Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Baseball Newsstand > Discussion
Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Sunday, November 29, 2020

Is the Hall leaving out too many players?

You should have two takeaways from these:

1) The lowest points in the first nine decades or so of the 21st century came during the two World Wars, understandably so. Bob Feller, for example, missed all of 1942-44 in America’s service, while Ted Williams missed all of 1943-45. Dozens of Hall of Famers missed the primes of their careers serving, plus several more (including Williams again) in the Korean War. The talent level had a meaningful drop.

2) The last three decades have consistently had the lowest Hall of Famers-per-year numbers in history outside of those wars, and it’s not just because not all of those players are eligible yet.

Taken all together, it does appear we’re being too stringent with the recent decades. We’re not hitting the “1.5% of all players” or “4.3% of regular players” thresholds, and the only times in history we’ve seen such a low percentage of Hall of Famers was literally during two world wars.

RoyalsRetro (AG#1F) Posted: November 29, 2020 at 10:27 AM | 423 comment(s) Login to Bookmark
  Tags: hall of fame

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.

Page 4 of 5 pages ‹ First  < 2 3 4 5 > 
   301. BDC Posted: December 16, 2020 at 06:07 PM (#5994564)
One factor I had never even heard of till it somehow influenced a recent NFL game was the K-Ball, special footballs that are set aside solely for kicking. They were introduced in 1999. Apparently the idea was to stop kickers from customizing their footballs with all kinds of weird treatments analogous to scuffing baseballs, thus to make it harder to kick FGs. But for all I know, the K-Ball may have had the opposite effect; maybe kickers had no idea how best to prepare the ball before that.
   302. SoSH U at work Posted: December 16, 2020 at 06:08 PM (#5994565)

What I think your data *does* suggest is that NYY pitchers took a lot of extra discretionary ground balls when Jeter played, balls that they might allow another SS to handle. (And maybe 3B and C did this too, to a much smaller extent.)


C?
   303. Lowry Seasoning Salt Posted: December 16, 2020 at 06:37 PM (#5994570)
Rally Posted: December 16, 2020 at 11:41 AM (#5994473)

While we're at it, I wrote The Defense of Derek Jeter about a year ago.


I'd forgotten about this. Good to reread it. Thank you.
   304. jacksone (AKA It's OK...) Posted: December 16, 2020 at 07:50 PM (#5994579)
If he was open to it (and remember, this is Sheff).

That's one thing that I think gets overlooked when it comes to the DH. A lot of guys don't want to do it (we've seen that with Bonilla, Giambi and others). They don't want to be seen as half a player, even if that's the role their best suited for. I imagine a manager appreciates when a team's best hitter, such as an Edgar or Ortiz, will accept the role without complaint.


I've been saying this for years re: just assuming anyone can DH. And Sheffield specifically said he did not want to do it - this from when he was begrudgingly a full time DH for the Tigers (finally! at 39!) -

"I can be in the outfield everyday...I don't want to DH, I don't feel like a baseball player when I DH." Sheffield said. "I don't know how to be the leader that I am from the bench...I can't be vocal, I don't feel right about it"


And note that he signed with an NL team for his last year in the league.

Other guys have said the same thing - this from Adam Dunn after he signed with the White Sox -

“That’s going to be the toughest adjustment for me,” Dunn said Wednesday.

“I know that. Luckily, I have spring training to figure that out. I’ve tried to talk with everybody who I thought might have DH’ed a game — Frank Thomas, Harold Baines, Paulie — and see what they do. I have no idea. I can’t sit on the bench, dude. Can’t do it. Ain’t happening.”


And you all really think someone like Jeter - who wouldn't move off his position for a much better Alex Rodriguez would willingly be a full time DH?
   305. bachslunch Posted: December 16, 2020 at 07:51 PM (#5994580)
Re the NFL kickers comments:

Not to turn this into an NFL kicking thread, but the analogy is very good. Also, I've got nothing against Jan Stenerud, who was finishing up his career when I was a kid, but compared to today's kickers...he wasn't very good. For his career, he made only 67% of his field goal attempts, including only 26% of 50+ (he attempted very few of those compared to today's kickers), and only 51% (!) from 40-49 yards out. For some reason, he was markedly more accurate in his late 30s and 40s than he was in from 1967-1980.

There are NFL writers like Chase Stuart who have made attempts to systematically codify players in skill positions like quarterback and place kicker, doing so in a more-or-less sabremetric manner. Adjusted for era, he has Jan Stenerud as the third most accurate all-time field goal kicker in NFL history (he looked at players from 1960 and later), behind only Nick Lowery (who absolutely should be a HoFer and probably won't be) and Morten Andersen (who is in). For his time period, Stenerud was extraordinarily accurate.

https://www.footballperspective.com/the-greatest-field-goal-kickers-of-all-time-ii-part-iii-career-rankings/

Ray Guy was the best punter, ever. Morten Andersen & Jan Stenerud - long, good careers, acknowledging that kicker really is its own position, and a fairly important one at that, these guys were good and deserved the HOF. There's probably two or so current kickers that will join them when eligible (Vinatieri and I am sure there's someone else out there)
George Blanda & Lou Groza were the Eck/Smoltz mold - they were more than just kickers.


Andersen and Stenerud are #2 and #3 all time in accuracy per Chase Stuart above, and very deserving HoFers. A fair bit of Groza's career predates 1950, so he ranks 15th in that poll. Much of his career came before then, and Rupert Patrick (who did so going back before 1960) lists his five greatest kickers adjusted for era as Groza, Stenerud, Andersen, Lowery, and Gary Anderson in what may still be an unpublished study. Adam Vinatieri is very likely to be elected 5 years after he retires, but his FG accuracy adjusted to era is not impressive, and frankly he would not be an especially strong choice. George Blanda ranks 38th adjusted for era, though some of his kicking occurred pre-1960. Blanda is borderline at best as a QB candidate, and he really needs the kicker boost to deserve his HoF bust. Groza was in fact a world-class level offensive tackle; either this or his kicking alone qualify his enshrinement easily.

The notion that Ray Guy was the Greatest Punter of All Time is in fact very debatable. The best one can say is that he was arguably the best punter of his era, as he has 6 first team all pro selections and went to 7 pro bowls, which are solid postseason honors. He unfortunately had a reputation for booming punts as far as he could, which reportedly meant a number of his kicks ended up in the end zone for a touchback rather than pinning the opposition back near the goal line by kicking out of bounds just short of the goal line. Those are called Coffin Corner punts, and something the best punters should be able to do. Sammy Baugh is pretty much universally considered the greatest punter in NFL history. All of the top 40 punters by average yards per punt started their careers in the year 2000 or later, with three exceptions: Baugh (23rd), Tommy Davis (tied for 26th), and Yale Lary (tied for 33rd). Baugh and Lary were respectively also elite at quarterback and safety respectively and are in the HoF. Davis was a combo punter-placekicker who should also be enshrined but likely won't ever get in -- he was an indifferent level field goal kicker, but devastatingly accurate on extra points, only missing two in his career (he long held the record for most consecutive XPTs without a miss). Davis should probably also get credit for kicking in Kezar Stadium, which was known for swirling winds and was considered a kicker's nightmare. Guy is actually tied for 78th all time in career average yards per punt, and his HoF case is largely the product of shameless and heavily sustained stumping on his behalf by the Raiders (and it may also get the very undeserving Tom Flores in the HoF this year). Myself, I think Guy's an okay HoF option, but shouldn't be in if Tommy Davis isn't.

https://www.pro-football-reference.com/leaders/punt_yds_per_punt_career.htm
   306. John DiFool2 Posted: December 16, 2020 at 09:05 PM (#5994586)
Adam Vinatieri is very likely to be elected 5 years after he retires, but his FG accuracy adjusted to era is not impressive, and frankly he would not be an especially strong choice.


At such high average percentages, it is much harder to rise above said average than when it is down around 65%. Everyone above him on the career list is younger than he is.

The notion that Ray Guy was the Greatest Punter of All Time is in fact very debatable.


Punters have also seen their averages skyrocket in this millenium: again, almost all punters above him are younger (plus some old-timers). They moved the goalposts back in his 2nd year (which would give a boost to the old-timers playing before that), which means more punts inside the 50 yard line. Plus the Raider's offense likely meant a lot of punts inside the 50, and a fair amount inside the 40 (given the crap abilities and ranges of his placekickers in the 70's at least: watch the 1977 Super Bowl to see some utterly pathetic kicks by straight-on kicker Fred Steinfort; I am certain than any current Division 2A kicker today would instantly chase him out of training camp if sent back in time).

