Just came across a new Bobby Richardson bio in the bookstore the other day. Has there ever been a more lauded not-very-good player? (His career WAR is 6.5. No, not sixty-five, but six point five.)

Anyone who puts together an All Time Cubs team which does not have Ernie Banks in the starting lineup, and has him on the bench as a back up first baseman, doesn't know what he's doing.

edit: More dumbness. The Cubs' bench is all infielders. The Red Sox bench is all outfielders except for Petrocelli. The Yankees bench has 3 catchers (in addition to the starter).

Which means they still left out someone good since the Yankees have had 5 very good catchers over the years: Dickey, Berra, Howard, Munson and Posada.

I think it's probably safe to say Banks was a better player than Joe Tinker. And on a related note, the idea that Evers and Tinker would both be on the bench most of the time would make that dugout...interesting. We might get a Zambrano-Barrett style throwdown or ten.

There's no way Levi Meyerle was better than Ezra Sutton! Is this the kind of debate this site is supposed to stir?

I recently realized that we’re rolling out features to address many of my favorites.

I love the wording of this. Makes it sound like hackers did it or something.

Explanation of the process.

-- MWE

Anyone who puts together an All Time Cubs team which does not have Ernie Banks in the starting lineup, and has him on the bench as a back up first baseman, doesn't know what he's doing.

I'm guessing he's using Walts version of which position to put a player at. Banks doesn't make it above Anson at first nor above Tinker at Short. In this case, I think instead of letting the numbers work by themselves, that the site should make a manual adjustment. 47.8 of his career 62.5 war was acquired at short. It doesn't matter if he played more games at first base, he should be listed as a shortstop if that is where he acquired the most value.

I'm just guessing but I think the Yanks would slide Dimaggio or Mantle over to LF and put Charlie Keller on the bench.

Not too many surprises on the Detroit page. I was glad to see John Hiller nosed out the perennial HOF wannabe Jack Morris.

Well, in 1951 DiMaggio and Mantle played together.
DiMaggio got center and The Mick played right.

Just came across a new Bobby Richardson bio in the bookstore the other day. Has there ever been a more lauded not-very-good player? (His career WAR is 6.5. No, not sixty-five, but six point five.)

Keys to fame:

1) Play on real good teams. People will mention how many rings you have all the time.
2) Have people hit line drives at you in critical situations.

I would have benched Veach and put Crawford in left field - but what to do about Heilmann?

Heilmann would make a nice DH, hitting .380 or so in his prime.

One of the greatest sports debates is whether or not a player should be in the Hall of Fame.

It's great like the flu.

I'm guessing he's using Walts version of which position to put a player at

I'm not sure how Bautista is listed as the third baseman for the Blue Jays. He's started 96 games there and it's never been his primary position. If they're doing that, they might as well slide Wells to left and start Devo in CF.

The one that perplexed me was Bo Jackson as the leftfielder in KC, with Alex Gordon slotted in at third. Gordon's played more games in left than he has at third, and virtually all of his value has been compiled there, and vastly exceeded Bo's.

The one that perplexed me was Bo Jackson as the leftfielder in KC, with Alex Gordon slotted in at third. Gordon's played more games in left than he has at third, and virtually all of his value has been compiled there, and vastly exceeded Bo's.

Wait...how bad does the history of your team have to be, to get Bo Jackson on the all time team at a corner outfield spot? That makes no sense to me, in 40 years or so, the Royals haven't had a leftfielder produce more than 6.2 war?

Edit: Alex Gordon alone has two individual seasons where he played 140+ games in leftfield and produced more war than Bo did in his career as a Royal.

Wait...how bad does the history of your team have to be, to get Bo Jackson on the all time team at a corner outfield spot? That makes no sense to me, in 40 years or so, the Royals haven't had a leftfielder produce more than 6.2 war?

Of course they have. It's just that those leftfielders are listed at other positions for whatever reason. Any sensible Royals All-Time team that includes Willie Wilson and Amos Otis will have Otis in center and Willie in left.

Wait...how bad does the history of your team have to be, to get Bo Jackson on the all time team at a corner outfield spot? That makes no sense to me, in 40 years or so, the Royals haven't had a leftfielder produce more than 6.2 war?

