Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Hall of Merit > Discussion
Hall of Merit
— A Look at Baseball's All-Time Best

Sunday, May 28, 2006

Bill Mazeroski

Eligible in 1978.

John (You Can Call Me Grandma) Murphy Posted: May 28, 2006 at 06:54 PM | 52 comment(s) Login to Bookmark
  Related News:

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

   1. John (You Can Call Me Grandma) Murphy Posted: May 28, 2006 at 07:14 PM (#2041248)
Okay, the fielding was great, but...
   2. DavidFoss Posted: May 29, 2006 at 01:08 AM (#2041769)
With the caveat that you can't go into the HOM on fielding alone and that I have no plans of voting for him, I'd still like to hear more about that amazing glove of his.

Stats guys have looked at his absurd DP totals and concluded they are indeed absurdly great -- not an artifact of his teams/staffs.

Anyone seen any good footage of his style? Did he have some sort of signature move? How was he able to do it? Why haven't we seen anyone dominate the DP since?
   3. Mike Emeigh Posted: May 29, 2006 at 01:27 AM (#2041783)
Anyone seen any good footage of his style? Did he have some sort of signature move? How was he able to do it? Why haven't we seen anyone dominate the DP since?


Mazeroski had a quick release - the ball barely touched his glove before it was off to first base. Plus, he didn't back off from anyone on the pivot - none of this "neighborhood" stuff for Maz - and he was almost impossible to dislodge.

I realize that the HOM is based on actual performance, and not "what-might-have-beens". Nonetheless, I will point out that Mazeroski was a relatively slow right-handed pull hitter, playing in a park which had a 365' left-field line. The park took away home runs and (very likely) some doubles as well. In a different home park, he'd have delivered 25-30 doubles and 15-20 HRs a season consistently, IMO.

Mazeroski was not making huge numbers of plays, however, considering the ground ball totals agianst the Pirates (where I've had data to evaluate it, that is) - almost all of his advantage in fielding stats comes from his extra DPs.

-- MWE
   4. Mark Shirk (jsch) Posted: May 29, 2006 at 04:28 AM (#2041952)
I take it that Emeigh is a Pirates fan?
   5. rawagman Posted: May 29, 2006 at 08:56 AM (#2042079)
I think Mazeroski need to be compared to Nellie Fox, primarily.
Both 2B's known primarily for their glove, with below average (but much better than replacement level) bats.
Fox is right below my ballot. If Maz is above in my system, he has a chance to ballot. If below, he will not.
   6. John (You Can Call Me Grandma) Murphy Posted: May 29, 2006 at 11:39 AM (#2042112)
Fox is right below my ballot. If Maz is above in my system, he has a chance to ballot. If below, he will not.

If Fox is not going in, than Maz certainly shouldn't, IMO.
   7. Howie Menckel Posted: May 29, 2006 at 01:52 PM (#2042178)
Wow, never once with even a league-average OPS+.

Now, he had six OPS+s in the 90s which as a 2B is really good - and as a GREAT fielding 2B is even better.

As for Fox vs Maz, it reminds me a little of Sisler vs Banks. Maybe some will downgrade Banks or Maz for the comparison, but one could just as easily upgrade Fox and Sisler instead.

Maz is only worth a damn from 1957-67 - but Fox is only worth that level, basically, from 1951-60 and 1964 (though he has a better argument for some lesser years than Maz does).
Of course, Fox has a 124-117-114-107-105 in there. When you're fielding at these levels, topping 100 OPS+ is just remarkably valuable.

Hmm, maybe this does make me think about adding Fox to the bottom of my ballot.
   8. rawagman Posted: May 29, 2006 at 01:56 PM (#2042182)
Maz is only worth a damn from 1957-67 - but Fox is only worth that level, basically, from 1951-60 and 1964


That's 11 years for each of them.
   9. yest Posted: May 29, 2006 at 05:59 PM (#2042371)
Both 2B's known primarily for their glove, with below average (but much better than replacement level) bats.

how many below average led his leauge in hits 4 times
   10. yest Posted: May 29, 2006 at 05:59 PM (#2042375)
Both 2B's known primarily for their glove, with below average (but much better than replacement level) bats.

how many below average led his leauge in hits 4 times
   11. favre Posted: May 29, 2006 at 08:03 PM (#2042555)
Bill Mazeroski: 8379 PA, 84 OPS+
Nellie Fox: 10,349 PA, 94 OPS+

And OPS doesn't tell the whole story. Seasons with OBP higher than the league:

Fox: 7 ('51, '54-55, '57-60)
Mazersoki: 0

Mazeroski never had a single season where his OBP was at or above league average; the closest he came was '68 (304/309). Fox's career OBP was higher than average (348/339).
   12. DavidFoss Posted: May 29, 2006 at 09:50 PM (#2042627)
how many below average led his leauge in hits 4 times