The only argument against him is the touchback thing (FBRef doesn't have that data for some odd reason tho I know the NFL kept track it back then), but, again, he likely had a lot more chances than most, at least on a percentage basis. But the guy who followed him 15 years later, Shane Lechler, may have more of a case (some filthy inside the 20 stats, 30% of all punts).
   307. Rally Posted: December 16, 2020 at 10:02 PM (#5994594)
What I think your data *does* suggest is that NYY pitchers took a lot of extra discretionary ground balls when Jeter played, balls that they might allow another SS to handle. (And maybe 3B and C did this too, to a much smaller extent.)


The data show that pitchers made more plays in front of Jeter, but I’m not convinced these are discretionary plays. A hard hit ball that a pitcher can get, he doesn’t have time to think about anything let alone the range of his shortstop. You just react. On soft hit balls, a pitcher might have time to think but if he doesn’t make the play, its going to be an infield hit by the time it gets to SS. I just see the data showing what happened, not necessarily that there was a conscious plan to get that outcome.
   308. Rally Posted: December 16, 2020 at 10:03 PM (#5994595)
I do think the Yankees were able to position their fielders in a way to somewhat mitigate Jeter’s lack of range.
   309. Rob_Wood Posted: December 16, 2020 at 10:17 PM (#5994596)
Erroll Mann
   310. alilisd Posted: December 16, 2020 at 11:54 PM (#5994602)
C?


Apparently Posada had much better range than we realized?
   311. GuyM Posted: December 17, 2020 at 08:50 AM (#5994615)
C?

Oops, meant to say 2B. In any case, I think it's really the pitchers who are doing something different with Jeter on the field. They account for 2/3 of all the "extra" plays made by non-Jeter fielders (and it's certainly possible the other fielders were simply a bit above average). And their extra 15 plays per season can easily account for the actual discrepancy between the extreme negative ratings in WOWY/GB+ (-25) and what appears to be his actual performance (c. -18). Maybe other fielders positioned themselves slightly differently to cover Jeter's shortcomings, but even if true, most of that would be offset by plays missed in the vacated territory (down the 3B line, or in 1B-2B hole) -- or else those fielders would assume the Jeter position all the time.

Back to the key point: we simply can't assume that Jeter's backups were league-average fielders. Nate Silver looked at the hitting and fielding performance of replacement players, and found SS were about -5 runs in fielding on average. Using DRS, I looked at starting SSs (750+ innings) from 2017-2019, and they were +3 fielders on average. That means backup SS were -6.4 runs (per 150) on average. Once we account for this, Jeter is still a -300 runs fielder for his career. Not quite as bad as WOWY indicates, but still in the "historically bad" category.

Rally also shows how many ground balls were fielded by NYY LF, CF, and RF with and without Jeter on the field. When Jeter plays, the LF handles vastly more ground ball hits (about 25 per season), the CF handles somewhat more, and the RF handles just slightly more. That is perfectly consistent with Jeter being the consistent weak link in the NYY infield.
   312. Howie Menckel Posted: December 17, 2020 at 09:04 AM (#5994616)
The notion that Ray Guy was the Greatest Punter of All Time is in fact very debatable.

Guy didn't even finish his career before he began to be mocked for his "kick it as far as I can and who cares about the end zone" style. and with that went any "GPOAT" talk.

Ex-Bears and Bengals QB Virgil Carter was lauded in WSJ a couple of weeks ago for his now-50-year-old primitive computer analysis of the game. how two biggest pet peeves, iirc, were punts into the end zone - a 20-yard penalty - and insisting on punting on 4th-and-1 almost no matter the game circumstances.

first coach I can remember consistently going for it there around midfield was Joe Gibbs with RB John Riggins and his massive Hogs up front. Bill Parcells soon followed with OJ Anderson.

as for Jeter, I should have patented my "gracefully slow" comment about him - as BBTF poster pastadivingJeter might agree. the ball was hit, and he just remained frozen for that extra tick or two. it's all there on the videotape.
   313. Rally Posted: December 17, 2020 at 09:38 AM (#5994620)
as for Jeter, I should have patented my "gracefully slow" comment about him - as BBTF poster pastadivingJeter might agree. the ball was hit, and he just remained frozen for that extra tick or two. it's all there on the videotape.


I remember that, and don't dispute it, but it's weird. Reactions are kind of important to hitting. How does a guy with terrible reaction time end up with 3000 hits?

Maybe he only has good reactions in the batter's box, but it's hard to understand why it that wouldn't translate.
   314. DL from MN Posted: December 17, 2020 at 10:01 AM (#5994626)
How does a guy with terrible reaction time end up with 3000 hits?


You don't have to move your legs when you hit. Jeter had the range of an average 3rd baseman. Plenty of 3B can hit.
   315. GuyM Posted: December 17, 2020 at 11:48 AM (#5994650)
Jeter had the range of an average 3rd baseman.

That seems unfair to 3rd basemen. On average, they would probably be something like -7 or -10 at SS -- significantly better than Jeter.
   316. DL from MN Posted: December 17, 2020 at 12:11 PM (#5994653)
Looking over qualified 3B in 2005, Alex Rodriguez was basically an average defensive 3B so you have a point. Jeter probably would have been a below average 3B but not nearly the worst. Troy Glaus, Shea Hillenbrand and Jorge Cantu were regulars. Jeter had decent glove-side range, plus arm and charged balls well. The difference in RPos between 3B and SS is about 5-6 runs a year. He would lose 9 career WAR shifting positions but probably more than make it up on defense.
   317. Slivers of Maranville descends into chaos (SdeB) Posted: December 17, 2020 at 12:19 PM (#5994654)
Guy didn't even finish his career before he began to be mocked for his "kick it as far as I can and who cares about the end zone" style. and with that went any "GPOAT" talk.


I encourage you not to Google "Greatest Punter of all time" then.
   318. Rally Posted: December 17, 2020 at 12:59 PM (#5994659)
You don't have to move your legs when you hit. Jeter had the range of an average 3rd baseman. Plenty of 3B can hit.


That would explain things if Jeter had a bit more in common with a typical slow 3B. But on the offensive side Jeter was good at moving his legs. He stole 358 bases, including 10 out of 12 when he was 40. He had a good number of infield singles and reached on error. I remember his home to first times being good even late into his career. There are just details of Jeter that just don’t seem to fit together. Maybe he was a secret DH, and sent his slow footed but otherwise identical twin brother out to the field every game and nobody ever saw the two of the, together.
   319. GuyM Posted: December 17, 2020 at 01:31 PM (#5994665)
Jeter probably would have been a below average 3B but not nearly the worst. Troy Glaus, Shea Hillenbrand and Jorge Cantu were regulars.

I think Rally makes a strong case in his article that Jeter should have been moved to 2B, not 3B:

If I had been running the Yankees I would have tried to move him to another position, the typical career path for good hitting, poor fielding shortstops. The 2004 season seemed like an obvious opportunity, as they acquired the reigning gold glove winner at shortstop and in doing so opened up a spot at second base. Jeter's weaknesses on defense were that he didn't react as quickly as other players, and his arm wasn't as good as the top shortstops. With those weaknesses, I don't think he would have been a good fit for the hot corner and if those were the only two options, it might have been the correct decision to leave him at short and play A-Rod at third. At second base, his arm would not have been an issue and his speed and solid fundamentals would have, in my opinion, enabled him to play solid defense there. The window closed a year later however as the Yankees came up with a very good second baseman.
   320. John DiFool2 Posted: December 17, 2020 at 10:33 PM (#5994811)
Guy didn't even finish his career before he began to be mocked for his "kick it as far as I can and who cares about the end zone" style. and with that went any "GPOAT" talk.

I encourage you not to Google "Greatest Punter of all time" then.


Just one more thing, and then I'll shut up.

In Guy's day EVERYBODY aimed to kick the ball out of bounds inside the 10. Nowadays we know it's better to hit it high and make the returner catch it around the 10 (or down it if he lets it bounce). You can blame, or credit, the older punters for that if you wish.
   321. GuyM Posted: December 18, 2020 at 08:32 AM (#5994830)
Did the Braves allow fewer outfield hits than other teams? How would I even find that information?

There are approaches to estimate things like that, but hard data is not in the public domain. Retrosheet has codes for batted ball type and who fielded each play. So you could look at hits where BBType = F or L, that are fielded by an outfielder. But Retrosheet only has complete data in these fields from 2003 on, it's just missing in too many cases for years before that.