Well, Gordon has, and he has played LF more than 3B as #16 points out. It also hurts that Willie Wilson spent a lot of time at LF, but more at CF, and Johnny Damon spent a lot of time at LF, but more at CF. And Lou Piniella spent like 5 years there and was barely above replacement level. The Royals have sucked, but not because they haven't had good left-fielders.

Well, Gordon has, and he has played LF more than 3B as #16 points out. It also hurts that Willie Wilson spent a lot of time at LF, but more at CF, and Johnny Damon spent a lot of time at LF, but more at CF. And Lou Piniella spent like 5 years there and was barely above replacement level. The Royals have sucked, but not because they haven't had good left-fielders.

I'm thinking people are taking my comment as an attack on the Royals history and not on an attack on the methodology that the website choose to use for determining positions. I can see why they thought that, but it was not intended to be a slight on the Royals history at all. (I save that for the Mets.) I just can't fathom how anyone could look at the Royals all time team for a second, see Bo Jackson in it, and not really think about redoing the methodology for determining the all time team.

The Pirates IF is crazy. I'd have done Stargell at 1b, stuck with Maz at second, let Vaughan play SS and slid Wagner to Third (He played Third on and off through 01)-that would help that team out tons.

And yeah-I'd have put Mantle in LF and Dimag in Center too.

I don't know anything about this "Hall Rating" stat, but presuming it's reasonable:

1. Babe Ruth, RF 322
17. Charlie Keller, LF 87
33. Tommy Henrich, RF 66

Didn't Ruth actually split his time on the Yankees roughly evenly between RF and LF?

And if you split him in half, he's the best Yankees RF and the best Yankees LF.

Hell, you could almost split him in quarters.

The accompanying article says this about the fact that ARod is listed at SS rather than 3B:

Graig Nettles is the Yankees third baseman over Alex Rodriguez, even though A-Rod’s Hall Rating with the team is higher. Why? A-Rod is still considered a shortstop primarily, as he’s played 110 more games there than third base. Therefore, he ranks behind Derek Jeter. After A-Rod’s games at third pass his games at short, he’ll take the third base spot from Nettles.

This despite him having played almost no games at SS for the Yankees. But anyway, yeah, it seems like the methodology described there should put Gordon in LF instead of at 3B, since he played 363 games in LF but only 329 at 3B. So something's screwy.

Maybe they do something wacky like first determining whether the person is "a corner outfielder", "a corner infielder", "a middle infielder" and so on, before then going on to determine a more specific position? In that case, Gordon would work out as follows:

LF = 363
3B = 329
1B = 40
RF = 3
DH = 3
SS = 1

Therefore:

CI = 329 + 40 = 369
CO = 363 + 3 = 366
DH = 3
MI = 1

Therefore he's a corner infielder, and since 329 is greater than 40, he's a third baseman, not a first baseman.

Graig Nettles is the Yankees third baseman over Alex Rodriguez, even though A-Rod’s Hall Rating with the team is higher. Why? A-Rod is still considered a shortstop primarily, as he’s played 110 more games there than third base. Therefore, he ranks behind Derek Jeter. After A-Rod’s games at third pass his games at short, he’ll take the third base spot from Nettles.

And how do you reconcile that with Jose Bautista, he of a 96 game career at the hot corner, at 3B?

Nettles over ARod at 3b is somewhat defensible, but if it's because he's had more career games at SS than 3B and thus has to be considered a SS, that's another piece of dumbness for this site. ARod has been almost exclusively a 3B for the Yankees, and though he trails Nettles in Yankee 3b games, has has compiled more value (BBREF WAR).

Maybe they do something wacky like first determining whether the person is "a corner outfielder", "a corner infielder", "a middle infielder" and so on, before then going on to determine a more specific position? In that case, Gordon would work out as follows:

LF = 363
3B = 329
1B = 40
RF = 3
DH = 3
SS = 1

Therefore:

CI = 329 + 40 = 369
CO = 363 + 3 = 366
DH = 3
MI = 1

Therefore he's a corner infielder, and since 329 is greater than 40, he's a third baseman, not a first baseman.

Jose Bautista:

RF - 415
LF - 55
3B - 383
1B - 13

Bautista has more games at CO than CI, and more games at RF than 3b, and far, far more games at RF than 3b for the Blue Jays.