Good question. How many 4 time hit-crown players have a career OPS+ of under 100. Fox's 1952 season wasn't that good by any standard.
   13. Cblau Posted: May 30, 2006 at 02:16 AM (#2043206)
Fox is it. I guess Harvey Kuenn came closest to matching him, with a career OPS+ of 108. Everyone else who led their (Major) league in hits at least 4 times was a good hitter.
   14. yest Posted: May 30, 2006 at 10:17 AM (#2043665)
OPS+ underates OBP esspeshilly in the 50's where slugging went way up
   15. jingoist Posted: May 30, 2006 at 04:03 PM (#2043815)
MWE got Maz's skill set correctly.
As a Pirate fan from the mid-50's til now he was clearly the best Pirate 2b-man these past 50+ years. Pittsburgh fans vaulted Mazeroski further into stardom than he perhaps deserved due to his walk-off HR in the 7th game of 1960 WS but he was a stead contributor at 2Base for 12 years.
Maz played almost every day during his age 20-30 peak; he went down hill quickly thereafter and was toast by age 33.
While his glove was Fox's equal neither his fielding range nor hitting was.

Fox is definately one of the top 250 players of all-time; Maz is on the outside looking in.
   16. Dr. Chaleeko Posted: May 30, 2006 at 05:49 PM (#2043930)
he was clearly the best Pirate 2b-man these past 50+ years
You mean better than Chico Lind, Carlos Garcia, and Warren Morris!?! No way!!!! ; )

On the other hand, I'd be interested to see whether Johnny Ray's bat was as good in his (what?) six years as Maz's was in his best six Pburgh years.

Now, he had six OPS+s in the 90s which as a 2B is really good - and as a GREAT fielding 2B is even better.

This is an interesting theoretical question of value in disguise! While we often talk about the defensive demands of a position being much larger or smaller (SS v. RF for instance) viz how much offensive performance we expect from the position. But what about when we already know the guy's position and, therefore, his expected offensive performance? Is it logical to say that Maz's performance is excellent for an excellent defender? Or is it more logical to say that Maz's defense and his offense in comparison to one another and to all 2Bs are closer to independent variables than dependents?

That is, once you are talking about guys at the same position, as defense improves do we reasonably expect offense to slacken? If you say Ozzie Smith versus Vern Stephens or Ray Dandridge versus Hojo I might feel compelled to agree. But i could counter and possibly spoil the argument with Ryne Sandberg v. Larry Doyle or Brooks Robinson v. Toby Harrah, or A-Rod v. Derek Jeter (Yankee fans, zip it) or Johnny Bench v. Javy Lopez.

So I think arguing Maz's case by saying he's got good years for a great 2B is possibly not the most accurate way to describe him relative to his positional peers.


Also, to support MWE's point about Maz's release, he is famously known as "No Hands."
   17. vortex of dissipation Posted: May 30, 2006 at 06:02 PM (#2043950)
The best anecdote I ever heard about Maz' hands came from Frank White (no slouch defensively himself). Many years after Mazeroski's retirement, a reporter asked White about the best DP pivot he'd ever seen. He replied that Mazeroski was by far the best he'd ever seen, and when the reporter asked him about seeing him play, White answered that he'd never actually seen him play. He'd seen him give instructional clinics for kids earlier that year, and his mind was boggled by how quick Mazeroski's release still was, even though he hadn't played for some time...
   18. sunnyday2 Posted: May 30, 2006 at 06:14 PM (#2043963)
Re. Maz, certainly it is more informative to say he was an average (peak/prime) to below average (shoulders/career) 2B on offense than it is to say he was a below average offensive player.

No question he was an above average 2B--ie. further above average on defense than he was (among 2Bs) below average on offense.

Still Fox is also way above average 2B on defense, and at least average or slightly above average on offense. Was he "more better" than Maz on offense than Maz was "more better" on defense? I'd say yes but it's not a huge or self-evident difference.

When the Twins won the WS in '87 and '91 there were those who said the Twins MVP (in the aggregate across the 2 seasons) was Greg Gagne, because (they said) Gagne made a below average pitching staff into an above average pitching staff pretty much single-handedly. Now, maybe that's true, maybe that's hyperbole. But let's say it's true for the sake of argument. How bad of a hitter does a guy have to be who plays defense at that level before you say, no, he can't be their MVP becuase he's a horsebleep hitter? If he's in the 90s, you wouldn't say no way (see Ozzie Smith). In the 80s, well, maybe you would. But that's a maybe, not a no way.

I don't Maz belongs in the HoF or the HoM. But unfortunately that believe lacks certainty because we don't know if we are properly valuing the truly great defender.
   19. Howie Menckel Posted: May 30, 2006 at 06:31 PM (#2043985)
Well, the way I see it, if a typical OPS+ for a 2B in a given year is 89, then 92 is good, for the position.
And if a 2B is a better fielder than his counterparts, that also is good, for the position.
And if he is both of these, that's even better.

Mazeroski would have been a useful player for the Pirates either as a great fielder, or as a slightly-above average hitter for the position. Owning both is a significant value.

