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 29, 2005

Ernie Lombardi

Ernie Lombardi

Eigible in 1953.

John (You Can Call Me Grandma) Murphy Posted: May 29, 2005 at 12:26 AM | 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 29, 2005 at 12:49 AM (#1368850)
IMO, he's a borderline candidate. The man could hit, though.
   2. CraigK Posted: May 29, 2005 at 01:31 AM (#1368914)
The thing I like most about Lombardi is that he is almost universally regarded as one of the slowest players in baseball history (8 career SB!), yet he managed to hit .306 in his career.

Also, how did he manage to come up with 27 career triples, being so slow?
   3. sunnyday2 Posted: May 29, 2005 at 03:14 AM (#1369051)
As different as they are, I've always kind of thought of Bresnahan and Lombardi as comparable. Both borderline HoFers--not the worst Cs, Ferrell is not even borderline, but further from being the best than from being the worst. Bresnahan has been anywhere from 20-35 on (or off ) my ballot over the years. Can't see Lombardi as any better than that, probably a little bit worse.
   4. Michael Bass Posted: May 29, 2005 at 03:31 AM (#1369077)
Not nearly enough peak for me. Simply a horrible, horrible defender apparently.
   5. karlmagnus Posted: May 29, 2005 at 02:00 PM (#1369580)
My system rather likes the Schnozz; he's tentatively ranked #10 on my ballot, just ahead of Schang, who appears fairly similar. But I'm open to argment.
   6. OCF Posted: May 30, 2005 at 04:26 AM (#1371291)
... just ahead of Schang, who appears fairly similar.

For offense only, my system also sees Lombardi and Schang as fairly similar, but with Schang slightly ahead. Since I have a hard time seeng Lombardi with a defensive edge on anyone, and since Schang is not on my ballot, then Lombardi won't be on my ballot, either.

In his two versions of the Historical Baseball Abstract, it's quite a motley cast of characters that Bill James found worthy of treating at great length, including Hal Chase, Mike Donlin, and Ernie Lombardi. By those standards, who from our own times might call for such treatment? Julio Franco? David Wells?
   7. Howie Menckel Posted: May 30, 2005 at 10:48 AM (#1371396)
I can't wait to see if people give Julio Franco Mexico/Japan credit.
If they do, he could be a HOMer, no?
   8. John (You Can Call Me Grandma) Murphy Posted: May 30, 2005 at 01:31 PM (#1371428)
If they do, he could be a HOMer, no?

Uh..., no.
   9. Howie Menckel Posted: May 30, 2005 at 03:14 PM (#1371492)
Well, I actually meant "on HOM ballots," but I'll save that argument for another 50 'years' or so.
   10. John (You Can Call Me Grandma) Murphy Posted: May 30, 2005 at 03:23 PM (#1371500)
Well, I actually meant "on HOM ballots," but I'll save that argument for another 50 'years' or so.

Maybe 75 more years. ;-)
   11. Brent Posted: June 05, 2005 at 03:34 AM (#1382277)
Ernie Lombardi’s five seasons in the Pacific Coast League deserve some consideration. He played for his hometown Oakland Oaks from 1927-30, where he was a teammate of Buzz Arlett, and returned in 1948 for a last season split between Sacramento and Oakland.

As a 19-year old in 1927 he played 16 games, going 3-for-20 with 1 home run. Just a bit player, but it was enough to get a ring for the Oaks’ first PCL pennant.

For the next three seasons, however, he was one of the best hitters in the league; here are his actual PCL statistics with the Oaks:

Yr __Age _G _AB _R _H 2B 3B HR RBI SB AVG SLG SH
1928 20 120 318 39 120 27 03 08 047 01 .377 .557 07
1929 21 164 516 70 189 36 03 24 109 03 .366 .587 15
1930 22 146 473 76 175 32 04 22 105 01 .370 .594 11

In 1928 he ranked 3rd in the league in batting average among players with at least 300 AB, in 1929 he was 8th, and in 1930 he was 4th.

