Is the WAR in this article based on number of games at first base, or number of games overall? Thome, at least, should be docked his years at DH.

Number of games overall. Is Bagwell the largest saber/mainstream disconnect of any recent player?

Finally, we know just what WAR is good for. (Say it again...!)

Is Bagwell the largest saber/mainstream disconnect of any recent player?

Well, Bagwell did very well in MVP voting, and despite playing on a team that never won a playoff series till his career with them had ended, he had a pretty high national profile when paired with Biggio, another great player. My father, who couldn't be less sabermetric, would get Who's Who in Baseball every spring, turn to Bagwell's page, and start reading me his Runs and RBI totals with reverence. If Bagwell has been forgotten by the MSM, it may be a curious case of them devaluing those sheer counting stats, which in Bagwell's case are superb.

But another reason he's being forgotten (if he really is; I mean, I think there's a natural lull while someone's retired but in the five-year HOF waiting period when people forget all about him) may just be that there are so many guys from his era with similar numbers. Manny, Sheff, Edgar, Jason Giambi, Thome, Helton, Berkman, Brian Giles, Delgado ... even if Bagwell was a shade better than that crowd, it's a big crowd. I imagine that Bagwell will do very well in HOF voting when his name comes up, though, unless pictures of him sucking down Mr Burns's Nerve Tonic surface.

As a rate-statter, the FA has Albert ahead of Lou Gehrig, 8.8 to 8.7 per 700 PA's. Wow. There's the "no deline phase yet"
caveat I suppose, but that's still impressive.

There's always been Lou Gehrig at 1B, then everybody else, in my book.

Number of games overall. Is Bagwell the largest saber/mainstream disconnect of any recent player?

what does that mean? from both perspectives he's a "first" ballot hofer. He may be a tad underrated mainstream because of the lack of publicity from the east coast media, but the fact that he played for only one team is a big plus. Even without the steroids issues or anything else, almost everyone I know considers Bagwell to be a hofer, a better choice than Palmiero and most of the mainstream guys will argue as a better choice than Frank Thomas(his "twin")due to superior defense and better character.

It's similar to why I think Biggio gets in easier than Roberto Alomar, the one team is a strong tie-breaker.

Is the WAR in this article based on number of games at first base, or number of games overall? Thome, at least, should be docked his years at DH.

It's career WAR at all positions, and Justin is counting players as 1B as long as they played more 1B than anywhere else. Thome is docked for his years as a DH, in that he gets a harsher position adjustment when he's a DH.

I thought Pete Rose's middle name was Edward, not Herman.

I thought he had two middle names, Charles & Hustle.

Is Bagwell the only player who's ever been helped by a baseball strike (other than in a financial sense, of course)? People are going to forget that he won the MVP because his injury just happened to coincid with the premature end of the season.

Is Bagwell the only player who's ever been helped by a baseball strike (other than in a financial sense, of course)? People are going to forget that he won the MVP because his injury just happened to coincid with the premature end of the season.

It's not a player, but the 1981 Reds had a better overall record than the Dodgers. But because Cincinnati didn't win either half of the season, they didn't go to the playoffs (they were 0.5 games behind the Dodgers in the first half, the result of playing one fewer game). The Dodgers ended up winning the World Series when, without the strike, they may not have even made the playoffs.

Interesting to see Jake Beckley clock in at #17. For those who follow our Hall of Merit discussion, we've inducted every eligible first basemen who ranks in the top 25.

took plenty of walks before it was fashionable to do so

Pet peeve of mine: I wish people would stop writing, saying, and thinking things like this. The notion that drawing a lot of walks was ever "out of fashion" or not highly-valued is utterly, historically false. The notion that the concept of OBP was discovered by Bill James or some other modern-day sabermetrician is utterly, historically false.

I just remember reading a few articles in SI/ESPN/whatever where Bagwell gets talked about as a tough case, with the tone being largely negative, because of his comparatively poor overall career counting stats; whereas he would actually be one of the best players not in the Hall if he's rejected (according to WAR, the only people not in the Hall with more career value are ARod and Bonds.)

The notion that drawing a lot of walks was ever "out of fashion" or not highly-valued is utterly, historically false.

But Steve, haven't you noticed that walk rates have simply SKYROCKETED since Moneyball came out? :-)

But I'll cut the author a little slack since he did say "fashionable." I think it's fair to say that the importance of walks is more widely recognized than it used to be and that low-BA, high-K, high-BB sluggers take less crap than they used to

But, yeah, McCovey was TEH FEAR.