All that said, he still won't make my ballot. But I do like the decent-hitting, great fielding combination here.
   20. BDC Posted: May 30, 2006 at 06:53 PM (#2044043)
Or is it more logical to say that Maz's defense and his offense in comparison to one another and to all 2Bs are closer to independent variables than dependents?

Odd as it might sound, that position has a logical attraction. One might even treat a single ballplayer as a sort of package combination of two separate platoon players. So Mazeroski is like being forced to play two guys: a magnificent defender and a guy who hit 8th in really strong lineups, though a creditable enough 8th. (Whereas Fox, though he had bad years, was good enough to hit 2nd for a pennant winner, however weak-hitting they were, and at his very best he was a fine #2 hitter.)

One problem, as sunnyday2 implies, is that it's common to acknowledge a magnificent hitter who should be hidden from view in a strong defensive lineup -- a Ted Williams, Lou Gehrig, or Hank Greenberg -- as one of the all-time greats. Why not the inverse? Clearly there are some prima facie reasons why not the inverse. The great hitter must come up 4 or 5 times a game, and you can try to put him behind other good hitters so he will be most effective. Whereas with a defender, you can't choose to have him make an important play: maybe nobody will hit a ball to second base all game, or maybe Maz will turn two DPs but they will be unchallenging one that any 2B might have turned. A lot of great fielding talent goes to waste for want of opportunity, probably more so than hitting talent; it's a truism that if you want to see a right fielder make a great throw, show up to watch practice, because he may not get the opportunity to do it in the game.

The other prima facie reason to think less of the great defender is that pitching has a lot more to do with team defense than does second-base play.

This is not a problem for Cooperstown, which has tended to think that being cosmically great at a key position is enough all by itself for Fame, and so has elected Maranville, Aparicio, Smith, Mazeroski, Schalk, Ferrell, no matter what they hit. (I don't know that Schalk and Ferrell are cosmically great defenders, but the Hall must have thought so.) But it is an interesting problem for y'all.
   21. Mike Emeigh Posted: May 30, 2006 at 07:23 PM (#2044078)
MWE got Maz's skill set correctly.


I should have, given how much I watched him growing up :)

Fox is also way above average 2B on defense, and at least average or slightly above average on offense.


It's not 100% clear to me that Fox was *way above average* at 2B defensively; he was certainly good defensively, and did win GG's in 1957 (major league), 1959 and 1960 (AL). The Pale Hose were recognized as having great defense, but references to that defense usually talked about the SS (Chico Carrasquel, then Aparicio) and the outfielders (Minoso and Jim Rivera) before mentioning Fox. Fox was usually talked about for his offensive skills (hitting, bunting, and getting on base) and his defensive skills usually mentioned only in passing (as in, "he's playing well on defense, too"). That might very well just be a manifestation of NLoCD applied to middle infielders, of course - but it was clear that Fox was considered an offensive sparkplug first.

-- MWE
   22. Mike Emeigh Posted: June 05, 2006 at 06:29 PM (#2051983)
In the 1978 ballot thread, yest posted:

Bill Mazeroski probably saved on average around 90 runs a year


Let's look at Mazeroski's best defensive season, 1966. Pittsburgh allowed 641 runs in 1966, with a pitching staff that included some guys who had pretty good years for other teams - Pete Mikkelsen, Don Cardwell, Billy O'Dell, Woddie Fryman - as well as some long-term Pirates like Bob Veale and Al McBean who didn't pitch a lot for other teams, and guys like Vernon Law and Steve Blass who didn't pitch for *any* other teams. That wasn't a bad pitching staff to have at that time. They allowed 641 runs and finished fourth in the NL in ERA.

You are claiming that "on average" Maz saved 90 runs a year. Add 90 runs to that season, and the Pirates suddenly have the third-worst staff in the NL, ahead of only the woeful Mets and Cubs. And 1966 (as I noted before) was Mazeroski's *best* defensive season - a season in which he very likely saved more runs than he did in any other year of his career.

I'm as much an admirer of Mazeroski's defense as anyone. But no matter how much I massage the data, I can never get to a conclusion that suggests that any defender saves 90 runs a season. *At best* (1966), Mazeroski was probably somewhere around 40-50 runs.

-- MWE
   23. yest Posted: June 05, 2006 at 10:59 PM (#2052329)
I hpe this sounds coherent
first of all are you using an average 2nd baseman or a replacment value 2nd baseman cause I'm using a replacment level
a replacement 2nd baseman would have abought 1 less putout or assist per game (and because he didn't get that out in addition to having a man on there is still at least 1 more PA that Maz wouldn't have had where for example 2 to 3 runs could have scored) and Maz played in around 95% of games 1957-68 (the years I was counting) and at least 1 third of those would be with men on base
   24. OCF Posted: June 05, 2006 at 11:26 PM (#2052388)
Not to overreact here: yest has already voted in 1978, and the comment MWE was responding to was attached to Mazeroski being placed 28th on his extended ballot. Even yest may never elevate Mazeroski to his top 15.
   25. John (You Can Call Me Grandma) Murphy Posted: June 06, 2006 at 04:19 PM (#2053391)
Not to overreact here: yest has already voted in 1978, and the comment MWE was responding to was attached to Mazeroski being placed 28th on his extended ballot. Even yest may never elevate Mazeroski to his top 15.