Calculating his MLEs was straightforward, reflecting the same run environment factors that I’ve previously used for Arlett. Oakland was a pitcher-friendly environment; I estimate that its run environment was 97.5 percent of the major league average for 1928, 94 percent for 1929, and for 1930 I estimate that its run environment was 90.6 percent of the National League average. (Following Chris Cobb’s convention, I converted his record for the 1920s to a major league average environment, and for 1930 to a National League average environment.) Here are his MLEs:

Year __G _AB _R __H 2B 3B HR RBI BB AVG OBP SLG OPS+

1928 096 259 28 091 21 02 06 033 18 .351 .394 .517 136
1929 125 403 49 139 27 02 17 077 30 .345 .390 .548 134
1930 112 373 56 132 25 02 16 077 27 .354 .398 .560 128

Are these estimates plausible? Lombardi’s career MLB batting average was .306 and his highest actual seasonal batting average was .343, so these averages may seem high. However, keep in mind that his major league career was played in the low scoring and relatively low average National League of the 1930s, whereas these MLEs are translated to the high average environment of 1928-30. The OPS+ statistics, which adjust for hitting environment, are in line with the rest of his career (career OPS+ of 125, with seasonal highs of 161, 153, and 147). Similarly, the line-drive power that shows up in the rest of his career is also apparent in these MLEs. Taking account of the changing run environment, these MLEs seem consistent with Lombardi’s MLB statistics.

As in the major league, the PCL data suggest that his fielding was average to below-average. For 1928 (his first full season) his fielding percentage was .953, the lowest among regular PCL catchers. For both 1929 and 1930 it was .975, about average for the league. He led the league’s catchers in assists both seasons, but also led in games caught both seasons; several other catchers had higher rates of assists per game.

To calculate fielding win shares, I will use his major league average of 1 fielding win share for each 35.7 games; I think this is conservative, since, at least for 1929-30, he was probably a better catcher than his major league average.

Here are my win shares estimates:

Year bWS fWS WS
1928 11.2 2.7 14
1929 16.1 3.5 20
1930 13.7 3.1 17

Lombardi’s final PCL experience was in 1948 at the end of his career, when as a 40-year old he played for Sacramento and Oakland. With Oakland, he was one of several former major leaguers managed by Casey Stengel who led the Oaks to their third pennant. I haven't attempted to calculate MLEs for this swan-song season.
   12. Brent Posted: June 05, 2005 at 03:51 PM (#1382574)
A correction - the Oaks' 1927 pennant was their second pennant, not their first (which they won in 1912).
   13. Joey Numbaz (Scruff) Posted: June 14, 2005 at 11:12 PM (#1404488)
Okay, am I missing something here? Looks like I have a new pet candidate. I posted this on the 1953 ballot thread, in response to comments about his OPS+ being more SLG than OBP . . . I'll repost here:

Regarding Lombardi and David Jones comments . . .

Defense? For a catcher in an era where the league leaders stole 25 bases a year? I'll take the hitter. 80% of a catcher's defensive value is just from being back there and letting another hitter play somewhere else. It's probably closer to 90% in the era Lombardi played in.

Plate discipline? First, who cares? We aren't projecting the guy, we are looking at the results. Even so, it's hard to knock plate discipline from a guy that struck out 262 times in his career, isn't it? The guy struck out more than 20 times in his career 3 times. His plate discipline was fine, he just hit everything.

His OBP was +.021. His SLG was +.074. I love Schang, he's just off my ballot, but he was not the hitter Lombardi was. His OBP was +.044, his SLG, +.018. Giving OBP double weight, you get Schang at +.106, Lombardi at +.116. Schang's defense was no prize either.

Using basic runs created from BB-Ref, Schang created 5.50 per 27 outs in a .280/.349/.383 environment. Lombardi created 5.96 in a .274/.337/.386 environment. There careers were of similar length, but Lombardi caught more. Lombardi was very definitely (new phrase?) the better hitter.

I think they are close, but Lombardi was clearly the better player, and is being vastly underrated by this group.

*********

I came to this conclusion with no thought of the minor league credit Brent mentions.

You've got a CATCHER that hit an awful lot like Jake Beckley (I realize he didn't triple). Their AVG/OBP/SLG are remarkably similar - though Schnozz of course did it in far fewer PA (only about 55% career length). But still, he smoked the ball like a Hall of Fame 1B and caught 1500 games. To me, that makes him one of the greats.

Compare him to Gabby Hartnett, a slam dunk HoMer.

Hartnett played in a much better offensive environment, and was a slightly better hitter, +.023 OBP, +.080 SLG (Lombardi +.021/+.074) in about 2 seasons extra playing time. Hartnett's OPS+ was 126, Lombardi's 125, same type of hitter.

Are you telling me that catcher defense in an era where SB were dropping like a brick makes one of these guys a slam dunk and the other struggling to be in the top 30? I just don't get it.

