site/comments/conclusions/
http://www.insidethebook.com/ee/index.php/site/comments/our_lab_bias_in_batted_ball_data/
http://www.insidethebook.com/ee/index.php/site/comments/conclusions/#137 (summary of Colin's findings)

Quick summary: There's a significant difference between the WOWY/PBP metrics and the batted-ball metrics, even over a period of several years when all random variation should even out. Colin Wyers builds a convincing case that this variation is due to "range bias": that is, stringers mark a ball as closer to the fielder when he fields it, and further away from the fielder when he does not field it (so someone like Jeter would be recorded as having harder chances on average). The upshot of all this is that UZR probably shouldn't be used as the "gold standard" of fielding metrics. So I would say that yes, there is significant doubt that batted-ball metrics are superior to the PBP fielding metrics.

Also, I would like to consider Colin's nFRAA, but I haven't been able to find it on BP's website (I don't have a subscription). Can someone with a subscription access this data? I remember being able to access it last year, but it seems to have moved/disappeared since then.

Dan (or anyone else), have you read Michael Humphreys' book? I'm thinking about buying it so I can use the DRA data, but if I don't want to enter all the DRA data manually--I'm hoping that if you buy the book, you also get a data table or something like that.

Vote for 10 this year.

It was a little disappointing to see TomH decide not to participate but understandable. I will say that this project lends itself well to popping in every once in a while. There is nothing wrong with deciding to vote when a year strikes you as particularly interesting.

I'd also be interested in hearing what Humphrey's book contains, though I may pick it up regardless.

Rules for 1961 voting:

Candidate Eligibility: Any North American professional baseball player is eligible for the MMP award including players in the top Negro Leagues or independent teams. Voters should consider the player's on-field contribution to MLB team(s) in 1961 only. If part of the season was spent outside MLB, that value may be considered as well. However, the player's on-field contribution should be judged in relation to the highest level major league, not relative to a minor league. A season may include playoff or World Series games but does not include spring training or exhibition games. No credit will be given for games not played due to injury, wartime service or contract holdouts.

Ballot Length: The length of the ballot will be 10.

Voter eligibility: All voters must post a preliminary ballot in the ballot discussion thread at least 2 days before voting ends. All voters must fill out a complete ballot. Voters must briefly explain their ballot choices. One person, one vote; anyone determined to have voted with multiple accounts will be banned and their votes will be disallowed. The MMP ballot committee has authority to exclude any ballot that does not meet these requirements.

Scoring: Points will be given in descending order with the #1 player receiving 15 points. For example, with a 10 player ballot the points will be 15-14-13-12-11-10-9-8-7-6. The player with the highest point total will be named the MMP for a particular year. In case of a tie, the tiebreaker will be number of 1st place votes. If the first tiebreaker does not determine a winner the players will be considered co-MMPs.

Some pitcher MLEs using the data from BR's minor league pitching leaderboard. These are a pain in the ass, so I'm only go to do a handful. I ignored anyone below AA. Also, BR doesn't have league pitching stats for the PCL or the IL, so can't really translate anyone who played in those. That's going to limit this to AA and the American Association. And as far as I can tell, it wasn't a good year for pitching performances in the American Association.

Howie Koplitz AA 23-3 230IP 2.11 ERA 171K (est.) 75BB 195 ERA+ MLE 24-4 245.1IP 2.50 ERA 142K 83BB 164 ERA+
John Boozer AA 19-9 207IP 2.61 ERA 166K (est.) 57BB 158 ERA+ MLE 18-10 209.2IP 3.09 ERA 137K 53BB 131 ERA+
Harry Fanok AA 16-7 185IP 2.24 ERA 158K 68BB 153 ERA+ MLE 17-9 203.2IP 3.59 ERA 144K 73BB 123 ERA+
Larry Maxie AA 17-7 203IP 2.08 ERA 47K 97BB 165+ MLE 18-8 223.1IP 3.09 ERA 36K 102BB 119 ERA+
Paul Toth AA 18-7 190IP 2.37 ERA 125K 57BB 144 ERA+ MLE 19-9 209IP 3.75 ERA 112K 58BB 118 ERA+
Lauren Pepper AAA 15-11 227IP 3.45 ERA 106K (est.) 94BB 118 ERA+ MLE 15-13 245.1IP 4.12 ERA 90K 128BB 109 ERA+
Don Randolph AAA 18-9 221IP 3.54 ERA 118K (est.) 46BB 116 ERA+ MLE 18-10 227IP 4.01 ERA 104K 50BB 102 ERA+

Huh. So, unlike batters, there could actually be a serious candidate here in Koplitz. No question he had a great year. He was second in the Southern Association in IP (behind the leader by 1 IP), and he put up his 2.11 ERA when the lgERA was 4.12 (he was technically fourth in the league in ERA but the 3 guys ahead of him were either relievers or had minuscule IP). That 164 ERA+ would have made him the MLB leader amongst starters, though 245.1 IP would only have been 15th. Of course, he actually did pitch in the majors as well this year, putting up a 188 ERA+, but in only 12 IP, so it doesn't mean too much (but does bump his total innings up a bit).