I hear you and I agree it's a moot point, OCF, but I'm glad that Mike chimed in anyway. Incorrect data or analysis should always be pointed out.
   26. Mike Emeigh Posted: June 06, 2006 at 04:43 PM (#2053415)
first of all are you using an average 2nd baseman or a replacment value 2nd baseman cause I'm using a replacment level


The problem is that we don't really know how much worse a replacement level *defender* would be when compared to a typical defender. Most analysts believe - with some justification - that a replacement level defender is not significantly worse than an average defender.

a replacement 2nd baseman would have abought 1 less putout or assist per game (and because he didn't get that out in addition to having a man on there is still at least 1 more PA that Maz wouldn't have had where for example 2 to 3 runs could have scored)


You're getting into the "false normalization of fielding statistics" issue that affects fielding stats - the idea that a play not made by player A is a play not made by the team. A play not made by player A may very well be a play not made by the team - but it is also possible (perhaps likely) that it was a play made by player B at another position, which player A never had a chance to make.

But let's suppose for a second that you are correct, and that Mazeroski was converting one more hit per game into an out. The Pirates, in 1966, allowed 1445 hits in 162 games. Replace Mazeroski with a replacement level defender, in your assumption, and the Pirates allow 1607 hits - 94 more than the worst team in the NL, the Cubs, did in reality - with a defense that still includes Clemente and Alley. Does that not seem to you to be an unreasonable result?

Mazeroski made a lot of plays in part because Pirate pitchers threw a lot of ground balls. A replacement-level second baseman, under those circumstances, in all likelihood would still have an above-average number of assists, and probably about the same number of putouts (because a good percentage of Maz's putouts were on DP grounders to the left side of the infield, which he converted into a DP where another 2B would have gotten just a force). Mazeroski deserves a level of credit for the number of DPs he turned - I've estimated that in 1966 Mazeroski turned somewhere around 40-50 more DPs than a typical 2B would have given his opportunities, and he was usually in the +20 range - but you have to deduct some assists from his total to avoid double-crediting him for those extra DPs. Once you do that, and consider the Pirates' ground-ball tendencies, Mazeroski was not making a large number of extra plays above and beyond what an average defender would make given his opportunities, from what I can tell. Some extra plays, yes, but hardly one per game - more like one every 10 games.

-- MWE
   27. Daryn Posted: June 06, 2006 at 04:52 PM (#2053429)
Most analysts believe - with some justification - that a replacement level defender is not significantly worse than an average defender.

Aren't there Gold Glove level fielders (particularly at first base and left field, but also at the other positions) who are in AAA or out of baseball because they can't hit? In that sense, isn't replacement level defence better than average ML defence (because the major leaguers are saddled with the requirement to have to hit).

And if the answer is we look at the fielding ability of players who are replacement level hitters to determine the replacement level for fielders, why would you do that? Isn't, for example, a Gold Glover with an EQA of .220 a better replacement player than an average fielder with a .230 EQA?

Can an argument be made that anything less than GG level fielders should be measured only on their hitting ability because their fileding could be replaced by literally 100s of non-MLers.
   28. DavidFoss Posted: June 06, 2006 at 05:10 PM (#2053466)
I think we've talked about this "replacement fielder not much different than average" here before.

About 50 elections ago, there was a general agreement here that WARP could be improved if FRAR was replaced with FRAA and then recalculated. A fair amount of work was done to post tweaked WARP numbers and then all WARP was completely overhauled forcing the work to start over. WARP has been overhauled a couple of more times since then as well which has been a high deterrent towards every trying that again. (Unless someone has a fancy internet scripting system that can do it overnight or something).
   29. yest Posted: June 06, 2006 at 05:31 PM (#2053550)
But let's suppose for a second that you are correct, and that Mazeroski was converting one more hit per game into an out. The Pirates, in 1966, allowed 1445 hits in 162 games. Replace Mazeroski with a replacement level defender, in your assumption, and the Pirates allow 1607 hits - 94 more than the worst team in the NL, the Cubs, did in reality - with a defense that still includes Clemente and Alley. Does that not seem to you to be an unreasonable result?
2 things would change this somewhat
1 you would have to knock of around 40 hits because when I said hits I meant times on base there would probley be around 40 of those being times on base(to exclude those double plays which would have otherwhise been fielders choices and outs that would have been errors)
you would also have to give the cubs a replacment level fielder at there most valuble postion to make it even

in 1966 Maz had a 5.86 range factor while the leauge average including his was 4.68
and Santo had a 3.56 range factor while the leauge average including his was 2.66
replace them both with replacment level fielders and I think the Pirates would still be ahead of the cubs
   30. yest Posted: June 06, 2006 at 05:39 PM (#2053566)
Aren't there Gold Glove level fielders (particularly at first base and left field, but also at the other positions) who are in AAA or out of baseball because they can't hit? In that sense, isn't replacement level defence better than average ML defence (because the major leaguers are saddled with the requirement to have to hit).