Please, someone convince me that I'm wrong here. If not, you're going to be hearing a lot about him from me over the next 50 years . . . this is the cause I needed to get back into the swing of things.
   14. Mark Shirk (jsch) Posted: June 15, 2005 at 12:31 AM (#1404713)
Joe, how do you explain teh lack of peak in both WS and WARP? I am confliscted on Lombardi because he seemed to be a great hitter at a position where offense was at a premium, but then he had only a few 20 WS seasons. Is bill Jamesmissing something? Or maybe he is overrating catcher defense?

Right now I have Lombardi just below Schang based on Schang's having a longer prime and far below Bresnahan and Mackey.
   15. OCF Posted: June 15, 2005 at 12:32 AM (#1404716)
I've been using RC and RC/out from a Stats handbook. One thing that does to Lombardi: it zings him pretty hard for his GIDP. The comparison to Schang, or Hartnett - for those guys, GIDP was mostly not recorded in their careers. So maybe that's not completely fair, although Schang was clearly much faster than Lombardi.
   16. sunnyday2 Posted: June 15, 2005 at 12:52 AM (#1404783)
Lombardi was not a terrible choice for the HoF, given where they had previoyusly drawn the line. I see he and Bresnahan as being very comparable in value despite their obvious differences in every other way.

OPS+ both 126
DefWS B 55 L 52

My top catchers to date:

1. Gibson
2. Hartnett
3. Dickey
4. Cochrane
5. Santop
6. Ewing
7. Bresnahan
8. Mackey
9. Lombardi
10. Bennett

11. Schang
12. Clapp
13. Clements
14. Schalk
15. Ferrell
16. Kling
17. Petway
18. Carroll
19. Meyers
20. McGuire

Top 6 plus Bennett are both HoM and PHoM. Bennett very deserving in his own time but the line has definitely moved up. The latter 4 all very very close anyway.
   17. Joey Numbaz (Scruff) Posted: June 15, 2005 at 12:57 AM (#1404806)
Lack of a peak - probably the fact that he never had 500 AB in a season. He has a huge peak in 1938.

Also, I think both of those metrics overstate the badness of his defense. I think the 'range' on defense for catchers is much smaller than either metric does. I think those metrics overrate the catchers they think are good defensively, and underrate the ones they think are bad. So that's going to suppress him in both stats. I've made no secret that I'll take a catcher that can hit over a supposedly good glove man any time.

I think Lombardi should get zinged for his DP (as they were probably more because he was slow than anything else). But he still rates as a better hitter according the RC than Schang, even with the DP included.

Over his career, compared to Dickey (same era), he had 76 more PB, made 35 more errors and had 109 fewer assists. That works out to about 5 PB, 2-3 E and 7 or 8 fewer assists per year. Putting context factors aside, I just don't see that as a huge deal. He gets a ding for his D, but I don't think it should be a huge one. Just being 'there' is most of the value of a catcher.
   18. Joey Numbaz (Scruff) Posted: June 15, 2005 at 01:04 AM (#1404833)
Another thing about DP, they were going up for everyone during the 1910-1940 period. In the deadball era, just about everyone could run at least a little, they sacrificed or hit & ran or tried to steal just about every time someone got on base.

As the home run came into play, teams started doing these things a lot less, and the players got slower and more powerful and DP totals went up. So while Lombardi certainly hit into more DP the Schang or Hartnett, he also had many more opportunities to as well.
   19. Joey Numbaz (Scruff) Posted: June 15, 2005 at 01:06 AM (#1404844)
"But he still rates as a better hitter according the RC than Schang, even with the DP included."

Actually, I was assuming the B-R data used DP. If it doesn't I could be off there.

I don't have access to my Sourcebook right now, could someone post the career RC/27, and league RC/27 for Schang, Lombardi, Bresnahan, Cochrane and Hartnett?
   20. Joey Numbaz (Scruff) Posted: June 15, 2005 at 01:17 AM (#1404880)
Nice list sunnyday. I agree with your catcher assessment pretty well, I'd bump Lombardi and Schang ahead of Bresnahan because they caught a lot more. But other than that I agree. And I love that you have John Clapp that high. He's not HoMer but he was a really good player.
   21. DavidFoss Posted: June 15, 2005 at 01:24 AM (#1404897)
Here is what Lee Sinin's encyclopedia spits out. He says that OWP is park adjusted, but I don't know aobut the other stuff. I'm guessing no because of the way that RC27/lgRC27 is not monotonic with OWP (see Cochrane/Dickey).

Anyhow, I'll post what I have. Should be able to back-calculate the lgRC27 from RC27 and OWP.