I didn't notice the Bautista example when I came up with that wacky hypothesis. Here's an even wackier hypothesis that accounts for the assigned positions of each of Bautista, Gordon, and ARod:

Perhaps instead of games played at each position, they're using seasons played at each position.

Bautista:
3B: 9
RF: 7
CF: 5
LF: 4
1B: 3
2B: 2

Gordon:
3B: 4
LF: 3
1B: 3
RF: 1
SS: 1

Rodriguez:
SS: 12
3B: 9

Maybe I shouldn't have said "even wackier"; it might be even wackier if intentional, but it's plausible as a bug.

No, wait, it doesn't account for DH. Looks like Rodriguez has played 13 seasons at DH, more than the 12 of SS.

OK, yet another wacky guess:

Chances. Probably as a bug rather than intentional.

Bautista:
3B: 1037
RF: 763
Others: less

Gordon:
3B: 860
LF: 809
Others: less

Rodriguez:
SS: 5750
3B: 2852

But then, how would he get (pretty much) anyone at DH?

My only guess is that it was, for some bizarre reeason, defensive games played through 2011.* That would account for both Bautista and Gordon being slotted at positions other than the ones they should be.

* Though it looks like value through the end of 2012, based on Gordon's overall standing.

Well, I made the site. I went here, saw the comments, got bummed, closed the window, but decided to come back.

So, one of the themes of this site is just seeing what a mathematical formula can do with doing things like:
- populating the Hall of Fame
- ranking players by position
- ranking players by franchise

A lot of people think it's neat. Some don't. That's cool.

The main focus of this feature was supposed to be the ranking of players by their contribution to the franchise. I thought I'd take that a step further and say, "hey, why don't I show the top guys at each position?" That is what people here don't seem to like—the use of a primary position.

I know it wasn't perfect. That's why I spelled out all the things that were weird in the blog post announcing the feature. I've got some ideas about how to improve primary position on a seasonal level and/or on a franchise level. That will improve these pages. Or the sidebars at least.

Just remember—I'm just another guy like you guys who likes baseball history. And also likes to make things and experiment. Calling it dumb when people try to create something new just stinks. Especially when I know the limitations and painstakingly spelled them out.

A bunch of people have really enjoyed the new update. Thanks a bunch to the folks sending nice words.

Just came across a new Bobby Richardson bio in the bookstore the other day. Has there ever been a more lauded not-very-good player? (His career WAR is 6.5. No, not sixty-five, but six point five.)

Two basic reasons for the lauding of a mediocre player:

1. He had the rep as a first rate defensive 2B, and in fact he and Kubek were a very good DP combo, in great part due to Richardson's quick release.

2a. 1960 World Series, 11 for 30, .367
2b. 1961 World Series, 9 for 23, .391
2c. 1964 World Series, 13 for 32, .406

The irony is that if Richardson hadn't booted easy DP balls in both games 4 and 5 of the 1964 World Series, the Yanks would have wrapped it up in 5 games. The first one in game 4 set up Ken Boyer's grand slam in a 4 to 3 game, and the second one in game 5 meant that Tom Tresh's 2-out game-tying homer in the 9th off Gibson wasn't a game winner, and the Yanks went on to lose in the 10th.

He's gotta be kidding, having Tony Murphy on the Metropolitans' bench. The guy played ONE game for them. Today's NY Times has an article about how businesses, search engines, etc., are using human judgement along with their algorithms to improve results. I think this Hall of Stats guy needs to read it.

Just remember—I'm just another guy like you guys who likes baseball history. And also likes to make things and experiment. Calling it dumb when people try to create something new just stinks. Especially when I know the limitations and painstakingly spelled them out.

I'm sorry your feelings were hurt, but the way you handle positions IS dumb, and hiding behind "that's how the computer handles it" is a lame excuse. Take ARod. Only once he plays 110 more games at 3B, does he become the all time Yankee 3B, despite the fact that he's already there on value. What's more, once he does that, he ceases to become the all time SS for the Mariners and Rangers, because he's now a 3B. Presumably, at that point he becomes the Mariners all time 3B, despite the fact that he never played a game at 3b and is the franchise leader at games at SS for the Mariners. How does that make any sense, and how is that not worthy of some ridicule?

He's gotta be kidding, having Tony Murphy on the Metropolitans' bench. The guy played ONE game for them.