Rally, couldn't you use WOWY to determine whether Andruw stole discretionary plays for ATL corner OF? For each significant teammate, measure their putouts with and without Andruw next to them -- not just on the Braves, but for their career. If Francoeur, Jordan, etc. make fewer outs when playing next to Andrew, then it's pretty clear what's happening. But if not.....
   322. Joyful Calculus Instructor Posted: December 18, 2020 at 10:54 AM (#5994851)
Dude, you just broke the page.
   323. SoSH U at work Posted: December 18, 2020 at 11:01 AM (#5994855)
Yup, we need RoyalsRetro to rescue us.

   324. GuyM Posted: December 18, 2020 at 12:40 PM (#5994883)
Sorry. I put italics inside a quote. Is that verboten?
   325. SoSH U at work Posted: December 18, 2020 at 12:52 PM (#5994890)
You probably hit code instead of quote.

Code exists specifically to break the page, which, in retrospect, might have been a site design flaw.

   326. Ron J Posted: December 18, 2020 at 03:27 PM (#5994934)
325. He did in fact hit code. I still have the keys even if I rarely use them. Fixed it.
   327. Mefisto Posted: December 18, 2020 at 03:39 PM (#5994938)
Thanks Ron.
   328. baxter Posted: December 20, 2020 at 02:42 AM (#5995187)
Gossage was a starter for one year; when Paul Richards came back to manage the Chisox.

Gossage was not good as a starter.
   329. Howie Menckel Posted: December 20, 2020 at 11:54 AM (#5995220)
I encourage you not to Google "Greatest Punter of all time" then.

I'm glad I did, because I did not realize how often Shane Lechler and Johnny Hekker's names would come up at Number One. So I learned something, thanks.
   330. Rally Posted: December 21, 2020 at 10:49 AM (#5995312)
Rally, couldn't you use WOWY to determine whether Andruw stole discretionary plays for ATL corner OF? For each significant teammate, measure their putouts with and without Andruw next to them -- not just on the Braves, but for their career. If Francoeur, Jordan, etc. make fewer outs when playing next to Andrew, then it's pretty clear what's happening. But if not.....


Could be done. If anyone wants to dig into that, I'd suggest looking at 2B/SS catching popups too - I think it was Chris Dial who said that since Andruw played so shallow he was catching popups that normally are caught by infielders.
   331. Mefisto Posted: December 21, 2020 at 11:14 AM (#5995313)
It was Chris.
   332. snapper (history's 42nd greatest monster) Posted: December 21, 2020 at 01:02 PM (#5995331)
I think it was Chris Dial who said that since Andruw played so shallow he was catching popups that normally are caught by infielders.

There's a good discussion of Dial's work here. Scroll down around two-thirds of the way.

jaws-and-the-2021-hall-of-fame-ballot-andruw-jones/

Great quote.

“Jones just took all the discretionary plays from the left fielder and continued to do so after he had lost his range. That’s not talent, it’s Kelly Leak,” referring to the ball-hogging star of the Bad News Bears. UZR and DRS “weight plays made by percentage for a position – when Andruw takes a discretionary play, he gets too much extra credit in those systems. Everything else tells us Andruw lost a step or three. His zone ratings (percentage of balls caught), his extra weight, his speed scores, his range factors. How the other metrics miss this, I cannot say,” Dial wrote.
   333. GuyM Posted: December 21, 2020 at 01:04 PM (#5995332)
OK, this is pretty crude, but as a first cut we can look at ATL putouts per 9 innings by position before and after Andruw arrived. These are the flyball putout averages for pre-Andruw (1994-96) and new-Andruw (1998-99 -- I skipped 1997 because Andruw split his time between CF and RF).

ATL Putouts (flyballs):
Pre-AJ / AJ
2B 0.58 / 0.57
SS 0.52 / 0.48
LF 1.47 / 1.58
CF 2.27 / 2.88
RF 2.09 / 2.06

There's a huge increase in CF, while every other position is stable (or small *increase*). Now, maybe there was a simultaneous change in pitchers and/or opposing hitters that is obscuring Andruw's theft of opportunities from teammates. But to my surprise, there is zero evidence here that Andruw did anything other than catch a ton of balls that other CFs wouldn't get.

1997 is also interesting, because Andruw got a decent number of innings in both CF and RF. Here are his putouts/9 and his teammates (mainly Lofton and Tucker):
CF: Andruw 3.01, Lofton/etc. 2.49
RF: Andruw 2.40, Tucker/etc. 1.95

I'm not saying Andruw actually made an extra out every two games (which is what it looks like here). But he only has to do half of that to be a +30 run CF -- about what the advanced metrics say.
   334. GuyM Posted: December 21, 2020 at 01:24 PM (#5995333)
Jones just took all the discretionary plays from the left fielder and continued to do so after he had lost his range. That’s not talent, it’s Kelly Leak,” referring to the ball-hogging star of the Bad News Bears. UZR and DRS “weight plays made by percentage for a position – when Andruw takes a discretionary play, he gets too much extra credit in those systems.

Chris knows way more about defensive metrics than I do. But he is wrong here, at least about UZR (I know less about DRS). In UZR, catching a discretionary ball that is nearly always caught (sometimes by the CF, sometimes by another fielder) does nothing at all to lift a player's rating.

There is actually no evidence presented that Jones actually took a lot of easy chances from teammates. And if you look at Chris' table in that article, comparing his metric (RED) to four other fielding metrics, the only possible conclusion you can reach is that RED is a total outlier, while the other four metrics are in agreement that Jones was fantastic.
   335. Rally Posted: December 21, 2020 at 03:33 PM (#5995359)
These are the flyball putout averages for pre-Andruw (1994-96) and new-Andruw (1998-99 -- I skipped 1997 because Andruw split his time between CF and RF).


Interesting. If he was indeed turning a hit into a putout every other game, that would certainly be huge enough to find evidence of in DER. That should be worth .020 on the DER. Looking at the team total I find:

1994 .697
1995 .696
1996 .689
1997 .700
1998 .700
1999 .694

So .694 the 3 years before he came up, .698 the 3 years after (though splitting time between CF/RF in 1997). A lot smaller, but still nothing to sneeze at, as over a full season 4 points of DER is the equivalent to 16 more outs than average. The last time MLB played a full season only two outfielders did better than +16, Robles and Kiermaier.
   336. GuyM Posted: December 21, 2020 at 04:48 PM (#5995378)
Interesting, but DER includes a lot of ground balls we don't care about. Do we know OF-DER for these years?

We can look at TZ for all ATL outfielders, which should account for Jones stealing plays from teammates (though not for their talent). Avg annual OF TZ for 1994-96 was +14, and for 1997-99 was +60 -- a net improvement of 46 runs per season with Jones. Not too shabby.....
   337. Ron J Posted: December 21, 2020 at 06:01 PM (#5995400)
#336 For that analysis to make sense though you'd have to believe that Atlanta suddenly became less good at turning ground balls into outs when Jones came up. I've always been suspicious of methods that don't try what Rally did -- reconcile at the team level as a sanity check.

Now it is absolutely possible that Atlanta did become less good at turning groundballs into outs. But I'd want to see evidence of this.
   338. GuyM Posted: December 21, 2020 at 06:13 PM (#5995407)
Now it is absolutely possible that Atlanta did become less good at turning groundballs into outs. But I'd want to see evidence of this.

Let's do this. TZ range rating for ATL infielders averaged +24/year from 1994-1996, but just +3 from 1997-99 after Jones arrived. So that's a net drop of 21 runs, or about 28 extra plays not made by the infield. That would *drop* team DER by .006, but in fact team DER improved by .0045. Add that all up and it implies the outfielders (AKA "Andruw Jones") raised team DER by something like .0105. Translated into runs, that's about +35 runs contributed by Jones.
   339. snapper (history's 42nd greatest monster) Posted: December 21, 2020 at 06:32 PM (#5995414)
Let's do this. TZ range rating for ATL infielders averaged +24/year from 1994-1996, but just +3 from 1997-99 after Jones arrived. So that's a net drop of 21 runs, or about 28 extra plays not made by the infield. That would *drop* team DER by .006, but in fact team DER improved by .0045. Add that all up and it implies the outfielders (AKA "Andruw Jones") raised team DER by something like .0105. Translated into runs, that's about +35 runs contributed by Jones.