RC/G
          RATE   PLAYER   LEAGUE     OWP      PA     
Bresnahan  141     6.35     4.51     .651     5373   
Cochrane   140     7.73     5.53     .643     6206   
Dickey     130     7.21     5.56     .644     7060   
Hartnett   127     6.57     5.17     .610     7297   
Schang     122     5.95     4.87     .595     6422   
Lombardi   116     5.64     4.86     .575     6347
   22. Joey Numbaz (Scruff) Posted: June 15, 2005 at 01:30 AM (#1404913)
Cool, if no one else posts it before she gets home in a couple of hours, I'll call my girlfriend and have her go through my All-Time Handbook to get the Stats, Inc. numbers for those guys. Or I'll post them tomorrow if she decides that's a stupid task . . .
   23. Joey Numbaz (Scruff) Posted: June 15, 2005 at 01:30 AM (#1404915)
That 'cool' was slang for, 'thanks David!' . . .
   24. DavidFoss Posted: June 15, 2005 at 02:54 AM (#1405187)
FWIW -- Here are the back-calculated-from-OWP contexts:

Bresnahan 4.65
Cochrane 5.76
Dickey 5.36
Hartnett 5.25
Schang 4.91
Lombardi 4.85

(these are contexts -- used to normalize the RC/27 above to get the OWP above)
   25. Chris Cobb Posted: June 15, 2005 at 03:53 AM (#1405298)
Over his career, compared to Dickey (same era), he had 76 more PB, made 35 more errors and had 109 fewer assists. That works out to about 5 PB, 2-3 E and 7 or 8 fewer assists per year. I just don't see that as a huge deal. He gets a ding for his D, but I don't think it should be a huge one. Just being 'there' is most of the value of a catcher.

It's not a HUGE deal: obviously, in the large majority of major-league baseball situations, if you can get Lombardi as your team's catcher, you want him. BUT for the purposes of the HoM, this difference is likely to be significant.

Win Shares awards Dickey 94.9 fielding win shares for his career, Lombardi 51.9. Dickey played more defensive innings: in Lombardi's innings, he earns 82.0 fielding win shares. So the difference between them is 30 win shares for their careers, or 2 fielding win shares per season over 15-year careers.

Joe D.'s math shows us 15 plays a year: errors, pb, missed assists, that Dickey makes that Lombardi doesn't. To that, I believe WARP and WS would find a number of missed put-outs as well. Bill James indicates that Lombardi's slowness cost his team putouts: "Lombardi was so slow that his speeed was a problem on defense, even at catcher. . . . Foul popups dropped a few feet from home plate, because he coudl not spring out of a crouch." Let's say that there are 5 lost putouts a year. That makes 20 plays that Dickey has over Lombardi. Are these 20 defensive plays worth 2/3 of a win per season? That seems quite likely to me: I'd even go so far as to say that they are probably worth more than that -- someone in the group probably has the empirical tools to estimate how much more. In any case, I think it is pretty likely that win shares is not overestimating the difference in defensive value here. (I would believe that WARP, which sees these plays as worth almost 2 wins a year, is overrating them, but it's possible they have the better estimate.)

So if Lombardi justly loses 2 fielding win shares per season when compared to a great defensive catcher, that's 30 win shares in a career.

If one goes on to consider that Dickey was 2 batting win shares per season better than Lombardi (roughly the WS assessment, which appears to be supported by OWP), then that's another 30 win shares in a career, and 4 full win shares per season, a significant peak and rate difference.

Bill Dickey is one of the great catchers of all time, so maybe being 75% of Dickey if Dickey is prorated to your career length or being 70% of Dickey with his greater durability figured in is enough to get a catcher elected.

Given the depth of the pool of eligibles at present, however, I think regarding Lombardi as a top candidate is not supported by the evidence. There are a number of eligibles right now who are almost as good as Bill Dickey. Ernie Lombardi is not one of them.

To make a persuasive case for him, one would need to show that he is the best catcher available. Mackey, Bresnahan, Schang, and Petway need to be addressed to make that case. Then one would need to show that catchers need to be rated more highly as a group than the electorate is now doing. I give catchers a bonus to peak and to career based on the difference between their average playing time and the average playing time at other positions, and only Mackey is making my ballot among current catcher eligibles. How much more of a bonus would it be reasonable to give to catchers??
   26. Joey Numbaz (Scruff) Posted: June 15, 2005 at 04:04 AM (#1405324)
"Foul popups dropped a few feet from home plate, because he coudl not spring out of a crouch." Let's say that there are 5 lost putouts a year"

Less. A foul out dropped is next to nothing, since the typical hitter is going to make an out 66-67% of the time later in the AB, not to mention that with an extra strike in the equation, this would likely be even higher. Every 3 missed putouts is probably an extra missed 'play' in reality.