So did Jones? (no first name).

But Tom Forster, who played 114 games didn't make the cut. Neither did ed Kennedy who played nearly 300.

Look, who is and who is not on the all time team for a franchise that played 6 seasons 140 years ago is trivial trivia. But if your algorithm spits out two guys who played one game as superior to another guy who played 300, that's a good reason to be skeptical about your more relevant choices.

I'm sorry your feelings were hurt, but the way you handle positions IS dumb, and hiding behind "that's how the computer handles it" is a lame excuse.

Did I say that? What I tried to say was that I went with an original release and I'm iterating to make it better.

how is that not worthy of some ridicule?

I dunno, just not the way I operate. If I see a person make an attempt at this, point out where the calculations have flaws, and also point out that he's working on addressing them, and I wouldn't ridicule. I'd perhaps offer feedback and see what the next release brings. I also don't actually ridicule people who are trying to create things, so maybe that's the key difference.

I only looked at the Yankees page, and there were more than a few howlers, such as Allie Reynolds stuck down at #68, Eddie Lopat at #77, and Vic Raschi at #100. Looking at those rankings, you'd never know that these three starters were the core of a team that won 5 straight World Series, while collectively in those 5 years they pitched to a postseason ERA that was about a full run lower than their regular seasons'. They're not loved by contemporary statisticians due to their relatively short peaks and quick fades, but it's still beyond ludicrous to rate Reynolds below Clete Boyer, Lopat below Frankie Crosetti, and Raschi below Brett Gardner. It shows what overreliance on counting stats can do to one's critical faculties.

That said, I do appreciate the time and effort that went into an obviously prodigious task. I know that I could never come up with anything even half as coherent as this.

I'm sorry your feelings were hurt, but the way you handle positions IS dumb, and hiding behind "that's how the computer handles it" is a lame excuse.



Did I say that? What I tried to say was that I went with an original release and I'm iterating to make it better.

OK, sorry, I misunderstood your post.

how is that not worthy of some ridicule?



I dunno, just not the way I operate. If I see a person make an attempt at this, point out where the calculations have flaws, and also point out that he's working on addressing them, and I wouldn't ridicule. I'd perhaps offer feedback and see what the next release brings. I also don't actually ridicule people who are trying to create things, so maybe that's the key difference.

Ridicule was too strong a word. Consider it harsh criticism and peer review. I'm glad you are taking the comments and critiques to heart

Cool, thanks. I'm just trying to make an awesome site. I knew the positions somewhat an issue. But I wanted to get what I had live before I iterated on that part of it (since it'll be a pretty massive calculation from multiple sources).

While I have you, here's what I want to do…

I've now got a Hall Rating value for each season. I'm going to get positions from the Lahman database and map them to each season, dividing that season's Hall Rating across the positions. This will give me the ability to:

1. Get a global primary position for the player based on VALUE not games played.
2. Get a primary position for each FRANCHISE.

That'll make things better all around. Of course, I should also add that I'm still learning all this database magic. I appreciate the feedback and you taking the time to check it out.

Who was more valuable to the 1962 Cubs? Ken Hubbs or Jim McAnany? Hubbs played 160 games at 2B, but was such a lousy hitter he racked up a -0.3 WAR. McAnany pinch hit 7 times and went 0/6 with a walk. He posted a -0.1 WAR. Relying solely on the computer, you'd be led to believe that McAnany was the better player, but we humans know better. Hubbs was no great shakes, but playing 160 games at 2b at a competent level (-4 fielding runs) is a hell of a lot more valuable than going 0/6 as a PH.

Fixed Alex Gordon and Jose Bautista, by the way. Yes, I had originally started from data through 2011. I thought I had gone back and updated the players who changed primary positions, but apparently that was an oversight. They'll all be cleared up when I do the adjustment above, though.

Cool, thanks. I'm just trying to make an awesome site. I knew the positions somewhat an issue. But I wanted to get what I had live before I iterated on that part of it (since it'll be a pretty massive calculation from multiple sources).

While I have you, here's what I want to do…

I've now got a Hall Rating value for each season. I'm going to get positions from the Lahman database and map them to each season, dividing that season's Hall Rating across the positions. This will give me the ability to:

1. Get a global primary position for the player based on VALUE not games played.
2. Get a primary position for each FRANCHISE.