How do you know that's not just the shallow playing Jones stealing pop-ups from IF?
   340. GuyM Posted: December 21, 2020 at 06:36 PM (#5995415)
339. See 333. ATL middle infielders caught *fewer* flyballs after Andruw arrived.
   341. Dog on the sidewalk has an ugly bracelet Posted: December 21, 2020 at 06:47 PM (#5995416)
Yeah, Guy, that's the argument the anti-Andruw folks are making. Though they only dropped by a combined .05/9, so even if we assume those numbers to be representative, it would only explain a fraction of what was going on.
   342. Eric J can SABER all he wants to Posted: December 21, 2020 at 07:29 PM (#5995419)
Atlanta's GB/FB ratio and K rate as a team, year-by-year:

'94: 0.95, 20.0%
'95: 0.97, 20.1%
'96: 0.94, 20.3%
'97: 0.88, 19.8%
'98: 0.91, 20.7%
'99: 0.83, 19.3%
'00: 0.79, 17.7%

Not sure how strongly these things are controlled for in what everyone is looking at (I know TZ at least ignores strikeouts), but the Braves' pitchers were becoming more fly ball heavy and less strikeout prone just as Jones was starting up.
   343. Mefisto Posted: December 21, 2020 at 07:51 PM (#5995420)
The issue with DER is that if the OF makes more POs, the IF must make fewer; it's a zero sum game at 27 outs/9 innings.
   344. GuyM Posted: December 21, 2020 at 08:29 PM (#5995424)
No, DER is not zero sum. The denominator is BIP not outs.
   345. Mefisto Posted: December 21, 2020 at 09:42 PM (#5995430)
I'm assuming outs other than BIP (e.g., Ks) remain the same. The numerator is fixed, so extra POs by any one player will necessarily reduce the POs elsewhere. This makes it hard to know if Jones was getting more FBs or the Braves IF was getting fewer GB.
   346. snapper (history's 42nd greatest monster) Posted: December 21, 2020 at 09:59 PM (#5995432)
The numerator is fixed, so extra POs by any one player will necessarily reduce the POs elsewhere.

Not if the player was converting hits into outs. That's the whole argument. If Jones was just taking PO from other fielders, he wasn't adding value. He had to be turning hits into outs.
   347. Mefisto Posted: December 21, 2020 at 10:35 PM (#5995435)
The problem is that DER can't tell you the answer. If a team's DER goes up, it might be due to Jones converting hits into outs that no other player could do. It might be from other players doing so. It might be from the pitchers allowing fewer BIP (reducing the denominator). It might be from general improvement by the team's defense. All of these are possible causes of higher DER and the DER itself can't tell us which is the actual cause.
   348. GuyM Posted: December 21, 2020 at 11:33 PM (#5995440)
The problem is that DER can't tell you the answer.

I don't think anyone said DER can "tell us the answer." I took Rally's point to be that, if a team adds a player as good as Jones appears to be, we should that in improved team DER. If we don't, then either Jones wasn't as good as purported, or by chance ATL got worse elsewhere on the field. And it turns out that A) team DER did improve somewhat, and B) the infield got a lot weaker, so C) DER is consistent with high-end estimates of Jones' fielding performance. DER doesn't prove the Jones metrics are right, but one plausible objection is removed.

And as far as I can tell, the story so far is this:
1) TZ says Jones was an historically great CF
2) the play-by-play metrics aren't available for a lot of his peak, but for the years we do have (2003+) they closely match TZ.
3) When WOWY controls for the identity of pitchers and hitters, Jones' numbers still look fantastic.
4) When we look at other positions from which Jones might have "stolen" easy plays, those positions record just as many (or more) outs after Jones joins ATL as they did before.

The fact is that Jones made many, many more plays than an average CF. Jones made many, many more plays than ATL CFs before him. Jones made many, many more plays than other ATL CFs while he was there. In RF, Jones made many, many more plays than ATL RFs before him, or other RFs while he was there. And there really isn't any evidence (yet) that this all resulted from a surge of additional CF opportunities in ATL or from Jones taking plays from other fielders -- and substantial evidence those things didn't happen.

So I don't think it's reasonable any longer to say "Andruw's ratings are too extreme to take at face value -- you need to prove he was really that good." Multiple methods independently say he was that good. And the theories that might explain away those ratings have not fared well under examination. I'd say the burden of proof has shifted: someone needs to prove the metrics are wrong, or at least provide strong evidence to that effect, or we should accept what the metrics say.
   349. Rally Posted: December 22, 2020 at 08:18 AM (#5995456)
Guy, good summary of where the debate stands.

I just want to point out that Jones' +230 TZ is actually only +173 if we're looking at catching flyballs. It's historically great but not that different from some others, like Paul Blair (+140) who played a similar number of games in CF. You get to +230 when adding in his historically great CF throwing arm, another 57 runs.

TZ, UZR, DRS, all of these are somewhat of a black box. At this point I'm not sure I could even recreate TZ. I've moved on to new computers a few times since that was published, and am just glad I sold the code to Sean Forman so he can do the updates. But the throwing arm stuff is very replicable, even without advanced programming skills. BB-ref shows the number of times a runner was on second and a ball hit to CF, how often he scored or was held at third, etc. Jones had 20 assists in 1998, and the system I used says that was worth 9 runs. Next year he only had 13 assists, but that's because people stopped running on him and his arm rating is even better at +12.

Anybody who put some time into it could take the numbers on BBref, compare to league averages, and with good assumptions for what a base advance, hold, or runner thrown out is worth should get something close to that +57.
   350. cookiedabookie Posted: December 22, 2020 at 08:19 AM (#5995457)
RE: JAWS, I actually think for HoF voting purposes, it's better to look at players better than the bottom quartile of HoFers, rather than better than half of the HoFers.
   351. Mefisto Posted: December 22, 2020 at 09:52 AM (#5995465)
I don't think anyone said DER can "tell us the answer." I took Rally's point to be that, if a team adds a player as good as Jones appears to be, we should that in improved team DER.


Ok, fair.
   352. DL from MN Posted: December 22, 2020 at 09:55 AM (#5995466)
Let's do this. TZ range rating for ATL infielders averaged +24/year from 1994-1996, but just +3 from 1997-99 after Jones arrived. So that's a net drop of 21 runs, or about 28 extra plays not made by the infield. That would *drop* team DER by .006, but in fact team DER improved by .0045. Add that all up and it implies the outfielders (AKA "Andruw Jones") raised team DER by something like .0105. Translated into runs, that's about +35 runs contributed by Jones.

How do you know that's not just the shallow playing Jones stealing pop-ups from IF?


That's exactly what stealing popups from the infielders would look like. I don't think anyone is arguing that Andruw Jones was not a terrific outfielder. The question is whether he's more than a standard deviation better than the next best outfielder in the history of baseball. If you regress him to being just a little better than the next best CF he loses 3 or more career WAR. The answer is probably something in between - the outfielders of the past are being regressed too much and Andruw may have some quirks in the data that overstate his performance.

The CF Andruw replaced were Marquis Grissom and Kenny Lofton - both above average defenders. The infielders who suddenly saw their performance crater were Jeff Blauser and Mark Lemke - they didn't replace the infielders until 1998. It's possible Walt Weiss and Bret Boone really did have lousy range going back on popups, Bobby Cox realized this and positioned Andruw shallow to pick up the slack (though I doubt it). However, this didn't improve the overall Braves defense all that much.

One thing that is different with Andruw compared to other terrific centerfielders through history is his base stealing was pretty lousy. I wonder how much of that was just the era he played in.
   353. Rally Posted: December 22, 2020 at 10:02 AM (#5995472)
No, that's not what stealing popups from infielders looks like. TotalZone infield ratings only look at ground balls. An infielder could catch a million popups, or none, and his TZ would remain unaffected. Call it a feature or a bug, but it is only measuring range in respect to ground balls.
   354. Rally Posted: December 22, 2020 at 10:08 AM (#5995476)
Before 2003 we don't have complete batted ball information, but it is mostly missing on hits, not outs. If an out was made we know if it was a flyball, grounder, liner, or popup. Or at least we have a code to that effect. The only way Andruw could steal TZ credit from an infielder is if he charged in front of the SS, fielded a ground ball, and got an out on the play.

But the only centerfielder I ever saw do that was not Jones, it was Eric Davis.
   355. snapper (history's 42nd greatest monster) Posted: December 22, 2020 at 10:11 AM (#5995478)
No, that's not what stealing popups from infielders looks like. TotalZone infield ratings only look at ground balls. An infielder could catch a million popups, or none, and his TZ would remain unaffected. Call it a feature or a bug, but it is only measuring range in respect to ground balls.

So what happens to pop-ups in that system?
   356. Rally Posted: December 22, 2020 at 10:29 AM (#5995483)
If caught by an infielder, they don't affect any fielder rating.
   357. snapper (history's 42nd greatest monster) Posted: December 22, 2020 at 10:30 AM (#5995485)
If caught by an infielder, they don't affect any fielder rating.