Also, some of those missed plays the PBs and Es aren't necessarily outs lost, they are just bases lost, which is much less harm. Many of the errors were on bad throws I'd imagine.

WS says 82 WS for Dickey, 52 for Lombardi (in the same time). Assuming this makes Dickey a 'good' catcher and Lombardi a 'bad' catcher, and an average catcher 67 WS in those innings, I'd be willing to guess that the true impact was more like making Dickey say 75 WS and Lombardi 59 WS, as I think the range for quality of catcher D is too wide.

James even admits that he is just guessing on catcher D, and that no one really has how to value it pegged yet. I'd think everyone should be nudged towards the mean here.

I still think Ernie Lombardi is damn close to Dickey and Hartnett, basically the same hitter and a season or two short. This last factor may be erased by PCL credit. I think there is very little difference between these guys.
   27. Chris Cobb Posted: June 15, 2005 at 05:10 AM (#1405439)
Less. A foul out dropped is next to nothing, since the typical hitter is going to make an out 66-67% of the time later in the AB, not to mention that with an extra strike in the equation, this would likely be even higher. Every 3 missed putouts is probably an extra missed 'play' in reality.

I meant that Dickey would average 5 more put-outs a year than Lombardi, which I picked as a conservative number. It could easily be 10 per year. What that's worth to the team is debatable, but I think you are underestimating the value of these plays.

Certainly Lombardi's fielding limitations in this respect don't cost his team every time, because somebody else will make the put-out later, but if you have _every_ defender on the team falling short in this way, the defensive consequences would add up fast. (This is an example of what James talks about as the "false normalization of fielding statistics" -- 27 outs are always going to be recorded.) And you can hardly say that it doesn't matter that Dickey gets his team five outs that a weak defensive catcher doesn't provide. Turning an out into a strike may not be a huge net loss, but turning a strike into an out is a large net gain.

James even admits that he is just guessing on catcher D, and that no one really has how to value it pegged yet.

What he's guessing on most is the _baseline_. The value of one catcher to relative to an average catcher is easier to assess.

I still think Ernie Lombardi is damn close to Dickey and Hartnett, basically the same hitter and a season or two short.

There are several problems with this view.

1) The only measure so far identified that supports this view of Lombardi as a hitter is OPS+. OWP doesn't, batting win shares doesn't. EQA supports it for Hartnett, but Hartnett has the longer major-league career.

2) The look at playing time leaves out Lombardi's weakness on seasonal durability, which matters a great deal for a catcher's peak value. Here are his top 8 seasons in games caught, compared to Cochrane, Hartnett, and Dickey:

Lombardi -- 123, 120, 116, 111, 108, 105, 101, 100
Cochrane -- 137, 135, 133, 130, 130, 128, 124, 123
Dickey -- 137, 127, 127, 126, 126, 125, 118, 108
Hartnett -- 140, 136, 129, 125, 118, 117, 114, 110

Harnett averaged 13 more games per season, Dickey 14, and Cochrane 19.5. This creates a significant difference in peak value that adding PCL time won't address.

Lombardi's pinch-hitting closes this seasonal playing time gap somewhat, but he still trails in this regard.

3) You accept above that 15 win shares as a reasonable estimate of the difference in defensive value between Lombardi and Dickey. I think the difference is twice that, but consider the significance of a difference of 15 career win shares right now in the rankings. On this ballot, 15 career win shares is a very significant difference. On my ballot, that's worth about 10 places: it's the difference in my system between John Beckwith at 3 and Stan Hack at 13. That's near the top of the rankings where the players are spread out a bit. Go down into the 30-60 backlog, and 15 career win shares moves a player even father up or down the rankings. If Lombardi is behind Dickey on the hitting side by another 15 win shares (a difference that also leads to a bit lower peak and a bit lower rate stat), that 30 win-share difference could be enough to drop him well off the ballot.
   28. Joey Numbaz (Scruff) Posted: June 15, 2005 at 05:32 AM (#1405474)
But a 30-WS distance from Dickey, who was at the top of the ballot is much less ranking wise than a 30-WS difference from the guy that finished 5th, no? Dickey was far ahead of the rest of the ballot (except for Greenberg). There's no way that should account for 30 places in the standing (when going from the top of the ballot).