I don't see much point in lists like this -- seems a straight accounting exercise. And again I'll say it's simply silly to include guys like Rose and Carew on this list ... and I'll assume there's a DH list coming which will be silly for lots of reasons. Rate Carew as a 1B for his time at 1B, rate him as a 2B for his time at 2B, rate him overall and compare to all others, rate him as a multi-position player and compare him to other multi-position players. This notion that we have to assign players to one and only one position for these sorts of comparison purposes is simply lazy.

Also, one assumes that this list would look a good bit different if it used WAA (which I think is more appropriate for "best of all-time" comparisons anyway) instead of WAR. Murray & Palmeiro obviously benefit from this.

Still I'm surprised Thome is ahead of McCovey in fewer PA. Also surprised that Mize is that high while Allen doesn't even make the list.

@Eamus, it is, and I'm embarrassed. :)

@Steve, fair enough.

@Walt, I like the position-by-position breakdown because it provides a convenient way to look back at player careers. To me, it's more approachable than looking at the huge overall rankings. I didn't want to break apart the careers of players within each position because you then miss out on something of how great Rose, Carew, Banks, etc, actually were. I think there are compelling arguments for doing it as you describe, but I prefer doing it this way. :)

Similarly, I completely understand the arguments for using WAA in a "greatness" project. Average is a very nice baseline. And someone else is welcome to do that--Rally's data is easy to parse out into WAA, as he reports replacement adjustment in a separate column. But there's value in all those years of playing time for Murray, Palmeiro, and similar players (Anson?), and that's part of what WAR recognizes. I did make the explicit statement in the catcher piece that I'm reporting value here, not necessarily greatness. WAR/700 PA gets a bit more at greatness, perhaps, as do the WAR plots, but technically that's not what these rankings show.

Re: Mize vs. Allen--it's the fielding. Mize is +18 runs in fielding (by JAARF), whereas Allen is -109 runs (by Total Zone). Allen gains 29 of those runs back with a smaller position adjustment (due to lots of time at 3B and OF), but that's still about a 100-run difference between them.
-j

@Steve, fair enough.

< fist bump >

and I'll assume there's a DH list coming which will be silly for lots of reasons.

Apparently not, based on the list in the series included at the bottom of TFA. So Frank Thomas, the poor fellow, will go unranked.

The notion that drawing a lot of walks was ever "out of fashion" or not highly-valued is utterly, historically false.

Balderdash! In my day, a "real man" had to "slug" his way onto the base-paths, to prove his mettle! "Walks" were for limp-wrists, anarchists and pickaninnies! Why, I once drew "four balls" against Lionel "No-Face" McErnany, and the manager was so enraged he whipped me within an inch of my life! No, sir, I "swang" at every pitch after that, no matter if they were ten feet outside! Yessiree...!

There will be a (abbreviated) look at DH's. -j

@ Eamus - or the '81 Reds might have lost to the Cardinals who had the best record in the NL East, but stayed home due to Kuhn's folly.

About Eddie Murray (and I'll admit that we're talking about my childhood favorite player) - I think he gets an exaggerated rep for having extended his career as a DH. He played 371 games as a DH after 35. A couple of things skew his rep, I think. One, Eddie left public consciousness after '86 as the Oriole Dynasty clearly ended, and when he emerged with Cleveland in '94-'95 and for his chase to 500, he was a DH. For anyone then under 18 & others with short memories, the Cleveland Eddie set his identity. (and, yes in those pre-universal cable & internet days, one could play in big markets and be easily out of mind.). Secondly, Eddie played 111 games at DH as a rookie which inflates his career numbers. Eddie moved through the minors and somewhat surprised the Orioles by being ready to break the camp where he'd turned 21 with the club. While his bat was ready, his defense was still unclear, Weaver toyed with ideas of LF and being the heir to Brooks. I guess those early games at DH extended his career at the other end, but I don't think that's what folks mean when they somewhat pejoritively speak of the length of Eddie's career.

The man played more games at 1b than anyone and deserves credit for that. Though "credit" for finding himself at the low-end of the defensive spectrum at 21 should be scant. (Albeit, in all likelihood, judging from his ability at first, Murray may well have been a passable 3b for some time - no shame in being judged inferior to DeCinces.) But Murray places so high on all-time GP lists by virtue of durability as well. From 21-40 Murray missed as many games due to work-stoppage as injury/manager's decision.