So then transferring 20 reasonably difficult pop-ups from IFs to the CF would increase the CF's ratings w/o changing the aggregate defensive efficiency at all?
   358. Rally Posted: December 22, 2020 at 10:36 AM (#5995489)
I think so. The DER would remain the same, as long as you are changing only from "caught by IF" to "caught by OF" and not "fell as a hit" to "caught by OF".

The TZ of infielders would remain the same. The TZ of the OF would increase.

But to make sure we stay on the same page, this scenario is not what we saw with the 1990s Braves. The infield defense on ground balls got worse, but the team DER improved anyway, so that strongly suggests improved defense in the outfield.
   359. DL from MN Posted: December 22, 2020 at 10:43 AM (#5995492)
An infielder could catch a million popups, or none, and his TZ would remain unaffected.


Is there a defensive system that actually measures this?
   360. snapper (history's 42nd greatest monster) Posted: December 22, 2020 at 10:55 AM (#5995495)
But to make sure we stay on the same page, this scenario is not what we saw with the 1990s Braves. The infield defense on ground balls got worse, but the team DER improved anyway, so that strongly suggests improved defense in the outfield.

Right, the OF DF improved, but not as much as Jones' individual numbers suggest. Pop-up stealing could account for some of that difference.
   361. Mefisto Posted: December 22, 2020 at 11:06 AM (#5995499)
But to make sure we stay on the same page, this scenario is not what we saw with the 1990s Braves. The infield defense on ground balls got worse, but the team DER improved anyway, so that strongly suggests improved defense in the outfield.


This is interesting. In general, FBs are easier to field than GBs. If (and it's a big if) the Braves' pitchers could induce more FBs without having them become EBH, then the team benefits. The IF would look worse and the OF would look better, with no net change (ceteris paribus).
   362. GuyM Posted: December 22, 2020 at 11:23 AM (#5995503)
Right, the OF DF improved, but not as much as Jones' individual numbers suggest. Pop-up stealing could account for some of that difference.

I'm not sure what you are comparing here -- the change in team defense is very consistent with Andruw's TZ rating.

And I don't see evidence that there was pop-up stealing. Again, after Andruw arrived the team's 2B and SS caught almost exactly the same number of flyballs per game.

ATL Putouts (flyballs):
Pre-AJ / AJ
2B 0.58 / 0.57
SS 0.52 / 0.48
   363. snapper (history's 42nd greatest monster) Posted: December 22, 2020 at 11:28 AM (#5995506)
I'm not sure what you are comparing here -- the change in team defense is very consistent with Andruw's TZ rating.

see [335] DER is not consistent with Jones being the greatest CF ever, by a mile. It's consistent with him being a great CF. But if Jones is merely great, like 5 or 6 other great CF, he has no business in the HoF. As "THE GREATEST" he's borderline.
   364. GuyM Posted: December 22, 2020 at 11:32 AM (#5995508)
363. See 338. DER was totally consistent with Jones' WAR rating, once you adjust for decreased IF performance after his arrival. I think Rally concurs.
   365. Rally Posted: December 22, 2020 at 12:00 PM (#5995512)
Yeah, I do concur. The way the DER numbers work out is, as Guy demonstrated, completely consistent with Jones being a +35 fielder, at least in his first few years as the regular CF. It is unusual for the data from different approaches to reconcile so well, but in this case it does.

The data are also consistent with what you would expect from aging, especially a guy who put on a lot of pounds early. From 1997-2001 he was a +25 to +35 fielder. From 2002-2007 more like +10 to +20. Then he got hurt, I'm sure the extra weight didn't help his recovery, and he wasn't a CF anymore.

2001 was also the last year he had double digit steals. He came up and was fast, plus had great instincts and read the ball better than anyone else. As his speed dropped off, he was still an excellent OF. Think Jackie Bradley Jr. - he's a plus defender even though his speed was never elite. Bradley's sprint speed has been between 27 and 28 ft/sec each year from 2015-2020. In scouting terms, he's a tick above average, but gets great jumps and reads and makes more plays than faster players.

The infield fly data Guy posted shows .05 fewer putouts combined for the SS/2B. Over 162 games, that works out to 8 catches. Jones might have been stealing some popups, but that is not a big part of what can be seen in the data. Most of it looks like turning hits into outs.
   366. GuyM Posted: December 22, 2020 at 01:28 PM (#5995530)
I don't think anyone is arguing that Andruw Jones was not a terrific outfielder. The question is whether he's more than a standard deviation better than the next best outfielder in the history of baseball.

I think this is really the core issue with Andruw: shouldn't we be skeptical of the claim that he was *by far* the best CF ever? We generally expect the best player in a category to surpass the 2nd-best by a little (Rose 4256, Cobb 4189) not by a lot -- that's how human talent is distributed. And that's fair enough: we *should* consider Andruw's with some skepticism. (I certainly have.) But skepticism should not mean dismissal -- sometimes exceptional things happen (Ryan 5714, RJ 4875). Here's why I think this objection fails to be persuasive.

First, Andruw's range rating is the best, but as Rally notes there is not a huge gulf between him and the next best CF. In runs saved per 150 games, it looks like this:
Andruw +15.7
Piersall +14.1
Blair +13.6
Devon +11.9
Pettis +11.5
....
Mays +7.5

Now, Jones also had an outstanding arm, which was not the case for all the others. It's certainly rare for a player to combine a great arm, exceptional jumps, and good (not elite) speed, as Andruw did. But is it impossible? No. Jones also discovered that he could play much shallower than other CF, an important innovation.

What drives a lot of the doubt about Andrew, I think, is that his WAR fielding rating is much better than Willie Mays'. That's seen as false almost by definition for many fans. But surprising <> impossible. For one thing, it's worth noting that 55% of Willie's fielding rating is based on performance after age 30, compared to 4% for Andruw -- so Andruw's youth is being compared to Willie's entire career. More importantly, what WAR (TZ) is really saying is that Willie wasn't as elite in the field as is commonly believed. Maybe TZ is wrong about Willie for some reason, but *that* is what people should be examining, rather than downgrading Andruw to avoid an uncomfortable comparison.

And, if we want to rule conclusions out because they seem unlikely, we should probably start not with Andruw>Willie but rather with the claim that the CF who hit 661 HRs also had the best glove in history. Given that fielding and hitting skills are negatively correlated, the odds of that being true are astronomically small. When I was growing up, I also learned that Mantle was an elite fielder -- but the data strongly suggests that wasn't true (even before his knees gave out). The reality is that the fielding ability of great hitters has consistently been overrated, as we can see clearly in the history of GG awards. It would not be surprising at all if Willie was "only" an excellent CF, rather than the best of all time. And again, even if you believe Willie's rating is too low, that's not a reason to punish Andruw.
   367. Mefisto Posted: December 22, 2020 at 01:58 PM (#5995538)
It's not just that Jones shows as better than Mays, it's that he shows as *20%* better than any other OF in history by TZ Runs (253 v 205 for Clemente). And we can't say that Andruw had a better arm than Clemente (I doubt Andruw's arm was better than Mays').

Then there's the fact that Andruw accumulated that total in much less playing time than most of his competitors. He has 26% more TZ runs than Paul Blair in similar playing time; maybe the arm accounts for that (I can't check it right now), but that's a lot.

I emailed Chris to see if he wanted to be heard on this, so maybe he'll join in.
   368. GuyM Posted: December 22, 2020 at 02:08 PM (#5995539)
It's not just that Jones shows as better than Mays, it's that he shows as *20%* better than any other OF in history by TZ Runs (253 v 205 for Clemente).

Jones rating better than Clemente would not make my list of top 10 concerns about Andruw's rating. Right fielders just don't have enough opportunities to match the range rating of the top CF. The fact that Clemente is +120 on range alone is amazing (close to Devon White) -- I assume that's by far the best in RF.

And we can't say that Andruw had a better arm than Clemente (I doubt Andruw's arm was better than Mays').

TZ agrees: Clemente is +84 with his arm, compared to +57 for Andruw (and +49 for Willie, +31 for Blair).
   369. Mefisto Posted: December 22, 2020 at 02:37 PM (#5995544)
And, if we want to rule conclusions out because they seem unlikely, we should probably start not with Andruw>Willie but rather with the claim that the CF who hit 661 HRs also had the best glove in history.


The match of HR and glove isn't that unusual. Mays is "only" 7th among all OF in TZ runs, and 2 of those ahead of him are Bonds and Yaz.
   370. Rally Posted: December 22, 2020 at 03:50 PM (#5995565)
Jones also discovered that he could play much shallower than other CF, an important innovation.