30-WS from #15 to #45, that's reasonable. But not from #1.

Your other points are well taken, especially the one on in-season durability.

But I find it hard to say that an MVP as a catcher loses peak value to more than a handful of people on this ballot. I just see a catcher with kind of offense as a very serious candidate, despite his defensive shortcomings. I have him well ahead of Schang and Bresnahan still at this point. I voted him 7th, not first. I can't fathom that he's off a ballot, out of the top 30 consensus wise.
   29. Joey Numbaz (Scruff) Posted: June 15, 2005 at 05:41 AM (#1405483)
"but turning a strike into an out is a large net gain"

Again, not as big as you'd think, because 2/3 of the time the guy is going to make an out anyway. 5 extra putouts, that would have been strikes, or 5 extra strikes, that would have been worth putouts are worth the equivalent of 1 2/3 outs, IMO.

"What he's guessing on most is the _baseline_. The value of one catcher to relative to an average catcher is easier to assess"

I 100% disagree, I think it's the exact opposite. He's guessing on the differences between catchers. He has 10 point categories, 30 point categories, etc.. He's guessing as to the relative weights of each. He admits it's the biggest guesswork of the whole system.

I think the baseline is where he has it right, as catchers being slightly more valuable than shortstops on average, which is supported by the replacement level offense at each position. The guesswork is in distinguishing the good catchers from the bad ones, with stats that aren't well suited to the task.

Giving him credit for a few years in the PCL would bring his career value nearly equal to the big guns, with a somewhat lower peak. That should be more than enough to get him in in my opinion.
   30. OCF Posted: June 15, 2005 at 05:43 AM (#1405484)
What David Foss said, from a different source - the Stats Handbook:

             PLAYER   LEAGUE       PA     
Bresnahan     6.07     4.04       5373   
Cochrane      7.60     5.08       6206   
Dickey        7.08     5.05       7060   
Hartnett      6.39     4.65       7297   
Schang        5.97     4.48       6422   
Lombardi      5.47     4.37       6347


Big differences in the "league" column - pitchers removed versus not?

I prefer looking at this stuff year-by-year rather than whole career.
   31. Joey Numbaz (Scruff) Posted: June 15, 2005 at 05:50 AM (#1405486)
Those DPs really do kill him wow. BR basically has his numbers and Schang's reversed, in terms of RC/27. Interesting. This drops his offense a notch for me, but I think the PCL credit probably offsets that.

It does explain, at least partially why he doesn't do as well in WS, since RC/27 is the metric of choice for WS.

Thanks OCF.
   32. TomH Posted: June 15, 2005 at 01:54 PM (#1405718)
good pts all

Schang is the best comp to Lombardi, in that they were both effective bats, similar career PA, not great defenders. I have Ernie slightly ahead with the stick (smal timeline bonus), and slightly ahead overall. Yes, the GIDP hurt, but a Lombardi double cleared the bases, and his singles pretty much always scored a runner from 2nd, didn't they? But both of them are stuck near #20-#25 on my ballot for now. It's a tough time to become eligible and crack a ballot, ain't it??
   33. Chris Cobb Posted: June 15, 2005 at 01:56 PM (#1405722)
But a 30-WS distance from Dickey, who was at the top of the ballot is much less ranking wise than a 30-WS difference from the guy that finished 5th, no?

Sure, though Beckwith is being underrated on his numbers by the electorate because of statistical uncertainty and because of "character issues."

However, the loss of 30 career win shares also entails a drop in peak and in rate stats also. If I were to accept every assumption in the discussion here -- Lombardi has equal career playing time to Bill Dickey with PCL play, Lombardi was one bws/season worse as a hitter, Lombardi was one fws/season worse on defense, Dickey's defense is overvalued by 7%, Lombardi's lesser durability cost him an additional 2.5 win shares of peak value per season in his top seasons when compared to Dickey -- and then rank Lombardi relative to Dickey, he would end up at . . . 15 on the current ballot, which is where Biz Mackey ranks at the moment on my preliminary ballot.

That's a lot higher than he currently ranks, but it would put him squarely in the backlog, not among the leading candidates. He might still fall short of the ballot, since both Bresnahan and Schang would benefit substantially from a recalibration of catcher defense, and both were better offensive players.

So even if I were to share this interpretation of Lombardi, I would still think a high ballot placement overrates him.

I'm open to persuasion on defensive value, but I'm not convinced by the arguments so far.

