His HoM candidacy? Frankly my dear electorate, I don't give a damn.

Actually, I haven't rung up his numbers yet, but it doesn't look promising.

I think that opposing outfielders played shallower for him than they did for any other batter, or for any batter now active. In particular, when he batted, the LF would come way, way in and hug the line - but there would still be some space between the LF and the 3B, since the 3B was in on the grass playing for the bunt.

He was certainly a good leadoff hitter, and his inclusion in the notorious Len Barker trade was one of several reasons that deal was such a disaster for Atlanta. In my offensive system, which should be friendly to his type, he comes in about level with Vada Pinson and Fielder Jones - and also Max Carey. Comparing him to Puckett is more of a peak versus career thing, with Puckett as the peak/prime guy. If Butler had been a better percentage base stealer, he'd be a step ahead of that group.

I think that opposing outfielders played shallower for him than they did for any other batter, or for any batter now active.

Jason Kendall begs to differ.

I liked Butler as a player even though he was a pretty bad basestealer. One of the best bunters of all time, I think. Teams knew he would probably try to bunt for a hit and it still worked a ridiculous amount of time. Also, a very headsy baserunner and more than dependable fielder. The Giants should have never let him go. Not a HOMer, but definitely in the Ray Lankford wing.

Butler had a stroke this weekend...

http://sports.yahoo.com/mlb/news?slug=ap-butlerstroke&prov=ap&type=lgns'

I liked Butler as a player even though he was a pretty bad basestealer.

68.4. Ouch, worse than I thought.

Best wishes to him. Very strange that he popped up here today.

Well, I've just run the numbers on Butler, and boy was I surprised -- my system (looking so far at WARP and WS but not yet at Dan R's WAR) sees him as a serious candidate, similar to but better than Bonds, Murphy, and Puckett! To give a sense of how this could be, here's how his 12-year offensive prime, 1984-95, compares to Puckett's 12 years.

First, EQA2, best to worst:

Puckett: .319, .317, .307, .307, .307, .304, .304, .295, .295, .292, .260, .244
Butler: .313, .311, .310, .304, .298, .298, .291, .284, .280, .280, .278, .276

These lead to WARP2 BRAA/BRAR totals of

Puckett: 342/574
Butler: 313/549

Next, FRAA/FRAR

Puckett: 48/288
Butler: 21/305

That's total BRAR+FRAR of

Puckett: 862
Butler: 854

Puckett is 1 win ahead.

Now, he is farther above average than Butler, but Butler closes the gap with durability and positional value.

Adjusting both players' 12-year primes for strikes, their games played are

Puckett, 1846 adj. games
Butler, 1856 adj. games

By defensive position, WARP totals their defensive playing time (unadjusted as)

Puckett, 267.4 g RF & LF / 1381.2 g CF
Butler, 1704.9 g CF

To these 12 prime seasons, Butler brings one more good full-time season (1983), and four part-time seasons of minimal value.

Am I being deceived by some strange statistical illusion to see him as comparable to a serious candidate like Puckett, or does this comparison stand up to further scrutiny?

(Win shares shows similarly, though it puts Puckett more definitely ahead.)

As I said, I couldn't have been more surprised at these numbers: Butler was not on my radar at all. So I'm not sure that they should be trusted. But I'd like to see them discussed!

After a quick glance I have Butler about equal to Puckett.

........WS/Top 3/Top 5 consecutive
Butler..307/81/127
Puckett.292/92/136

EQA has them almost identical: .283/.284.

As mediocre as Butler was stealing bases, Puckett was worse.

Butler - 558 steals at a 68% success rate
Puckett - 134 at 63%

Comparable to Puckett? Sure. Throw in comparable to Dawson while we're at it. Serious candidate? That I'm not so sure of. We're nearing the end of the usefulness of my yearly scaled RCAA charts, but I might as well run this one. Remember: this is offense only. But Butler was a pretty good defensive CF.

My system sees Butler as well superior to Puckett--but neither as anything resembling electable in my opinion.