I wouldn't call that an innovation. Tris Speaker was doing that 90 years before Jones. A lot of stories about Joe Joe DiMaggio playing shallow. I'm sure there are others.
   371. Rally Posted: December 22, 2020 at 04:30 PM (#5995577)
One very fair criticism is that the TZ for Willie Mays is based on a completely different method than for Jones, simply because of data availability.

I have played around with some alternative defensive metrics with the goal of keeping things simple so the same code can be used for any year. And since I published TZ, Retrosheet has added a lot of back years, at last check they had every season back to 1916.

The first one is mostly what range factor should have been, using the denominator of balls in play instead of innings. I don't have park adjustments, the only thing I've added to a simple calculation is to split plays vs. LHB and RHB, so a third baseman won't be penalized if his team sees a larger than normal share of lefty hitters. The numbers are bigger than what we've come to expect as the range between good and bad defenders. Anyway, in CF Jones ranks 6th:

Ashburn +595 (376 home, 218 road, Bill James pointed out the advantage his home park gave him many years ago)
Max Carey +374
Mays +361
Flood +323
Blair +321
Andruw +320
Mike Cameron +315
Garry Maddox +295

The other metric produces much more conservative numbers. What it does is to attribute a share of all hits to a fielder along with plays made. It reconciles to Team DER.

Mays +268
Blair +215
Cameron +202
Ashburn +191
Flood +185
Andruw +178
Carey +144
Willie Davis +143
Dwayne Murphy +140

Even this conservative measure though works out well for Jones. With a .85 run value that's +151, but keep in mind he's also +57 in throwing. So using that would change his rfield from +235 to +208, making him a 60 WAR guy instead of 63. And to get there you don't need to believe that he is far above every other great outfielder, he's just one among the elites.

Whether a 60 or 63 WAR for a player who did almost nothing after the age of 30 is good enough for the hall is up to you. I'm in the middle here. I would not oppose him going in, but I don't think he's the best candidate either.

   372. Mefisto Posted: December 22, 2020 at 04:39 PM (#5995580)
If even your more conservative metric is valid, then I guess that may answer the question DL in MN asked in 236.
   373. Rally Posted: December 22, 2020 at 04:45 PM (#5995581)
By rate stats, Andruw isn't even the best outfielder to debut in 1996. That would be Darin Erstad. Looking at defensive runs per inning for all outfield positions, Andruw comes out at +22 per 162 games. Erstad is +26. But he hasn't even healthy enough to play half as many innings as Andruw.
   374. Rally Posted: December 22, 2020 at 04:54 PM (#5995585)
It's a long way from being valid. Just something I've played around with after seeing how much new old stuff is available on Retrosheet.

   375. GuyM Posted: December 22, 2020 at 05:43 PM (#5995587)
One of the odd things about May's TZ ratings is that he didn't excel in his youth as much as you'd expect:
Age / TZ (per 150G)
23-29 +11.2
30-35 +16.1
36+ +0.2

Can you tell if your new metrics display a similar or different pattern?
   376. Eric J can SABER all he wants to Posted: December 22, 2020 at 07:39 PM (#5995604)
One of the odd things about May's TZ ratings is that he didn't excel in his youth as much as you'd expect:
Age / TZ (per 150G)
23-29 +11.2
30-35 +16.1
36+ +0.2

Can you tell if your new metrics display a similar or different pattern?


This could partly be a park thing; the Giants moved into Candlestick Park in 1960 (Mays's age 29 season). The Polo Grounds in particular had strange enough dimensions that they could easily have messed with defensive ratings of outfielders.
   377. Mefisto Posted: December 22, 2020 at 07:48 PM (#5995606)
They also played in Seals Stadium in 1958-9.

Mays said in one of his biographies that it took him a year to figure out the winds at Candlestick.
   378. alilisd Posted: December 22, 2020 at 08:41 PM (#5995611)
I wouldn't call that an innovation. Tris Speaker was doing that 90 years before Jones. A lot of stories about Joe Joe DiMaggio playing shallow. I'm sure there are others.


Growing up I used to hear announcers say this about Garry Maddox frequently.
   379. snapper (history's 42nd greatest monster) Posted: December 22, 2020 at 09:14 PM (#5995618)
Whether a 60 or 63 WAR for a player who did almost nothing after the age of 30 is good enough for the hall is up to you. I'm in the middle here. I would not oppose him going in, but I don't think he's the best candidate either.

He's Bobby Abreu or Jim Edmonds, basically. No reason to be upset any of them aren't in the Hall. Lofton is better than all of them.
   380. Rally Posted: December 22, 2020 at 09:30 PM (#5995622)
Can you tell if your new metrics display a similar or different pattern?


Yeah, I’ll do that when I get a chance.
   381. Rally Posted: December 22, 2020 at 11:01 PM (#5995637)
OK, first the outs per BIP method, per season

age, plays, (home/road)
23-29 +15 (9/6)
30-35 +33 (10/23)
36+ +4

Now the DER method

23-29 +11 (10/1)
30-35 +21 (11.5/9.5)
36+ +6

It's showing he was about the same in his home parks at 30-35 as he was at 23-29, but he got a lot better on the road.
So the same pattern.
   382. GuyM Posted: December 23, 2020 at 09:11 AM (#5995652)
This could partly be a park thing; the Giants moved into Candlestick Park in 1960 (Mays's age 29 season). The Polo Grounds in particular had strange enough dimensions that they could easily have messed with defensive ratings of outfielders.

In six seasons at the Polo Grounds Mays made 2.79 outs per 9 innings, while other SFG CF made 2.59. That +0.20 for Mays would be worth about +25 runs/season. But if we compare Mays' first two years at the Polo Grounds to the two preceding years (Bobby Thomson in CF), the gain is less impressive: Mays 2.95 vs Thomson 2.86, an improvement of 11 runs/season. Mays' range at the Polo Grounds was clearly well above average, but unclear to me whether it was otherworldly.

In his first 3 years in CA, Mays was actually no better than his backups in CF, but the non-Mays sample is quite small.

He's Bobby Abreu or Jim Edmonds, basically. No reason to be upset any of them aren't in the Hall. Lofton is better than all of them.

Yes, if you look only at WAR. But one could certainly argue that the best fielder of all time at a key defensive position, who was also a 111 OPS+, 400+ HR, 1200+ RBI hitter, has a stronger claim than other 65ish WAR players. I'd be happy to see Lofton go in -- just saying that Jones has a distinct reasonable claim to make beyond his total WAR.
   383. pikepredator Posted: December 23, 2020 at 09:55 AM (#5995663)
This discussion is flat-out amazing. Thanks to the contributors for all the work they are putting in.
   384. GuyM Posted: December 23, 2020 at 11:21 AM (#5995679)
Rally, two TZ questions for you, if you have the time:
1) B-Ref doesn't list TZ fielding results pre-1953, but does show pre-1953 WAR fielding. Do you know what the latter is based on?
2) Is the TZ catcher metric explained somewhere? Can you provide a link?
Thanks.
   385. Rally Posted: December 23, 2020 at 12:08 PM (#5995691)
1. JAARF - just another adjusted range factor
2. Sb/cs/pb/wp, compared to league average and adjusted for catching lhp and rhp
   386. GuyM Posted: December 23, 2020 at 12:24 PM (#5995694)
The match of HR and glove isn't that unusual. Mays is "only" 7th among all OF in TZ runs, and 2 of those ahead of him are Bonds and Yaz.

Among the top 150 players in career dWAR, only Schmidt (49th dWAR, 148 OPS+) and Mays (51st dWAR, 156 OPS+) were also all-time great hitters. It happens, but it's a rare combination. And as good as they were with the glove, I don't believe either was the all-time best fielder at his position.
   387. GuyM Posted: December 23, 2020 at 12:27 PM (#5995695)
385: thanks. Should we consider JAARF to have significantly more potential bias than TZ, or would you say similar reliability?
   388. Rally Posted: December 23, 2020 at 01:38 PM (#5995707)
385: thanks. Should we consider JAARF to have significantly more potential bias than TZ, or would you say similar reliability?


I consider it inferior to TZ. It's just a stand in for not having much data to go on, but needing to put something there for fielding if I wanted an all-time WAR leaderboard.
   389. alilisd Posted: December 23, 2020 at 02:09 PM (#5995712)
This discussion is flat-out amazing. Thanks to the contributors for all the work they are putting in.