In reviewing the PCL history, I think he deserves a season and a half of MLE credit for his PCL play. That will raise him in my rankings, but won't bring him past Mackey, Bresnahan, or Schang.
   34. andrew siegel Posted: June 15, 2005 at 02:08 PM (#1405744)
On the most sophisticated metrics, Lombardi wasn't Schang's equal at the bat or in the field and didn't play any longer. How can he be a serious ballot candidate for anyone who doesn't have Schang in the top 10?
   35. TomH Posted: June 15, 2005 at 02:30 PM (#1405799)
By the sophistictaed metrics BRAR and BRAA (adjusted for league strength), Lombardi leads Schang by over 80 runs.
   36. Chris Cobb Posted: June 15, 2005 at 03:09 PM (#1405888)
By the sophistictaed metrics BRAR and BRAA (adjusted for league strength)

I am about done looking at the WARP1 to WARP2 "league strength" adjustment as at all reliable. Right now, it shows Lombardi for the 1944 and 1945NY Giants going from 1.8 WARP1 to 1.9 WARP1 and 5.7 WARP1 to 5.9 WARP2. He gains value in 1944 and 1945?? Yes, apparently because, in "all-time" context used for WARP2, catcher defense was more valuable than in was in the mid-1940s, so Lombardi gets a boost in the FRAR adjustment that outweighs the drop in batting adjustment due to competition issues.

Lombardi was only able to make a career at catcher because he was playing at a time when the defensive demands on catchers were at a historic low. Should he get a bonus, then, for being projected into a defensive environment in which it is doubtful that he could have played??

This doesn't address the batting value issue Andrew and Tom were touching on, I know, but I think skepticism about WARP2-3 needs to be heightened. It's is not just a league strength adjustment, it is a projection into an "all-time" context that just does not make sense for evaluating fielding value.

Consider: Schang was catching in the teens and early twenties, when catching was highly important defensively. He was generally, an average defensive catcher but good enough to catch when the position was demanding. He loses value as a catcher when projected into this all-time context, which sets catcher as less important defensively. Lombardi was catching in the 1930s and 1940s, and he was barely good enough to catch when the position was at its least demanding. He gains value as a catcher when projected into this all-time context. Surely that's not right?

Even if it's in some way defensible, believing that WARP3 is a "league strength" will lead to a mistaken impression its meaning.
   37. Chris Cobb Posted: June 15, 2005 at 03:11 PM (#1405893)
Too much of a hurry: Here's the first paragraph again, after better proof-reading:

I am about done looking at the WARP1 to WARP2 "league strength" adjustment as at all reliable. Right now, it shows Lombardi for the 1944 and 1945 NY Giants going from 1.8 WARP1 to 1.9 WARP2 and 5.7 WARP1 to 5.9 WARP2. He gains value in 1944 and 1945?? Yes, apparently because, in the "all-time" context used for WARP2, catcher defense was more valuable than it was in the mid-1940s, so Lombardi gets a boost in the FRAR adjustment that outweighs the drop in batting adjustment due to competition issues.
   38. andrew siegel Posted: June 15, 2005 at 03:14 PM (#1405901)
A small percentage of the lead Lombardi has over Schang on BRAR and BRAA is an illusion of season-length disparities (by my calculations, Schang picks up about 300 plate appearances if you make this adjustment). However, TomH is right--the WARP folks have Lombardi as a slightly better offensive player (.294 to .291 career EQA's, between 65 and 75 batting runs). I still don't think that his slight batting lead is enough to offset Schang's better (though so-so) defense or the hidden effects of the huge baserunning difference between the two. I currently have Schang in the late 20's on my ballot. I had originally had Lombardi out of the top 50; on reconsideration he probably should rank in the low 40s.
   39. sunnyday2 Posted: June 15, 2005 at 03:14 PM (#1405902)
Chris,

Herb Brooks was famous for the saying, "Fool me once, shame on you. Fool me twice, shame on me."

This is apropos of the fact that you are only now deciding that WARP has no clothes? I haven't looked at it in 30 "years."

TPR has become passé since, well, since when? Whoda thunk that after a couple years of HoM discussions, exposing more and more weaknesses in WS and WARP, that TPR would start looking pretty good again? It is beginning to strike me again as a more reliable (understandable, malleable) metric.
   40. TomH Posted: June 15, 2005 at 03:31 PM (#1405949)
Sunny, I'll ask some WARP-related Qs to you on the 'battle of uber-stats thread'
   41. Chris Cobb Posted: June 15, 2005 at 04:15 PM (#1406146)
This is apropos of the fact that you are only now deciding that WARP has no clothes? I haven't looked at it in 30 "years."