I doubt that very much to be true in the 60's at a defensive position
the only reason that would be done would be because his bat's a liabilty and in the 60's they were looking primeraly for defensive players more then any other time in history
   31. DCW3 Posted: June 07, 2006 at 07:05 AM (#2054990)
I doubt that very much to be true in the 60's at a defensive position
the only reason that would be done would be because his bat's a liabilty and in the 60's they were looking primeraly for defensive players more then any other time in history


But...wouldn't that mean that replacement level for defense would be higher in the '60s than it would be now, since teams would be more willing to accept a weaker bat to go along with it? And doesn't that make it even more unlikely that Mazeroski was really that many runs above replacement level defensively?
   32. jingoist Posted: June 07, 2006 at 08:46 AM (#2055022)
Maz was not fleet of foot or quick to a ball as say an Alomar; he was more like Biggio.
Good instincts and quick pivot and release were his strong suits.

I think MWE has it about right; Maz probably made 10 to 15 extra plays a year based upon good positioning and his above referenced skills.
That said, he still probably saved his pitching staff at least 20 or 30 runs (impossible to ever accurately compute) due to rallies snuffed via premature inning-ending DPs.
His defensive proficiency made up for his inability to hit higher than .260 or drive in more than 50 RBIs a year batting 7th or 8th as a below replacement hitter so he becomes a net-neutral 2B-man.

I do think he gets bonus points as a positive influence toward total team chemistry.
Strong work ethic coupled with his consistent level of play undoubtedly acted as a calming influence to the Pirate pitching staffs and fellow infielders over the years making them a more efficient defensive unit.

Hall of the Very Good but not one of the top 250-300 players.
   33. Ardo Posted: June 07, 2006 at 05:20 PM (#2055246)
1) Mazeroski's 1966 is among the greatest defensive seasons by anyone, anytime, any position.

2) If I had a good Diamond Mind team, but replacement-level 2B and 3Bman, I'd select Mazeroski before Pie Traynor.

That said, my 2B candidate chart looks like Fox-Doyle--Childs-Mazeroski-Evers--Monroe-Lazzeri.
   34. fra paolo Posted: October 21, 2006 at 11:30 PM (#2220450)
Here's Mazeroski's OPS+ against position, for his prime years:

1958 148
1959 75
1960 125
1961 110
1962 120
1963 96
1964 111
1965 97
1966 107
1967 96
1969 97

Only one clunker of a season, which followed a great one. League-averagish over career, with historically great defence.
   35. fra paolo Posted: November 20, 2006 at 12:41 PM (#2242069)
Although I've been silent on the matter of Mazeroski v Fox, I haven't been idle. I asked how Mazeroski's high assist totals compared to historic levels, and now I've compared him to his big rival in my HoM comparison set, Nellie Fox.

Using Charles Saeger's Context Adjusted system, I've done estimates of the number of ground balls for their teams and for their leagues, covering the period 1950-1962 for Fox, and 1958-68 for Mazeroski, periods I defined as their primes, based on their fielding rates.

For those unfamiliar with Saeger's formula, it works as follows:

Take team assists, subtract catcher assists, first base double plays and outfield assists. This gives you GB outs. Then, calculate a GB/FB ratio by subtracting assists and strikeouts from put outs, which will give FB outs. Apply this ratio to hits minus home runs, to create GB hits and FB hits categories. Add GB hits to GB outs to get an estimate of total team GBs. (There's also a stage where one deducts 2b and 3b, but Retrosheet data doesn't include this for pitchers before 1957, so in order to be consistent I used the method I adopted for 1950-6 for both players.)

While not ideal, comparing a middle infielder's assists to an estimate of team GBs must be better than nothing in trying to get a handle on a player's range in an era before PBP data. (We also know that the average number of ZR opportunities for the period covered by Chris Dial's recent research is fairly consistent, and given that there has been no radical shift in positioning, I'd suspect that it remains the case for the distant past, so you could take this team estimate further to at least provide a range of likely outcomes at each position.)

Having gotten a team GB estimate, I also took a league GB estimate, subtracting Fox's and Mazeroski's team where appropriate. However, I had to subtract their team's assists at 2b, and not just theirs, because of the data limitations.