One simple way to generate PRAA is ((lgRA - RA) * IP) / 9. By this method, Spahn lead MLB with 37.38. He was only 15th in MLB in RA+, but sixth in IP, and none of the guys ahead of him had a better RA+, so he squeaked by Ford, who was #1 in IP, but only 21st in RA+. Koplitz's MLE RA was 2.86 (163 RA+, which would have been 1st in MLB amongst starters), and the AL's was 4.58, so using this same method Koplitz's PRAA would be 46.88, far and away the best. And again, that's not counting the 12 innings he did pitch in MLB.

Still, this is all a pretty crude measure of value. No data on Ks for individual pitchers was listed for the Southern Association this year, so I had to estimate them based on his A and AAA K/9 rates. I used KJOK's MLE spreadsheet for the conversion, a .84 conversion factor, and 100 for the minor league PF, so if Birmingham was a serious pitcher's park, that would take some of the air out of his numbers (I have no idea if it was or not). His performance is projected for the Tigers (his parent club) for this year.

I'll have to think about this some more before deciding if I want to put him on my ballot. There's always a lot of uncertainity with MLEs, regardless of how well they're done, and that's just magnified when someone is being projected up two levels rather than one. His translation seems a bit too good to be true, and the lack of a real PF for his minor league park is a pretty serious issue. His MLB career was very brief, and none of his other MiL years look nearly as good, so this seems like something of a fluke year (which doesn't necessarily mean it wasn't meritorious). Haven't really looked at his hitting or fielding to get a complete picture of value. I encourage anyone else who's interested to take a look at him to see if I screwed up somewhere, because, even with those doubts, I still think it's clear he had an excellent year in AA.

And ignoring Koplitz for the moment, here's my preliminary ballot. It's strongly based on the last consideration set I posted, with adjustments for season length and league strength for the NL, but I have made some subjective tweaks.

1) Mickey Mantle (#1 in every system I checked, usually by a healthy margin. Don't see how anyone else can be the choice.)
2) Norm Cash (Even with schedule and strength boosts for the NL, still an easy #2.)
3) Willie Mays (Mays, Aaron and Robinson were very close, with no consensus amongst the systems about who was better. So I'll go with the one who probably had the most defensive value as a tiebreaker.)
4) Hank Aaron
5) Frank Robinson
6) Rocky Colavito (Colavito and Gentile are passed by the previous trio due to the AL's relative weakness.)
7) Jim Gentile
8) Eddie Mathews (Not that far behind the previous two, when the NL is adjusted, but can't quite catch them.)
9) Ken Boyer (Very close to Mathews in value, and directly comparable to him, since they played same position in same league. But I think Mathews' stronger hitting trumps Boyer's likely defensive edge.)
10) Elston Howard (Ended up giving him a 12% C bonus, which landed him right about here.)

11) Roger Maris (Moves up here after I tossed my WAA system's hatred of his defense, but still can't crack top 10.)
12) Vada Pinson
13) Frank Lary (Spahn had the best results this year, but I think that was a bit of an illusion. His peripherals weren't good. His FIP, in particular, was brutal. So Lary ends up my top MLB pitcher this year. Not that any of them are making my ballot.)
14) Al Kaline
15) Orlando Cepeda
16) Harmon Killebrew
17) Roberto Clemente
18) Warren Spahn
19) Whitey Ford
20) Juan Pizarro

I really wasn't sure how to start with my preliminary ballot, so I simply put in the numbers from the systems at the top of this thread and in post 75, then added in WAR from Fangraphs. From that all I've done is create percentages for each and added them up. I went ahead and plugged in the schedule adjustment and league adjustment that were already done in post 75 and this is what spat out:

1- Mickey Mantle
2- Norm Cash
3- Hank Aaron
4- Willie Mays
5- Frank Robinson
6- Eddie Mathews
7- Jim Gentile
8- Rocky Colavito
9- Vada Pinson
10-Roger Maris
11-Ken Boyer
12-Al Kaline
13-Orlando Cepeda
14-Elston Howard

From here I need to decide a few things...
1) How much catcher bonus to apply. I'd love to hear suggestions!
2) Frankly I don't like the top 2 being from the weaker league... but don't know if I should do anything about that.
3) Figure out which systems to drop (if any) or which systems to emphasize (if any).