The support for Puckett is baffling to me--despite getting a bonus for playing in low-stdev leagues, I see him with two excellent years (high All-Star/weak MVP candidate) in '88 and '92, two more All-Star type seasons in '86 and '89, and then little more than filler. He was a great CF his rookie year, and an above-average one as a sophomore, but basically average after that (Chris Dial's Zone Rating numbers make Puckett look rather bad in the field; FRAA and FWS are slightly positive). He didn't do anything particularly remarkable on the basepaths. His hitting was far from spectacular for an outfielder. His career was short. What is there to like about this guy?

For that matter, what makes Pucket so much better than Andy Van Slyke? He has about 130 games and 5 points of OPS+ on Van Slyke, but Van Slyke is OBP-heavier, hit into literally half as many double plays, and was an excellent basestealer, while Puckett's SB/CS actually hurt the Twins. For defensive reputation, Puckett had six Gold Gloves, Van Slyke 5; BP FRAA says both were 50-60 runs above average for their careers. Both had their biggest years in 1988 and 1992, and they were of similar value. Puckett and Van Slyke seem absolutely indistinguishable to me. Can someone enlighten me?

Of modern outfielders, Reggie Smith (200 more games than Puckett, 13 more OPS+ points than Puckett) seems *vastly* superior. So do guys I don't support like Bobby Bonds (70 more games and 6 more OPS+ points than Puckett, with much better baserunning and fewer DP's), Fred Lynn (200 more games and 6 more OPS+ points, although he didn't play CF), and Chet Lemon (200 more games than Puckett, 3 fewer OPS+ points but WAY better defense, full career in CF).

Butler is well above Puckett, but no better than Lemon.

My system's got Butler behind Puckett by a slight margin, but it sees neither one as a really serious candidate.

I've got the in/out line just above Jimmy Ryan right now (based on theoretical positional balance), and runners-up go like this:

Ryan
t-Bell
t-Averill
Browning
Wynn
Carey
Puckett
t-Ashburn
t-Berger
t-Murphy
(Bernie)
F Jones
H Wilson
Butler
t-Cedeno
t-Poles
t-R Thomas

CF is just plain stacked.

If I understand, that is the top ten CF in Eric Chalek's personal PHOM queue. Van Haltren is already in or Eric is EOGVH #1.
HOMers bold. Browning and Puckett --and Dawson, Duffy, and Leach-- are among the first twelve in the backlog, the group someone calls viable at this stage. If you think the first fourteen in the backlog are viable, down to a gap in the 2002 results, then Oms and Van Haltren make seven who are roughly half-career CFs, or more.

Ryan
t-Bell
t-Averill
Browning
Wynn
Carey
Puckett
t-Ashburn
t-Berger
t-Murphy

>My system sees Butler as well superior to Puckett

I figured there was probably somebody here who thought that.

There are no valid mathematical equations that can make it so, however. (Or Willie Randolph over Puckett, or Randolph over Mattingly). On this one I am with the neanderthals.

This of course is my neanderthalic opinion, having merely watched them but not done what Paul W. calls "original sabermetrics."

This of course is my neanderthalic opinion, having merely watched them

So what did you see? What were the qualitative differences between them, aside from their obviously and dramatic different profiles as hitters?

Brett Butler was certainly a nice ballplayer, but if Brett Butler why not Kenny Lofton. I mean, I have never looked at Lofton's numbers and the way it's going I may not have to until about 2020, but they seem like similar players and Lofton seems better. Maybe Lofton will be a HoFer and/or a HoMer someday (alongside buddy Omar Vizquel maybe). But I don't know anybody who thinks of Lofton (or Butler) as a HoFer or a HoMer.

Of Kirby Puckett, OTOH, Bill James says, Is he a HoFer? "Sure." He had a 62 percent chance of 3,000 hits at the time of his retirement, for what that is worth. James also ranks Puckett #8 and Butler #25 among CFers, though he says really nice stuff about Butler. Butler is "one of (James') all-time favorite players," he "hustled" and played "smart" and kept himself in "tremendous shape," and with those characteristics he was the 4th best "old" CF (over age 33) of all-time. Nice. Even if he has Puckett too high by a factor of 2 (IOW, if Puckett were only the #16 CF of all-time, which might be more realistic than #8), I think he's got Butler about right. And #16 still trumps #25 comfortably. Here are the numbers.