Indeed! Thanks to the guys putting the numbers and discussion together!
   390. alilisd Posted: December 23, 2020 at 02:25 PM (#5995716)
The match of HR and glove isn't that unusual. Mays is "only" 7th among all OF in TZ runs, and 2 of those ahead of him are Bonds and Yaz.

Among the top 150 players in career dWAR, only Schmidt (49th dWAR, 148 OPS+) and Mays (51st dWAR, 156 OPS+) were also all-time great hitters. It happens, but it's a rare combination


I don't know that this is really a fair way to measure this though, as so many players give back a lot of dWAR late in their careers. Griffey, for example, through age 30 is at 11.6 dWAR, with over 400 HR and 1200 RBI, plus leading the league in HR 4 times. Now after 30 he gives back 9 dWAR, but he was an elite fielder and a HR bat throughout his peak/prime.

Actually, maybe I should scratch all of that and just ask you how you ranked the players, because I do not get either one of those guys at 49 and 51 for dWAR. Did you use rField? Nope, that doesn't seem to be it either. FanGraphs instead of B-R? Anyway, it doesn't seem quite right. A-Rod would fit this bill, maybe Mathews, Banks, and probably a good few more who were much better bats, though perhaps lesser fielders, than Jones. Guess it depends on how high you're setting the bar for the two components. But I'm not sure how you're ranking career dWAR as it's not matching up with what I'm getting using Stathead.
   391. GuyM Posted: December 23, 2020 at 02:56 PM (#5995722)
390. Ranking career dWAR, 1918 to present, 1000+ games.

Yes, Griffey was much better defensively before age 30, as were some other players. If we take the top 100 in DWAR/season through age 30, three were elite hitters: Schmidt, Bonds, and Banks. Griffey doesn't make the dWAR cut (he's #131). There are a bunch more who were very good fielders in their 20s (ARod, Hornsby, Boggs, Mays, Brett), but none of them make the top 100. So I think it's still fair to say that our Bayesian prior should be *not* to expect a top-20 all time hitter to also be a top-20 all time fielder. Not impossible, but highly unlikely.

And I don't think you can seriously dispute that the players most overrated defensively tend to be very good or great hitters. Alomar (-37 rfield) and Joe Morgan (-48) don't win 15 GG at 2B if they don't have good bats. Same for the 15 GG won by Mattingly (+33) and JT Snow (-22). Jeter (-253) surely doesn't win 5 GG if he isn't an elite hitter for a SS. And then there's Winfield (7 GG, -91 runs), Puckett (6, -14), Finley 5, -1), Gwynn (5, +6), and Dale Murphy (5, -33) in the OF.
   392. Bleed the Freak Posted: December 23, 2020 at 03:33 PM (#5995736)
Echoing the others, great discussion all.

Adding that Kiko's stat utilizes Retrosheet data, 1916 to 2019.

Putting the cutoff at #41, Rally's mention spot on here for Darin Erstad, an awesome per game rating.
https://baseball.tomthress.com/Leaders/Leaders.php?
ex=0&y=&y1;=&y2;=&l=&a=c&s=f8&w0=0.333333333333&w5=0&wl=0.333333333333&ws=0.333333333333
&p0=0.333333333333&p1=0.333333333333&p2=0.333333333333&n=99

Andruw neck and neck for #1 all-time

eWins eLosses Net Wins Rk Player
72.3 64.2 8.1 01 Amos Otis
85.6 77.6 8.0 02 Willie Davis
71.2 63.5 7.7 03 Andruw Jones
65.4 57.9 7.5 04 Curt Flood
51.2 44.5 6.8 05 Duke Snider
58.1 51.7 6.3 06 Paul Blair
69.1 62.9 6.3 07 Jim Edmonds
74.7 68.6 6.1 08 Max Carey
51.7 46.0 5.6 09 Vince DiMaggio
63.4 58.1 5.3 10 Mickey Mantle
66.4 61.2 5.2 11 Joe DiMaggio
63.6 58.6 5.0 12 Carlos Beltran
29.2 24.4 4.8 13 Leonys Martin
45.6 40.9 4.7 14 Jim Piersall
37.7 33.0 4.7 15 Johnny Mostil
41.1 36.7 4.4 16 Dave Henderson
58.7 54.3 4.4 17 Dom DiMaggio
23.3 19.0 4.3 18 Kevin Kiermaier
43.3 39.0 4.3 19 Andre Dawson
39.6 35.3 4.3 20 Gary Pettis
44.9 40.7 4.2 21 Taylor Douthit
71.5 67.7 3.8 22 Kenny Lofton
36.1 32.5 3.7 23 Mike Trout
60.8 57.2 3.6 24 Devon White
66.8 63.2 3.5 25 Mike Cameron
20.8 17.3 3.4 26 Juan Lagares
24.8 21.4 3.4 27 Brian L. Hunter
18.5 15.1 3.4 28 Franklin Gutierrez
48.2 44.9 3.4 29 Dwayne Murphy
42.0 38.8 3.2 30 B.J. Upton
27.2 24.0 3.2 31 Jigger Statz
26.2 23.1 3.1 32 Ben Chapman
20.6 17.6 3.0 33 Happy Felsch
25.2 22.3 3.0 34 Vic Davalillo
37.2 34.3 2.9 35 Cesar Geronimo
18.1 15.2 2.9 36 Hal Jeffcoat
73.7 70.9 2.8 37 Lloyd Waner
11.7 9.0 2.7 38 Carl Furillo
68.1 65.4 2.7 39 Sam West
16.6 14.0 2.7 40 Endy Chavez
21.5 18.8 2.7 41 Darin Erstad

At this rate, Kiko has Jones as HOF level worthy
https://baseball.tomthress.com/Leaders/UberLeaders.php

For a normal size HOF, if he's in the discussion for greatest ever in CF or close, he's at least high 50s/low 60s WAR with a strong peak/prime, he's worthy for me.
   393. alilisd Posted: December 23, 2020 at 05:51 PM (#5995750)
390. Ranking career dWAR, 1918 to present, 1000+ games.

Yes, Griffey was much better defensively before age 30, as were some other players. If we take the top 100 in DWAR/season through age 30, three were elite hitters: Schmidt, Bonds, and Banks. Griffey doesn't make the dWAR cut (he's #131). There are a bunch more who were very good fielders in their 20s (ARod, Hornsby, Boggs, Mays, Brett), but none of them make the top 100. So I think it's still fair to say that our Bayesian prior should be *not* to expect a top-20 all time hitter to also be a top-20 all time fielder. Not impossible, but highly unlikely.


But how are you defining top 20? Is it on a seasonal basis? Career, peak, or prime? By your criteria of dWAR the list ends up being filled with a bunch of guys who were up the middle defenders for a reasonably long career. Now that's in part due to them being very good to great, at least at times, defenders, but it's largely due to positional adjustment, not necessarily to being a great defender. Think of guys like Garry Templeton, or Rick Burleson, or Freddie Patek, or Leo Cardenas, or Butch Wynegar.

Using career rField instead brings Clemente to 6th, with 3 batting titles and a 10 year prime OPS+ of 150. Mays 8th, Yaz, 9th, Bonds 13th, Kaline 20th, and Pujols 26th. Schmidt is still there, Boggs is 64th, Aaron 74th. Is Sosa an elite bat, Jackie Robinson? Ruth is 106th. Totally on board with elite glove plus elite bat being rare, it's why HOF players are rare, it's why 5 tool players are rare. I just don't think it's quite as unusual as you're making it out to be. Apologies if I'm mischaracterizing your position. I am coming to this part of the discussion a bit late and missed some of the earlier posts on this I'm sure
   394. GuyM Posted: December 24, 2020 at 08:34 AM (#5995806)
Now that's in part due to them being very good to great, at least at times, defenders, but it's largely due to positional adjustment, not necessarily to being a great defender. Using career rField instead brings Clemente to 6th, Mays 8th, Yaz, 9th, Bonds 13th, Kaline 20th, and Pujols 26th.

But the positional adjustment is there for a reason: the guys who play SS really are better fielders on average than those who play 3B, and vastly better than the guys in LF or RF. And rfield measures players only against others at their position. If we take the top 20 career hitters (rbat), 13 were above average fielders at their positions but just 4 were above-average *defenders* (dWAR): Bonds, ARod, Hornsby, and Mays. On average they are -7 wins in the field. I mean I love Pujols, but the idea that he's the 26th-best fielder in MLB history is absurd on its face. You could probably find 30 minor league SS today (if we had minor leagues) who are better fielders than Albert in his prime.