I then used the OPS+ technique to compare each player's percent of team GBs fielded against the league average. I get the following results:

Fox
1950 78
1951 97
1952 100
1953 94
1954 103
1955 119
1956 96
1957 106
1958 89
1959 109
1960 98
1961 86
1962 91

Mazeroski
1958 120
1959 82
1960 97
1961 103
1962 107
1963 100
1964 107
1965 81
1966 111
1967 97
1968 92

In both cases, the really low numbers in their range cover seasons where the player didn't play a full season for some reason. Compared to league averages, there isn't much difference between them in terms of fielding range. Mazeroski's massive lead in assists doesn't actually reflect anything significant in terms of comparison to league averages. Mazeroski's advantage is due to a high ratio of GB to FB for the team, just eyeballing the data. Fox played for a team that regualarly gave up more flyballs than groundballs - the opposite of the typical Mazeroski team in the 1963-8 phase of his career.

Saeger also has a method for estimating men on first, and it would be interesting to see how the two compare. However, Bill James wrote somewhere that Mazeroski's value was largely down to his double plays, so I'm not sure that it's worth my putting in the effort. I'm somewhat more interested to see how Larry Doyle compares on this range measure.

What I did find tremendously interesting, though, was that fielding range on this measure doesn't appear to be a strongly repeatable skill, which surprised me. League averages are fairly steady, but Fox in particular goes up and down quite a bit. (So does Aparicio in the 57-62 period.) Given that the gap between replacement level and average level for fielders is accepted not to be very large, I kind of expected something more like Mazeroski's 1961-4 to be reflected across both the careers of these excellent defenders.
   36. Devin has a deep burning passion for fuzzy socks Posted: November 26, 2006 at 11:27 PM (#2246171)
Since you say playing time affects these statistics in some measure, would it be more accurate if you were comparing GBs/game rather than a seasonal total? Or were you doing that already?
   37. David Concepcion de la Desviacion Estandar (Dan R) Posted: November 27, 2006 at 12:29 AM (#2246190)
40 to 50 runs, Mike??? Even for a historically great defender, that's beyond belief. To my knowledge UZR doesn't have *anyone* at that level in any season...the best, I think, was 2002 Darin Erstad at +35. Particularly since a 2B can't take away extra-base hits....60 net plays in what 800 chances seems to me a bit beyond the realm of possibility...no? I was always under the impression that the absolute greatest defenders are 20-25 runs true talent, 30-40 runs in their best individual seasons. That might not have been true in, say, the 1890's, when infielders had so many more chances than they do today, but certainly should apply to the 1960's, no?
   38. fra paolo Posted: November 27, 2006 at 02:00 PM (#2246379)
Since you say playing time affects these statistics in some measure, would it be more accurate if you were comparing GBs/game rather than a seasonal total?

It would probably help Mazeroski a lot, who had two 'short' seasons in 1959 and 1965, and would greatly narrow the gap between his highest and lowest CRF+ (to give this stat a name of Contextual Range Factor+). I'm not sure it would do Fox any favours outside of 1950. His 1958, for example, is his third-worst season, but he played 155 games. Whether doing opportunities per game makes it more accurate is something for debate.

However, knowing how the system works, and wanting to save time for other work, one can just simply note that those seasons can't be treated at face value, and mentally adjust one's estimations. In terms of range, Mazeroski doesn't really have a big advantage over Fox at the peak end of the scale, although I think he has a noteworthy edge in the middle and low end. In other words, Fox may have a small advantage in Peak value, but Maz clearly edges him on Career.
   39. Mike Green Posted: November 27, 2006 at 04:15 PM (#2246453)
60 excess net plays in 800 chances does not seem beyond the realm of possibility for Mazeroski. The difference in double play frequency between Mazeroski and the rest of the second basemen in the league was so large and so consistent that the figure is credible. If Erstad was +35 at his peak, I see no reason that Maz couldn't have been +42 at his.

It should be possible to mine annual pitching staff GB/FB and the ratio of L/R opponent batters from retrosheet back to 1957. With this data, it would be possible to perform a much more accurate evaluation of Maz's defence compared with other second basemen in the league.
   40. Mike Emeigh Posted: November 27, 2006 at 04:43 PM (#2246485)
It should be possible to mine annual pitching staff GB/FB and the ratio of L/R opponent batters from retrosheet back to 1957.


There are two issues here, both related to the source data (and not the efforts of the Retrosheet crew):

1. Retrosheet's PBPs are incomplete for most years prior to 1974, and Pirate games, in particular, are frequently among the missing. For 1957, as an example, Retrosheet is missing 35 NL games, 22 of which involved the Pirates. There are other occasions pre-1974 where Retrosheet has only a partial PBP account, and what is usually missing from the partial accounts are the identity of the fielder(s) who made the play - there are a good number of 99s (which is how Retrosheet typically records an out with fielders unknown) in the files as you go back from 1974.
2. Most of the time, ball type information is missing for hits.

You can do a little better by treating singles as ground balls and doubles/triples as fly balls - the number of fly ball singles usually comes pretty close to offsetting the number of ground ball doubles and triples, although there are usually more of the former than the latter so it's not 100% accurate - and to the extent that Retrosheet has complete data there's value in doing that. The missing games and partial accounts complicate the issue, however.