Comparison of this list to the Hall of Merit ranking vote:



The HoM ranked players by the value of their entire career, at the position at which they accumulated the plurality of their value, which may not be the same as the plurality of playing time (note the definition differences with Carew, Banks, and Rose). And we included Negro League players in the list, which this article doesn't. Given that, our biggest disagreements seem to be over Mize and Greenberg. Also, this list doesn't have Sisler in the top 25. We did elect him but his election was sharply controversial with us.

But Jake Beckley was also a controversial choice with us. He stood out as a candidate for career value voters - extremely long career, unusually low peak for such a long career. His name is a sign that this is a career-leaning list; other signs of that are the relatively high rankings of Murray and Palmiero and the relatively low rankings of Greenberg and Mize.

Given that, our biggest disagreements seem to be over Mize and Greenberg.

War credit.

The notion that drawing a lot of walks was ever "out of fashion" or not highly-valued is utterly, historically false.

Paraphasing-out here

Gary Cohen brought this up to Keith Hernandez the other day...about how nobody paid any attention to OBP when Keith was playing.

Naturally, Keith said HE looked at it all the time and said that "being a three hole hitter I knew my job was to get on base any way possible and let the big boppers drive me in."

Cohen then asked him if could would he go back and change anything about his approach to hitting.

Keith sez..."Yeah, I regret walking so much."

~~(!!!!)~~

I don't remember the rest as I was half way through feasting on my mostly surprised cat by then.

Also, one assumes that this list would look a good bit different if it used WAA (which I think is more appropriate for "best of all-time" comparisons anyway) instead of WAR. Murray & Palmeiro obviously benefit from this.

Absolutely. Which is how you get such complete nonsense as a list that ranks Carlton Fisk Gary Carter ahead of Yogi Berra, while the author prates about how Berra was overrated because he's a Yankee. It is hard to take him seriously after that.

Absolutely. Which is how you get such complete nonsense as a list that ranks Carlton Fisk Gary Carter ahead of Yogi Berra, while the author prates about how Berra was overrated because he's a Yankee. It is hard to take him seriously after that.

Yup, 100% agree. He also has Carter and Rodriguez as having much more defensive value than Bench and Berra which I find very hard to believe. Bench and Berra were both known to be superlative C's.

So.. that Lou Gehrig guys' pretty good ehe.

Yup, 100% agree. He also has Carter and Rodriguez as having much more defensive value than Bench and Berra which I find very hard to believe. Bench and Berra were both known to be superlative C's.

Bench yes, but Berra's early part of his career he was a known less than good catcher, he improved quickly though. I still don't know if Irod takes a back seat to Bench though. Carter I seriously doubt enters the discussion except for his longevity.

Roger Connor was only ranked #22 by Bill James...what gives!

Heavily dinged in the subjective side of the rankings. 10th in WS/162, 6th in WS and his peak ranks well.

If anything he's hardened his opinion, now arguing that 19th century baseball should not be considered major league baseball. Talent pool was far too shallow.

@OCF, thanks for the list. I like the use of value to determine position better than what I'm doing, but I've already started the series so I'm going to continue as-is. FWIW, Sisler shows up 30th on "my" list of 1B's by WAR.

@Srul & snapper, the overrated quip on Berra was more a joke than anything else. If you read the entry again, I praised the hell out of the guy. I like to make fun of Yankees a bit now and then, because it makes Yankee fans mad. And there is some truth to it: like it or not, when someone's a Yankee legend, they get more attention in baseball's collective conscience than other comparable players.

As for WAR vs. WAA, as I said above, it depends on what one is trying to evaluate. Vs. average is a great baseline, and for some kinds of questions, it's better than anything else. But replacement also is a relevant and important baseline. Fisk played for a long, long time, was outstanding for a few years, and was decent for much of his career. There's a lot of value to teams in that career.

Perhaps when I finish the series, I'll post top-25 lists based on WAA instead of WAR for each position. It would be interesting to see how much everything changes. I'm sure a fair bit will. For now, though, I'm sticking to what I'm doing.

Also, I think you're making a mistake to discount Carter and Rodriguez's fielding contributions. They were exceptional (IRod's still playing, of course, but his arm isn't what it once was). That said, catcher defense is a hard thing to measure quantitatively, so rating them above Bench might be an error. You should note that we don't have even the basic fielding data that we have on the other three for Berra prior to 1955, so there's legitimate reason to think that he's missing a fair bit of his value in these rankings. We do the best we can with the data we have.
-j

Bench yes, but Berra's early part of his career he was a known less than good catcher, he improved quickly though. I still don't know if Irod takes a back seat to Bench though. Carter I seriously doubt enters the discussion except for his longevity.