All I'm saying is that we should not expect truly elite hitters to also be elite fielders very often. That means we should be shocked if the best fielding SS or C were an elite hitter, very surprised if the best CF/2B/3B is elite at the plate, and somewhat surprised in LF/RF/1B. And I think history largely confirms that. Obviously, it would be less shocking for the best hitter and best fielder at an individual position to be the same person, but still surprising. (I'm defining "best fielder at a position" as something like the best 10-year peak, since, as you point out, poor late-career performance doesn't make you worse in your 20s; plus, a player's ability to keep accumulating rfield in his 30s depends in part on how good his bat is).

Anyway, I don't believe we should rule out any claim just because it's unlikely. But it should affect our skepticism and demand for evidence. And the idea that the best fielder ever to roam CF is also the 7th most productive hitter in history (Mays) is much, much less likely than the possibility that it's hitter #381 (Andruw). But only one of these claims is viewed with skepticism.
   395. Mefisto Posted: December 24, 2020 at 10:59 AM (#5995834)
I think the term "best" is a bit ambiguous here. It could mean "highest peak" or it could mean "longest period of excellence". I always understood the case for Mays to be the latter. I've never seen any data to suggest that he had the highest peak of any CF, though observational reports might suggest that.
   396. Joyful Calculus Instructor Posted: December 24, 2020 at 11:23 AM (#5995838)
Some mentions of Banks made me decide to look up his defensive numbers. I thought he was considered a middling fielder, which is why he got moved to first base halfway through his career, but I see that he was actually an excellent short stop and still above average at the time he got moved.
   397. GuyM Posted: December 24, 2020 at 11:27 AM (#5995839)
I think the term "best" is a bit ambiguous here. It could mean "highest peak" or it could mean "longest period of excellence". I always understood the case for Mays to be the latter.

That's certainly a valid approach too. That would lift Mays much closer to the #1 spot (though not quite there, it appears). This definition will of course favor great hitters, since they have longer careers. For example, Yaz or Bonds would rank #1 in LF rather than Alex Gordon. So it then becomes more likely that one player could be both the best hitter and best fielder at a position (Bonds and Mays have the best cases, I suppose).
   398. alilisd Posted: December 24, 2020 at 12:47 PM (#5995849)
But the positional adjustment is there for a reason: the guys who play SS really are better fielders on average than those who play 3B, and vastly better than the guys in LF or RF. And rfield measures players only against others at their position.


I look at it somewhat differently. The guys who play SS are not necessarily better fielders, they do have a different skill set than other positions though. An elite SS might not be able to play at an elite level in the corner OF. They may not have the top end speed to get to difficult fly balls even though they have quick feet and reaction times to get to hard hit ground balls. They may not have the arm to make the long throws required although they may have enough arm and a quick release for the infield. It's certainly more important to have a quality SS than a quality corner OF because of the number of balls hit, but it's not necessarily that the SS is a better fielder per se, although they do have a different skill set required of them, and you can't hide a bad defender at SS the way you can in LF.

The SS is not necessarily a better fielder then a 3B, but they likely have more range. Top fielding 3B have the glove and the arm, but may not have the range (although they may, think Machado, or A-Rod when he moved to 3B) of a SS. 2B likely have similar range and glove, but may not have the arm. Both 2B and 3B at the top end are just as good at fielding, but range is the differentiator. If you're going to look at it as ONLY the best defenders can play SS because range is crucial to the discussion for infield (and frequency of balls in play relative to OF), then you would be right to use positional adjustment, but then there cannot be ANY elite fielders at any other position because they didn't have the range to play SS (or at least they didn't display it on the field because they played elsewhere). Pujols is an apropos player for this discussion. Could he play an up the middle position, clearly not. However, I don't think there is any doubt that he is a stellar defender at 1B. Does this mean he's not an elite defender over all though? I would say he is, but of course YMMV.

Anyway, I don't believe we should rule out any claim just because it's unlikely. But it should affect our skepticism and demand for evidence.


Definitely agree
   399. GuyM Posted: December 24, 2020 at 02:16 PM (#5995856)
The guys who play SS are not necessarily better fielders, they do have a different skill set than other positions though.

Clearly there is not just a single dimension to fielding skill. An excellent 3B might not be even better in LF, even though good 3B are harder to find than good LF. So that's fair enough. But the position adjustment accounts for two crucial realities:
1) the average fielding ability at SS/2B/3B/CF is much more scarce than the fielding ability found at 1B/LF/RF. That has big consequences for teams in terms of reduced offense at those positions. So unless you really believe there are a bunch of 1B and LF capable of providing adequate defense at 2B, and MLB teams are just blind to this, you have to account for this value difference.
2) If you rely on rfield alone, then 1B/LF/RF are being compared to much inferior players.

Albert Pujols was an excellent first baseman, but I don't see how he can be an "elite defender" when we know there have been hundreds and hundreds of professional players with a better glove. It's like saying Phil Rizzuto (93 OPS+) was an elite hitter because he's the 5th-best 8-hole hitter in history, or Steve Avery was a great pitcher because he was the best fifth starter of the 20th century.

At a minimum, I would put all OFs in one pool, and all IFs in another. DWAR favors CF, as it should, but there are a number of corner OF among the top 25 OF (Barfield, Heyward, Clemente, Jordan, Bonds, Gordon). And a number of 2B and 3B among the top 25 non-OF (though C and SS are favored). If one had to choose between rfield and dWAR, certainly dWAR comes closer to measure true fielding value (though neither is perfect).
   400. Fred Lynn Nolan Ryan Sweeney Agonistes Posted: December 24, 2020 at 02:31 PM (#5995858)
Reaching back to add -

I used to be one of Wilhelm's greatest boosters as Best Reliever Ever, but some people helped me see that his career ERA+ of 147 is effectively more like 133 when you add in the disproportionately large # of unearned runs; largely his fault, as his catchers were charged with oh-so-many passed balls.


- knuckleballers also tend to be extreme ground-ballers, leading to more infield errors behind them and more unearned runs that way.

I've long suspected this was part of the historical bias against knuckleballers: managers couldn't necessarily quantify it, but it always seemed like with a knuckler on the mound things would go wrong.
Page 4 of 5 pages ‹ First  < 2 3 4 5 > 

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Dynasty League Baseball

Support BBTF

donate

Thanks to
Randy Jones
for his generous support.

You must be logged in to view your Bookmarks.

Hot Topics

NewsblogTo dream the impossible dream - and then decide it's time to let it go
(50 - 8:13pm, Oct 03)
Last: Zach

NewsblogHector Lopez, Who Broke a Baseball Color Barrier, Dies at 93
(24 - 8:12pm, Oct 03)
Last: BDC

Newsblog2022-23 Preseason NBA thread
(362 - 8:11pm, Oct 03)
Last: Fourth True Outcome

NewsblogShohei Ohtani to make $30 million in 2023, record amount for arbitration-eligible player
(17 - 8:07pm, Oct 03)
Last: Never Give an Inge (Dave)

NewsblogOMNICHATTER for the week of September 26 - October 5, 2022
(202 - 8:05pm, Oct 03)
Last: cardsfanboy

NewsblogThree impacts of baseball's new 12-team postseason format
(22 - 7:30pm, Oct 03)
Last: Ron J

NewsblogRoger Maris Jr. blasts MLB, says Aaron Judge’s potential 62nd home run should be single-season record
(122 - 7:06pm, Oct 03)
Last: SoSH U at work

NewsblogTony La Russa expected to announce retirement Monday as White Sox manager
(15 - 7:02pm, Oct 03)
Last: asinwreck

NewsblogAtlanta Braves sign RHP Charlie Morton to 1-year, $20 million deal for 2023
(4 - 5:16pm, Oct 03)
Last: The Yankee Clapper

NewsblogOT Soccer Thread - European Leages Return
(304 - 5:14pm, Oct 03)
Last: AuntBea odeurs de parfum de distance sociale

NewsblogThere's a new longest last name in MLB history
(15 - 4:19pm, Oct 03)
Last: Edmundo got dem ol' Kozma blues again mama

NewsblogTrevor Bauer’s defamation case against accuser’s former attorney could be thrown out
(30 - 3:40pm, Oct 03)
Last: The Yankee Clapper

NewsblogOT: Wrestling Thread November 2014
(2543 - 2:26pm, Oct 03)
Last: Tubbs is Bobby Grich when he flys off the handle

NewsblogOT - October 2022 College Football thread
(48 - 1:49pm, Oct 02)
Last: AuntBea odeurs de parfum de distance sociale

Sox TherapyPredictions of Ridiculousness
(143 - 12:59pm, Oct 02)
Last: Captain Joe Bivens, Pointless and Wonderful

Page rendered in 1.1462 seconds
48 querie(s) executed