That'll make things better all around. Of course, I should also add that I'm still learning all this database magic. I appreciate the feedback and you taking the time to check it out.

You might also want to tweak the all time team so that the bench isn't all OF (Red Sox), IF (Cubs), or C (Yankees). They might be the most valuable leftover players, but a team is not going to carry only 1 IF and no C on the bench. It's more interesting to know who the Red Sox second best catcher is rather than their 8th best OF.

Wow, Rico Brogna lists as the worst 1B in Phillies history, number 96 out of 96. In spite of two 100 RBI seasons. I don't know if I agree, but I like a system that spits out interesting "conclusions" like this.

Hard to see how he ranks 63 spots behind Travis Lee, however, who was a near identical player in his three seasons
OPS+
Lee: 82, 102, 96
Brogna: 88, 97, 95

It's great that you fixed Bautista. If you're still looking at the Jays, you might want to swap Shannon Stewart for Aaron Hill. They've got the same rating, but Stewart's buried as the 8th outfielder and the Jays have no backup infielders.

Hard to see how he ranks 63 spots behind Travis Lee, however, who was a near identical player in his three seasons
OPS+
Lee: 82, 102, 96
Brogna: 88, 97, 95

WAR (as a Phillie)

Brogna -3.7
Lee +1.7

Wasn't Brogna the one who his manager bragged saved them 100 runs a year with his glove?

Perhaps the value thing will correct this, but Smokey Joe Wood is not an OF. He's a pitcher. The all time Red Sox team should have Ruth and Wood as pitchers, but instead it has Ruth as an OF and Wood left out. Wood went 117-56 with a 150 ERA+ as a Red Sox pitcher, but because he played a bunch of games as a Indians OF, he's not a pitcher. The Sox staff includes instead Tex Hughson, who went 96-54 125 in fewer innings than Wood. Ironically, if Wood had been a little better pitcher, he would have made the Sox roster, as an OF.

OK, enough bashing.

OK, enough bashing.

(there's never enough)

Just needs a little tweaking with the logic. How about:

-Minimum 500 games played at a position for a team to qualify as starter for that team
-Every position needs at least one backup. To qualify as a backup, you need to have played at least ten games for the team at that position.
-5 starting pitchers = pitchers who started at least 150 games for the team, and started more games than they relieved during their career.
-5 relievers = pitchers with less starts than games they relieved.
-15 position players, 10 pitchers.

That would make the Cubs' roster:
c: Gabby Hartnett. Backup: Anson, Chance, Williamson
1b: Cap Anson. Backups: Banks, Hartnett, Williams, Hack, Chance, Grace
2b: Ryne Sandberg. Backup: Anson, Evers, Herman
3b: Ron Santo. Backups: Anson, Sandberg, Banks, Hack, Tinker, Williamson, Evers
SS: Ernie Banks. Backup: Santo, Tinker, Williamson, Evers
LF: Billy Williams. Backups: Anson, Banks
CF: George Gore. Backup: Sosa, Williams
RF: Sammy Sosa. Backup: Anson, Williams, Chance

Bench: Stan Hack, Frank Chance, Joe Tinker, Ned Williamson, Mark Grace, Johnny Evers, Billy Herman

SP1: Fergie Jenkins
SP2: Rick Reuschel
SP3: Mordecai Brown
SP4: Pete Alexander
SP5: Clark Griffith
RP1: Lee Smith
RP2: Bruce Sutter
RP3: Ryan Dempster
RP4: Carlos Marmol
RP5: Sean Marshall

I'd say skip the "primary position for a franchise", and instead break it apart by actual positions by franchise. Or at least as an option. Obviously it would be tough to get exact numbers in a lot of the older cases, but you could easily do rough approximations just by stuff like "percent of innings as a shortstop that season times overall value that season" or whatever.

I mean, assuming that the basic "Hall Rating" calculation is feasible, then it really does seem like it's saying that Babe Ruth was the Yankees' most productive right fielder ever AND the Yankees' most productive left fielder ever. That sort of thing would be neat to see, rather than just glomming it all together as him being a right fielder.

As long as we're nitpicking, I think the Padres look weird with Trevor Hoffman as their starting pitcher and all the real starters coming out of the bullpen.