8. Puckett 281/32-31-29/136/25.5
25. Butler 295/27-27-26/124/21.6

For a peakster or a primester, there's not much to compare there. But looking closer.

Puckett 281/32-31-29-27-26-22-21-20-20-19-18-16
Butler 295/27-27-26-24-23-23-20-20-20-19-19-16-16

Butler catches Puckett in their 6th and 8th best years; before that it's all Kirby. And Butler earns those 14 extra career WS in his 13th year, after Puckett is forced to retire after 12 years. But the 4th best old CF of all-time only has that one additional year (a 13th) over Puckett with 10 or more WS. I mean, I can see where a career voter might prefer a guy with a 295-281 edge, but for a peak/prime voter, that's not nearly enough to overcome an 18 WS edge for Puckett over 5 years, which he maintains through their 10th year.

Then there's OPS+.

Puckett 123/151-38-37-31-30-30-29-19-19-18-(2 yrs < 100)
Butler 110/134-33-31-22-22-19-14-11-6-5-3-(3 yrs < 100)

All Kirby all the time. Why?

Puckett .318/.360/.477/123
Butler .290/.377/.376/110

Butler walked 700 more times than Kirby. Some might like that .017 OBA edge. But Kirby would have to go 0 for 900 to match Butler's career totals for hits, AB and BA. Kirby had more than 200 more XBH in those 900 fewer AB.

On SB, Kirby was a mediocre 134 with 67 CS. But what, then, can you say about Butler's 558 with 257 (count 'em) CS. His percentage is just barely better than Kirby's and I don't know that an extra 400 steals balances an extra 200 CS.

And for 9 years Kirby played ≥146 games and in 8 of those he played >150. The 10th year was 1994 and in 1995 he got hit in the face with a pitch. Butler had a similar streak of 11 years with ≥ 137 games and I think 10 of those years he played 150. But Kirby was just as durable in the 12 years he played.

Then, WS rates Kirby an A+ fielder and Butler a B+. The Baseball Encyclopedia has Kirby at +46 FR and Butler at -13. Kirby's range index is just 101 (Butler's is 99) but Kirby's throwing index is 146 to Butler's 100.

So I just don't see much to commend Butler. On Butler's strongest suits, Puckett is pretty much right there. On Kirby's strongest suits--hitting for power, throwing the ball--Butler is nowhere to be seen. Kirby was 5 WS better at his peak and Butler wasn't any better than Kirby until Kirby went blind.

That's what the numbers say to me, if I had never seen them play. But having seen them play, Kirby had 2 obvious advantages. The numbers don't lie. He hit for power and he threw the ball as well as almost any OF I've ever seen. He came up throwing and put the ball on a dime. His arm was strong enough but mainly it was just an insanely quick release. For those of you who have seen Torii Hunter, Torii is damn good, but Kirby was better. He climbed the fences like Torii (see 1991 World Series 6th game, not only did he win it with a walk off HR, but he sent it into extra innings by scaling the fence to catch a ball that had left the playing field; and it was far from the only time he did it). He caught the ball like Torii, and he threw it better. A lot better than Butler.

Those are all the things I look at and, like I say, if I had never seen them play it would be no contest. Having seen them play, it is no contest. I know some of you will say that you have a more sophisticated methodology, but sometimes it's hard to find the forest amongst the trees.

Rather than Butler, I think Fred Lynn is a better test for Kirby. I won't go through the whole thing but Lynn was better when he was healthy. He just was hardly ever healthy and, ironically (considering he died at an early age), Kirby was a freakin' horse for 12 years.

1995 he got hit in the face with a pitch.

That was a short season (144 G). There were only three games left when that happened. No durability issues in 1995 for Kirby.