Sounds like a good start. How did you deal with no WAR for pitchers when adding in fWAR, though?

@Nate the Neptunian re:comment #88

Umm.... Would I need to add in fWAR in one of the systems? [I double checked my spreadsheet and should note that I didn't use ALL of the systems in your comment #75 as I did not include your own system for various reasons]. The way I dealt with not having pitching numbers in all the systems was to look over the rankings of the systems with pitching numbers and look at the pitching leaders in 1961 at baseball-reference and I decided that no pitchers were going to be in my top 10 anyway, so I just ignored them.

So... which systems would I need to add in fielding to myself? I thought the systems I used all incorporated fielding...
Thanks!

I think Nate means Fangraphs' WAR by fWAR, not fielding WAR.

Yeah, sorry, that's what I meant.

> probably had the most defensive value as a tiebreaker

Aaron was a pretty damned good defensive CF in 1961.

Re: comments 88-92:

Yeah, a couple hours after I posted my response (#89) I was doing something else and suddenly it struck me (D'OH) that you were referring to fangraphs WAR
and not fielding WAR. As to why I didn't realize that at the time... I wish I knew!

Sorry for the confusion (and sounding like an idiot).

So.... yeah, 1961 was pretty easy to include fangraphs WAR since it seemed there would be no pitchers in the top 10 anyway. If I were to do the same
thing in a year where I would have to consider pitchers then perhaps I would do position players and pitchers seperately and then try and combine them.

> 1) How much catcher bonus

Catchers typically play 20% fewer games than other position players. No catcher bonus should exceed that for playing time. Then there is the question of whether a WAR calculator fully captures the defensive contribution of a catcher (unlikely).

Aaron was a pretty damned good defensive CF in 1961.

Yeah, on taking another look, '61 was one of Aaron's best defensive years, while there's conflicting evidence on Mays. Mays had the better offensive year, but I'm not sure if that's enough to compensate. Admittedly, Mays played CF all year, while Aaron split time between CF and RF, so he'd have a larger negative positional adjustment.

Hmm. Total Zone actually kinda likes Mays' defense in '61. +14 Runs for Mays, +24 for Aaron. DRA disagrees; -1 Runs for Mays, +26 runs for Aaron. BG WAR thinks they were both above average defenders; .8 dWAR for Mays, 1.1 dWAR for Aaron. But it uses a top down approach similar to fielding WS to calculate dWAR that I'm not completely sold on. BP's FRAA1 is +5 for Mays, +16 for Aaron, but then its FRAR1 is +30 for Mays, +36 for Aaron, closing the gap (because Aaron spent time in RF?), but I don't know what their methodology is. I can't find the SFR data on BP's site, that might be subscribers only.

Pretty much everything agrees Aaron had the better defensive year, but differs on how much. Having said that, I'm leaning towards changing my mind and ranking Aaron above Mays.

SFR data can be found at this link, but it's only available for infielders 1957-2007, with gaps in 1984, 1985, and 1999.

Ah, OK. Won't help with Aaron vs Mays, then.

meta-discussion of effects of changing ballparks on offense and defense (may apply to 1961?):

We all know ballaprks affect value. Primarily offesne, and we have metrics such as OPS+ and ERA+ which account for this in the large sense. Some would use these metrics as the 100% solution to adjust, using the "value = merit" position. Others would use what I term the "whatif?" position that guys like DiMaggio were unfairly penalized by their home park configuration, and give them some bonus.

What do you do about players who are traded, or whose teams move to new stadiums? There could be adjustment period in either the batters' swing (offense) or learning to play the conditions (defense) if the park plays very differently than what he was used to. Does this affect "merit" if a player experienced unusual difficulty with a new stadium?

Willie Mays: it is somewhat well known that Mays altered his swing when he went to Candlestick, as hitting home runs to left-center became harder than to right-center. Some items that aren't noted as much are
1) Mays switched parks twice, first to Seals stadium in 58, then Candlestick in 60.
2) By many of the advanced metrics (going from memory, do not have them with me), Mays had a "trough" in defensive value from 1958-61 or -62. By this I mean his FRAA or def WS were higher in the mid 50s, and again in the early-to-mid 60s, than in between. This could easily (although not provably!) be due to his having to learn two new stadiums' patterns. And of course learning the winds in S.F. was a well-known challege!

When assessing defensive range/value, are the best ##s to use the 1-year estimates, or averaging over longer periods? If longer, it might be wise to consider the extra challenges imposed by unique conditions.

Voting will be 5/30-6/6. Make sure you have a prelim ballot posted by 6/4 in this thread.

DL - Thanks for answering my question about voting timing. I REALLY appreciate it, as I had started to sweat. - Brock