Very early on Berra was poor, before "Bill Dickey learned him everything he knew". Bench's first 2 seasons have some pretty hideous PB/WP #'s if you check it out. So, he may not have been so great right away either.

Looking at the available stats (only partial before 1954) Berra is superior to Bench in CS%, PB and WP. Looking at the pitching staffs he caught, I can't imagine anyone arguing that he didn't call a great game.

I don't think Berra is second to anyone as a defensive C. He's superior to even IRod on PB/WP and equal in CS% (although the game was somewhat different). To me, he and Bench are a flatfooted tie as #1 C ever.

You should note that we don't have even the basic fielding data that we have on the other three for Berra prior to 1955, so there's legitimate reason to think that he's missing a fair bit of his value in these rankings. We do the best we can with the data we have.

You don't prorate his stats at all? It's hard to believe he'd be much worse when he had youth on his side.

you then miss out on something of how great Rose, Carew, Banks, etc, actually were.

I'd argue you've badly mis-characterized their careers. Carew, as a 1B, is nothing special. If you set his career replacement level as a 1B, he won't make your list (I wouldn't think ... not that a 131 career OPS+ would suck at 1B). He's a WAR (or WAA) stud mainly because of the years he was posting a 140-150 OPS+ at 2B. This all goes double for Banks -- as a 1B he was league average. His WAR is from the 8 seasons he was the first incarnation of AROD. Rose played less than 1/3 of his games at 1B ... oh no, you're trying to seperate LF and RF aren't you? Silly boy. :-)

Similarly, I completely understand the arguments for using WAA in a "greatness" project. Average is a very nice baseline. And someone else is welcome to do that--Rally's data is easy to parse out into WAA, as he reports replacement adjustment in a separate column. But there's value in all those years of playing time for Murray, Palmeiro, and similar players (Anson?), and that's part of what WAR recognizes.

Sure ... but you said you're after value. It is absolutely absurd to assume that Eddie Murray (or whoever) would have been replaced by a replacement-level 1B for 13,000 PA. On average his teams would have had an average 1B over those 13,000 PA. His value, be it peak or career, is relative to average.

(and that's before we even get into the fact that the average level actually exists while replacement level is a guesstimate ... except maybe in some of Dan R's work ... and that replacement-level makes no more sense as "zero" than using true zero ... e.g. 0 RC)

Replacement level is fine for a single season under the thought experiment of "if Murray got hit by a bus on his way to opening day, the O's would have had to replace him with whatever they could find." Replacement level might even be fine if what you're going for is a career-level "was he worth the money he was paid" sort of measure. But, in general, for multi-year comparisons, WAR just makes no sense to me.

Now I have no problem if you want to say something like "players A and B were 40 wins above average, but A played 500 more games and that longer career means he was the more valuable player" -- I'd probably even agree with that.

It's hard to believe he'd be much worse when he had youth on his side.

We're talking Berra right. He was very raw. I have a quote some place by Vic Raschi complaining that Berra wouldn't call anything but fastballs with a fast runner on.

Also look at the 1947 World Series. DOdgers ran him out of the catcher position. Bucky Harris lost confidence in his ability to contain the Dodger running game.

Yes, he had a brutal series with the bat -- making it easier to start guys with a better defensive rep. But Berra started game 7 in right. In other words Harris hadn't lost confidence in his bat.

I have a quote some place by Vic Raschi complaining that Berra wouldn't call anything but fastballs with a fast runner on.

You can find similar quotes about the young I-Rod, and I'd be willing to bet you could find people who would say the same thing about almost every young catchers with a good arm.

@snapper, I pretty much made the decision to not modify any of Rally's ratings when doing this series--essentially, my goal is to explore the best players of all time according to his data. It's not that they're sacred or something, it's just that I want to spend my time looking at baseball history rather than trying to improve what he's done.

That said, you're welcome to try to pro-rate Berra's CS run values if you wish to do so, however. Rally only released an overall catcher rating, but he might be willing to separate out the SB/CS from the PB/WP/Error data for the retrosheet years. This would let you pro-rate. Or you can calculate them yourself from Bref, retrosheet, or whatever source you prefer.

As you said, Berra may have changed a lot in the first few years, so it's a bit dangerous to do. But even then, it might be a better estimate of his defensive prowess than what we have. I'd be interested to see the data.

@Walt,

I'd argue you've badly mis-characterized their careers. Carew, as a 1B, is nothing special. If you set his career replacement level as a 1B, he won't make your list (I wouldn't think ... not that a 131 career OPS+ would suck at 1B). He's a WAR (or WAA) stud mainly because of the years he was posting a 140-150 OPS+ at 2B.