I like Puckett a lot better than Butler, myself, but it seems silly in Hall of Merit voting for fans of any team to push for their guys in a sense of "I know what I saw, and even though I'm a fan, never mind."

Sunnyday and I have gotten along for "over a century" now, so I think he'll take this post with aplomb..

I hope.

Bill James gives Puckett six Gold Gloves --all legitimate top threes, where James expands the listed leaders from three to five in 1989.

Kirby Puckett and leading contemporary OFs by number of Bill James gold gloves
(contemporary: later than Garry Maddox, Amos Otis, Omar Moreno [5 each], earlier than Bernie Williams)
(column one, number of listings, expansion from 3 to 5 per league-season in 1989)
(column two, number of top three listings)
(parentheses, rank among the listed top three or top five)

7 6 Marquis Grissom (3-1-1-1-1-1-5)
6 6 Kirby Puckett (1-1-2-2-3-3)
6 6 Willie Wilson (2-1-2-3-2-2)
5 5 Andre Dawson (2-3-1-1-3)
5 4 Lance Johnson (1-2-4-2-1)
5 3 Devon White (1-1-2-5-5)

Howie, I have spoken on behalf of Kirby carelessly, as a fan. But the point of #16 is that, for me, the numbers more than back up my impressions. Obviously others use different numbers but those are mine and they make Kirby a solid HoMere. My other favorite Twin of all-time is Tony Oliva and I've been pretty objective there and elsewhere. I'm probably more of a Harmon Killebrew fan than some, but also less of a Rod Carew fan than some.

If a player (Butler?) has lots of bunt hit, his singles then had less base-advancing ability than the typical players, hence will be overrated by most stats like OPS, RC, LW, etc. which assume all singles are created equal, endowed by their perpetrators with the same abilities to drive runners in.

Brett Butler career high in RBI: 51.
Brett Butler career RBI per 27 outs, per bb-ref: 2.47.

Add in that leadoff hitters are generally under-leveraged to begin with (batting with fewest baserunners on), I'd say Butler's case is less than whatever WS or WARP claims it to be.

If a player (Butler?) has lots of bunt hit, his singles then had less base-advancing ability than the typical players, hence will be overrated by most stats like OPS, RC, LW, etc. which assume all singles are created equal, endowed by their perpetrators with the same abilities to drive runners in.

At the same time OPS undervalues OBP heavy players, and RC & LW tends to undervalue FAST baserunners who have greater ability to advance from 1st to 3rd on a single, score from 1st or 2nd than other runners.

In Butler's case I suspect the biases tend to even out, his singles did not have as much value in advancing runners than the average hitter's, but his being on base was somewhat mroe valuable than the average runner being on base.

BJ once took one of Butler's seasons (I think it was 1991- but it doesn't really matter) and mentioned that Butler got a lot of grief for his CS- he said ok, Butler is .296/.401/.343 he has no power and he was caught stealing 28 times. Lets use those 28 CS to erase 28 walks. What do you get? A .296/.377/.343 batter with 38 Sb and 0 cs, "and people wonder why he scores 100 runs every year".

If a player (Butler?) has lots of bunt hit, his singles then had less base-advancing ability than the typical players, hence will be overrated by most stats like OPS, RC, LW, etc.

Maybe his outs include more reached first base on error. There are three general reasons why I guess so. Among the infielders pitchers are less reliable fielders of ground balls and throwers to the bases, and they handle a high share of bunts. The lefty batter's drag bunt is a common type and it often provides a bad throwing angle, but official scorers are not generous. The bunt is a deliberate challenge by the batter, selected by those who run the first basepath more quickly than average (time pressure is universal) against those pitchers more likely to err.

For Jake Daubert, too, representing the class of batters with a now-incredible number of sacrifice bunts, his plate appearances may include many reached first base on errors. Some of the same reasons pertain.

Well, a CS is a bigger negative than walks are positives. But the general point holds.