-- MWE
   41. Mike Emeigh Posted: November 27, 2006 at 05:19 PM (#2246514)
The difference in double play frequency between Mazeroski and the rest of the second basemen in the league was so large and so consistent that the figure is credible.


1966 was clearly Mazeroski's peak defensive season, and that was the season where I'd estimated 40-50 runs (as a crude estimate, without PBP data). The Pirates turned 215 double plays that season; only two other NL teams (the woeful Mets and the Cardinals) were within 50 of that total. The Bucs' pitchers had around 100 more runners on first than did the Cardinals, which at the Cardinals' DP rate would have yielded about 12 more DPs. The Pirates beat that by 37. The Mets had more runners on first than anyone in the NL (about 60 more than the Bucs had); they were actually less efficient at turning the DP than were the Cardinals, although they turned 5 more DPs overall. Given Pittsburgh's baserunners, New York would have turned about 164 DPs; the Pirates beat that by 51. Those are just huge advantages at the team level - usually the difference top-to-bottom in the league isn't much more than 40-50 DPs when you account for runners on 1B - and this is the difference between the Pirates and the two next best DP teams in the league. Pittsburgh removed about 1 of every 7.5 runners on first by a DP; St. Louis, about 1 in 8.5; the Mets, about 1 in 9.5.

It is primarily because of that difference that I find an argument for Maz being 40-50 runs above average in 1966 to be reasonable. I don't find it reasonable for the rest of his career - that's why I said *at best* when I posted this estimate earlier - but 1966 was just so far beyond anything we typically see that I don't think we can rule out the possibility of it being a +40 season just because we haven't seen one from UZR.

-- MWE
   42. David Concepcion de la Desviacion Estandar (Dan R) Posted: November 27, 2006 at 07:54 PM (#2246614)
For whatever it's worth, BP's FRAA has him at +30 for that year. In addition to accounting for the team's groundball tendencies, don't you also have to account for the proficiency of the other Pirates' infielders at turning them? FRAA has Gene Alley at +15 at SS in 1966. The *combined* FRAA for Maz and Alley of +45 is right in line with your estimate, Mike. Does that seem plausible to you?
   43. Mike Emeigh Posted: November 27, 2006 at 08:32 PM (#2246649)
In addition to accounting for the team's groundball tendencies, don't you also have to account for the proficiency of the other Pirates' infielders at turning them?


You do, but the second baseman is the most important member of the combo. He's on the pivot on most of the team's infield DPs, simply because there are more RHB than LHB, and he's also moving away from 1B while he's on the pivot. A 2B who is slow on the pivot will cost his team more DPs than a SS will.

Another example I like to use: In 1913, Rabbit Maranville was a rookie shortstop. Working with 2B Bill Sweeney, behind a pitching staff that included Dick Rudolph, Lefty Tyler, Bill James, Hub Perdue, and Otto Hess, Maranville turned 49 DPs in 143 games at shortstop; Sweeney turned 42 in 137 games at 2B. In 1914, most of those same pitchers were still on the staff (although the innings were distributed a bit differently; Bill James pitched more, Hess and Perdue, who was traded in May, pitched less). Maranville turned 92 DPs in 156 games at SS; Johnny Evers, replacing Sweeney, turned 73 in 139 games at 2B. (This was an era in which SS often turned as many or more DPs as did 2B, and I don't really have a good idea why.) A year earlier, Evers had turned 70 in 136 games with the Cubs; Sweeney, moving over to the Cubs in 1914, turned 40 in 134 games, again without a lot of turnover on the pitching staff that Chicago had used in 1913 (Hippo Vaughn pitched more, Charlie Smith less) and Cub shortstops went from 58 DPs in 1913 to 47 in 1914. Now it's *possible* that the redistribution of innings had some effect on DPs, and also that the other changes in the defense (most notably for the Cubs, at SS, where 1913 regular Al Bridwell also departed, replaced mostly by Red Corriden) had some effect. But the most important thing to me is that the 2Bs were swapped between teams - and the DPs went with them. And while other analysts (notably Bill James) credit the platoon arrangements of George Stallings for the Miracle Braves of 1914, Stallings himself recognized how important the defensive change was, and was quick to credit his middle infield combo. The Chalmers Award voters did as well - Evers and Maranville were 1-2 in the voting.

The *combined* FRAA for Maz and Alley of +45 is right in line with your estimate, Mike. Does that seem plausible to you?


It's not unreasonable; the question is how much Mazeroski is credited vs Alley for the DPs. BPros, I think, divides them 50-50; I'm suggesting that it's more appropriate to give credit to the 2B than the SS.

-- MWE
   44. David Concepcion de la Desviacion Estandar (Dan R) Posted: November 28, 2006 at 01:13 AM (#2246898)
OK, well if you shift 5 runs to Maz, you get Maz at +35 and Alley at +10, which does seem to me to line up with what I think of as the outer limits of defensive performance. It's totally inconsequential, obviously, but I'm just skeptical that a post-deadball player could really put up a +40 defensive season. No reason for me to keep quibbling.
   45. fra paolo Posted: September 28, 2007 at 10:12 PM (#2548346)
Well.