For my series (and you're welcome to do your own series by your own rules), I wanted to look at the entirety of a player's career, regardless of where they were playing. The position classification is more secondary--it's a convenient way to separate out the top-500 list Rally posted into manageable increments, and has the side merit of allowing some looks at the "best" at a given position over time. If I would have thought of it, I probably would have preferred to use value at positions (e.g. WAR as a 2B vs. WAR as a 1B in Carew's case) as the determinant of career position. But I didn't think of it, and am already ~20 hours into this series, so I'll stick with what I have. It's easy enough for someone to look up a player's WAR and see where they would rank in these lists had they been included. We've done this very thing for Thomas, for example, at BtB. It's not a big deal.

Rose played less than 1/3 of his games at 1B ... oh no, you're trying to seperate LF and RF aren't you? Silly boy. :-)

LF and RF are two different positions. Are they more closely related in required skillset than 1B vs. LF or RF? Sure. But they're still two different positions. As I said above, this is how I decided to do my series. You're welcome to do your own series with your own rules.

Looks like Hall of Merit concurred on that point (they listed him as a RF, not OF), so I'm not alone in this.

...I think your last little quip in the quote was rather unnecessary. I'm happy to talk about this stuff, but please be polite.

Replacement level is fine for a single season under the thought experiment of "if Murray got hit by a bus on his way to opening day, the O's would have had to replace him with whatever they could find." Replacement level might even be fine if what you're going for is a career-level "was he worth the money he was paid" sort of measure.

Technically, that's not what WAR does either, because of chaining effects within teams. I increasingly think of it more as value above the minimum performance a big leaguer can provide without assuring himself of a demotion. Colin Wyers had a nice post on this at THT a few weeks back. That definition works the best to me of any others I've seen, and I'm reasonably well read on this topic.

But, in general, for multi-year comparisons, WAR just makes no sense to me.

My take on it is this: Wins Above Replacement is an established, well-vetted method of assessing player value. It corresponds in a fairly linear fashion to player salaries, which is a good check.

It sounds like you're ok with using it to evaluate a player season, at least when answering certain kinds of questions. I'm simply defining a player's career value as the sum of his seasonal values. That might not be the way you prefer to define it, but that's ok.

Now I have no problem if you want to say something like "players A and B were 40 wins above average, but A played 500 more games and that longer career means he was the more valuable player" -- I'd probably even agree with that.

But what about player A with 39 WAA vs a player B with 40 WAA, but player A played 500 more games?

WAR gives us a tool to make that call. It's an intermediate between WAA and an absolute stat like wRC. It is sort of a compromise measure of value between rewarding performance (which is largely, though not entirely, what WAA does) and rewarding playing time (which is largely, though not entirely, what wRC does). Is the baseline more nebulous than WAA's? Absolutely. But it's not purely imaginative either, and works well when answering a variety of questions.

And let me re-iterate: just because someone ranks above someone else in WAR does not mean they were the "better" player. I think it probably does mean that they provided more value (as I'm defining it) to their teams over their careers. But I still might take the player with a shorter, higher peak career over the player with the longer, lower peak career, even if the longer, lower peak career player had a higher WAR. Career WAR's part of the equation, in my mind, but it's not everything.

I think in all likelihood, we're not going to agree on this. That's too bad, but it happens. :)
-j

I think in all likelihood, we're not going to agree on this. That's too bad, but it happens. :)

So future generations are still going to have to argue about the relative merits of ballplayers in barrooms, on internet boards, and whatever other forums for discussion are available to them? That sucks.

If Rose is a 1Bman by these rules (not arguin whether the rules are the best or not, just for consistency), then the list completely misses the best first baseman ever.

Let me pick my all-star team, and it's very clear who gets to play first, and it ain't Lou Gehrig.

Stan the Man.

@TomH, Good point with Musial. I don't have good outfield position data on him in my database. It looks like I'll stick him in left field, but perhaps I should have included him at 1B. He would rank at the top of the list--128 to Gehrig's 118 WAR.

For fun, here's a WAR graph. Gehrig did have the stronger peak, so I probably would still take him over Musial if I got to choose one of them in their best 10 years. But Musial is just amazing in his longevity...and leaves a pretty decent player in Foxx in the dust every season of his career.
-j

Dale Stephenson did something very similar for peak. He made the decision to list players where they accumulated the plurality of their value. I think his decision works a lot better than the one taken in this series. (The other thing he did was to list percentage of playing time at the position listed and at other positions.)

Makes Stan Musial a RF, but that's no big deal.