Yes, faster players are undervalued by standard stats formulae. However, I suspect in Butler's extreme case, it doesn't quite even out. One year (1992) he had 40 (!) bunt base hits. His splits that season were

runners AB . H 2B 3B HR RBI BB HBP ROE BA OBP SLG OPS
Men On 192 .59 6 . 4 . 1 .37 .29 .. 0 ... 5 .307 .396 .396 .792
none on361 112 8 . 7 . 2 ..2 .41 .. 3 ... 6 .310 .421 .388 .809

Butler batted with a total of 118 men on first, 106 on 2nd, and 56 men on 3rd. He drove less than 13% (37 minus 1, divided by (118+106+56) ) of them. Eh. Not a lot of value in that 792 OPS. And his OBP with no one on is higher; I'd suggest it is likley that he had many bunt hits with none on, which are less valuable than hits with men on. Wow, I feel like I'm a cranky Joe Morgan, and that DOESN'T feel good.

I think Brett Butler, like many high OBP guys, is vastly underrated by most fans. But there is SOME value in team-dependent stats like RBI, especially if we can infer the reason why a person's RBI may be low.

Actually the avg would drop also. Turn the 28 CS into outs and assume the walks turned into outs. Turn the 38 SB into doubles (add to TB) and the final line is: .283/.363/.405.

Oops, .283/.363/.387, forgot to add outs to the slugging.

Maybe his outs include more reached first base on error. There are three general reasons why I guess so. Among the infielders pitchers are less reliable fielders of ground balls and throwers to the bases, and they handle a high share of bunts. The lefty batter's drag bunt is a common type and it often provides a bad throwing angle, but official scorers are not generous. The bunt is a deliberate challenge by the batter, selected by those who run the first basepath more quickly than average (time pressure is universal) against those pitchers more likely to err.

there is a forth reason being since if a player is fast there is less recovery time (time it takes to make the play at first) thus faster players would be more likley to reach on an error

True. That's true in general, even for a fast slugger who hits hard ground balls but never bunts --maybe Canseco or Bonds? The fielder may err because of time pressure. And I'm sure the infielder is more likely to err than with Boog Powell or Greg Luzinski at bat; they miss out on some errors made in haste for a faster runner and they also get thrown out on more unforced "errors" --boots and muffs that would be scored as errors if a faster runner were reaching first in time.

Brett Butler, Otis Nixon, and other speedster bunters do bunt so frequently that a better understanding of the bunt plays is important to understanding their game. For young Canseco, the speedster slugger who never bunts (i suppose, contrast Mickey Mantle), the effect of extra errors in comparison with Luzinski is likely to be statistically significant but baseballistically negligible.

Caught stealing would conceal many safe on error if the scoring were not generous. But the scoring is generous: unless the runner advances an extra base, it is a steal for the runner and against the catcher, not an error.

The bunt "for a hit" and the steal are similar plays: at discretion of the batter/runner, occurring in bulk for a few players, always under time pressure that induces hasty throws, commonly with angles that induce poor throws or hit runners (commonly, a good throw must be close to hitting the runner).

--
No doubt, my interest in this play is influenced by a small number of close plays including some scored errors on Otis Nixon in 1994 (his season in Boston) and by then-recent expressed opinion that it was stupid for him to make the last out of the World Series on a bunt. And one play by Pete Richert in World Series ~1969. And for the sacrifice bunt, one play by Joe Hoerner in the Zamboni game ~1972 (SLN@PHI). And one play by Nolan Ryan in the 1980 semifinal (HOU v PHI), maybe on a swinging bunt. Otis Nixon 1994 plus a small number of memorable plays.

I'm not sure what the point is about Butler giving up RBIs by bunting for hits. Did he often bunt for hits with runners in scoring position? That would certainly be surprising. A bunt single with the bases empty is just as valuable as a line-drive single (or a walk, for that matter). Or are you saying that because his basehitting ability depended so much on his bunting ability, that he was a worse hitter when he had to swing away? Then we'd expect to see it in his splits, with a lower AVG with runners in scoring position. Over his career, his average was .280 with RISP, compared to .290 overall, suggesting that there may have been a modest effect in that direction.