Off and on, since my last post on this thread nearly a year ago, I've been wrestling with the concept of estimated Zone Rating, an attempt to ascertain whether using bits and pieces of information we know can help us make an estimate of a Zone Rating. Unfortunately, I don't think this is possible for individual players. However, thinking about what we do know, and not fretting about what we don't, I've come up with something.

Going back to Charlie Saeger's Contextually Adjusted Defense provides the main foundation. I also used the distribution of zone opportunities that Chris Dial provided in his rating system. Finally, I used only 70 percent of the possible hits as being fielders' responsibilities, in accord with the 70/30 split for runs described in Baseball Prospectus article on Davenport Fielding Translations published in the 2002 edition.

It's therefore possible to use a team's total stat line at a position to determine something relative to the league. The basic formula is Assists/(Errors + ((GB outs+IF hits * 0.7)*0.29)). It gets sketchier before 1957, when there aren't so many Retrosheet splits. One then goes straight to Dial's formula and calculates the Runs Saved by the league as a whole. Each team's position is then n runs above or below the average. You can take this number and plug it into a Pythagenpat formula to get a Defensive Winning Percentage.

My campaign for Bill Mazeroski has been based around the election of Nellie Fox. If Fox is in the HoM, I thought, so should Mazeroski because Maz's offensive shortcomings appeared to be made up by his defensive superiority. I've not done Fox's 1950 season, but I can do a table for the rest.
   46. fra paolo Posted: September 28, 2007 at 10:22 PM (#2548360)
Fox (White Sox 1951-63, Astros 1964)
Season   DWP  Gold Glove team
1951   .492   Cle .509 (Bobby Avila 136)
1952   .403   Was .512 (Floyd Baker 68, H Moderlein 58)
1953   .500   Cle .511 (Bobby Avila 140)
1954   .496   Cle .521 (Bobby Avila 141)
1955   .515   Chi (Fox 154)
1956   .495   Cle .509 (Avila 135)
1957   .501   Det .510 (Frank Bolling 146)
1958   .495   Det .514 (Bolling 154)
1959   .503   Det .504 (Bolling 126)
1960   .496   Bos .506 (Pete Runnels 129)
1961   .493   Was .5068 (Chuck Cottier 100/D. O'Connell 61)
1962   .502   Cle .516 (Jerry Kindall 154)
1963   .495   Cle .512 (Woodie Held 96, Kindall 37)
1964   .505   StL .511 (Julian Javier 154)
   47. fra paolo Posted: September 28, 2007 at 10:26 PM (#2548362)
Mazeroski (Pirates)
Season DWP Gold Glove Team
1957 .503 Cubs .512 (Bobby Morgan 116)
1958 .517 Pit
1959 .500 Phi .509 (Sparky Anderson 152)
1960 .501 Cubs .513 (Kindall 82, Don Zimmer 75)
1961 .503 Mil .510 (Bolling 148)
1962 .508 Pit
1963 .513 Pit
1964 .505 StL .511 (Javier 154)
1965 .511 Pit
1966 .5078   Pit
1967 .498 Cin .513 (Tommy Helms 88, Chico Ruiz 56)
1968 .510 Pit
   48. fra paolo Posted: September 28, 2007 at 10:31 PM (#2548366)
Remember, these DWP relate to the team at that position, not a player per se. That's why included games played at that position when listing the "Gold Glovers". As a rule of thumb, I used 115 as the cut-off point.

Obviously, the system has its problems, but I think it goes further to giving us a glimpse of range in PBP days than FRAA does. Furthermore, using the DWP allows one to estimate the total number of wins above average the players at a position contributed with the glove. Avila's 1954 season is the best in my data set so far, and it's worth +3 wins. Three of the worst I've found are

Charlie Neal 1960 Dodgers .485
Bobby Richardson 1961 Yankees .487
Bobby Young 1953 Browns .488

Richardson is a particularly notable example. His 1961 season looks to me as a useful example of exactly what we mean by a 'replacement level player'.
   49. fra paolo Posted: September 28, 2007 at 10:32 PM (#2548370)
Oops! Fox's 1952 should read .503
   50. fra paolo Posted: September 28, 2007 at 10:35 PM (#2548374)
A clarification. That basic formula is for 2bs, 3bs, and SSs only. One uses different methods at the other positions. Also, I deduct pitcher assists from the GB outs, because Dial doesn't give an average number of Zone Opportunities for them.
   51. fra paolo Posted: September 28, 2007 at 10:42 PM (#2548380)
Aargh. The formula is only for 2bs. I'm too tired to be doing this.
   52. fra paolo Posted: September 28, 2007 at 10:47 PM (#2548385)
Another bad fielder was Glenn Beckert in 1966, 162 games at second for the Cubs, .485.

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
Adam S
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Syndicate

Page rendered in 0.7253 seconds
49 querie(s) executed