For 1992, it appears Butler just had bad luck with RISP, hitting .262/.347/.349 in 150 PA (compared to .309/.413/.391 overall). With runners only on first, on the other hand, he hit fantastic: .394/.487/.485 in 96 PA -- true, only 3 RBI (on 3 EBHs), but he did a great job in that situation of keeping the inning alive and moving the batter along. For his full career, he hit .307 (or .017 above his overall average) with runners just on first, which again compares to .280 (or .010 below his overall average) with RISP.

Tom sees Butler as doubly damned--low leverage, because he seldom hit with runners on base, and poor at driving them in when he did. I prefer to see the glass half (or three-fourths) full: a player whose job was to get on base and score runs and fulfilled that role very well. While I can see the leverage argument for a leadoff hitter like Soriano, who arguably would be more valuable using his power with more runners on base, for a high OBP, no-power hitter like Butler it seems the leverage argument may actually work the other way. His skill was getting on base, and by batting leadoff he was using that skill more often in situations (bases empty, no outs) where it was most valuable. It seems to me that batter leverage must be multidimensional, and that a low leverage situation for a power hitter (bases empty) may be a high leverage situation for an OBP specialist.

Butler reached base on error overall in 1.4% of his plate appearances, which doesn't seem to have been much different from the norms for his era. It would be interesting to see his ROE rate while bunting, but I don't know how to look up that split.

oops--"keeping the inning alive and moving the runner along..."

Somebody (maybe Bill James) said Butler scored 100 runs every year. Clarification: He did so 6 times. Not bad, certainly, but not every year.

I prefer to see the glass half (or three-fourths) full: a player whose job was to get on base and score runs and fulfilled that role very well. While I can see the leverage argument for a leadoff hitter like Soriano, who arguably would be more valuable using his power with more runners on base, for a high OBP, no-power hitter like Butler it seems the leverage argument may actually work the other way.

That's a good point, Brent. The reverse argument would apply to someone with an opposite profile, like Andre Dawson (who gets raked over the coals by many metric oriented analysis for being a low OBP, power hitter situated in the spot where he is best at, getting the runners like a Butler or a Nixon in).

someone with an opposite profile, like Andre Dawson (who gets raked over the coals by many metric oriented analysis for being a low OBP, power hitter situated in the spot where he is best at, getting the runners like a Butler or a Nixon in).

Dawson isn't really the opposite pole, nor is Juan Gonzalez, another slugger who is deservedly famous for few bases on balls. Why? Because these players, while great out-makers among HOM candidates, are actually about average. It seems to me that Brett Butler is close to one pole among star ballplayers and the other pole is Dave Kingman or (young) Bo Jackson. They are outmakers so extreme that their managers would consider batting positions seven and eight (even nine in the DH league), ahead of no one but the outmakers who lack power. Another at the opposite pole is Joe Carter, in his prime, when he was still well above league average in slugging.

Brett Butler is speedy, so Dave Kingman gets the nod as his opposite.
Joe Carter is speedy, so Mike Hargrove rather than Butler gets the nod.

ok, good answer, sunnyday....

Dawson isn't really the opposite pole, nor is Juan Gonzalez, another slugger who is deservedly famous for few bases on balls. Why? Because these players, while great out-makers among HOM candidates, are actually about average. It seems to me that Brett Butler is close to one pole among star ballplayers and the other pole is Dave Kingman or (young) Bo Jackson.

Butler's SLG was 5% below average. Dawson's OBP was 3% below. Kingman's OBP was 8% below.

As full-career averages, relying on baseball-reference, I get -6% -3% -9%.
But I actually found good players with on-base averages extremely high and extremely low relative to their slugging averages. Thus Hargrove, slugging 0% below league-average, but 5 points below his on-base. Even if done systematically, rather than by looking up a few extreme players, I'm not sure that is a good way to identify players whose value is most augmented by batting first and last in a rally, which was sort of the point (although I didn't say so).

Looks like I'm still on Saint Louis time. Good night.