Well, I've never based my rankings on WARP, but I take their WARP1 numbers seriously as a cross-check on win shares, especially for fielding and for good players on bad teams.

WARP2-3 I have long distrusted. However, since a portion of the electorate is using WARP3 as the primary metric, it seems important to point out serious, evident errors whenever I notice them, in hopes of diminishing the effect of WARP2-3 biases on our elections.
   42. sunnyday2 Posted: June 15, 2005 at 04:18 PM (#1406159)
>it seems important to point out serious, evident errors whenever I notice them,

Well, you've got a point there!
   43. Mark Shirk (jsch) Posted: June 19, 2005 at 09:09 PM (#1416189)
Having looked over Lombardi I must say that his candidacy falls short for two reasons despite his high OPS and OPS+ totals, especially for a catcher.

1. His GIDP really hurt him when looking at more advanced metrics like Eqa, RC and RC/27. To some degree Jim Rice will run into the same thing.

2. He didnt' play many games in-season. This keeps his single season WS and WARP totals (not to mention RC and whatever other measure you may choose to use that has a playing time element). His peak is therefore not as impresssive as it may look by OPS+. I dont' like to look at anything smaller than a season and when a catcher only plays 110 games he is doing less to push his team to a pennant than a catcher of the same ability who plays 130 games. Edd Roush suffers from this same problem as well.

Right now Lombardi is outside of my top 50, meaning he sint' a really serious candidate right now. I like him and wish he was higher, but wishing won't make it so.
   44. Brent Posted: June 22, 2005 at 04:02 AM (#1422415)
There was an interesting article by Steve Treder at hardballtimes.com on the Casey Stangel-managed Oakland Oaks of 1946-48.
   45. yest Posted: July 03, 2005 at 10:28 AM (#1446576)
every time I think a bought Lombardi I wonder what his stats would have been if he was a good runner.
   46. Mark Shirk (jsch) Posted: July 03, 2005 at 01:10 PM (#1446582)
That's a good question, could his being so slow afoot have cost him enough offense to make him a HOMer? I know that his GIDP's bring his RC and Eqa down quite a bit.

How would one answer this without watching game tape or timing him to first vs. other players?
   47. Paul Wendt Posted: July 07, 2005 at 02:42 PM (#1454894)
Chris Cobb #27
I meant that Dickey would average 5 more put-outs a year than Lombardi, which I picked as a conservative number. It could easily be 10 per year.

I agree. 5 putouts --net of those handled by 3B and 1B in lieu of C-- is conservative. And some of the credit for what is netted out should go to 3B and 1B. (Not all. Are Werber and McCormick overrated because they played with Lombardi? Doesn't matter here.)

What that's worth to the team is debatable, but I think you are underestimating the value of these plays.

Certainly Lombardi's fielding limitations in this respect don't cost his team every time, because somebody else will make the put-out later, but if you have _every_ defender on the team falling short in this way, the defensive consequences would add up fast. (This is an example of what James talks about as the "false normalization of fielding statistics" -- 27 outs are always going to be recorded.)


Joe Dimino #28 seems to concede.
I disagree.
This, Joe's point about Lombardi's missed putouts, is not an example of "false normalization" (see James, Win Shares). Lombardi's misses must be devalued relative to putouts missed by his teammates because he misses putouts on foul balls, almost always. The cost is an out vs., usually, a strike; compare an out vs. a baserunner. (Teammates field foul balls, too, that should be a baserunner at CF and P; almost always a baserunner at LF and RF; usually a baserunner at SS and 2B; frequently a baserunner at 3B and 1B.)

I don't agree with all that follows below, either, but that isn't important enough to be said. Joe has said some of it.

And you can hardly say that it doesn't matter that Dickey gets his team five outs that a weak defensive catcher doesn't provide. Turning an out into a strike may not be a huge net loss, but turning a strike into an out is a large net gain.

> James even admits that he is just guessing on
> catcher D, and that no one really has how to
> value it pegged yet.

What he's guessing on most is the _baseline_. The value of one catcher to relative to an average catcher is easier to assess.
   48. Chris Cobb Posted: July 07, 2005 at 11:38 PM (#1456615)
Some ruminations in response to Paul:

1) So the fact that 3B-men and 1B-men make a higher percentage of their plays on foul balls is part of what makes defense at these positions less important?

2) It has been my understanding that a large foul ground around the infield is one of the features that slants a park heavily towards pitchers rather than hitters. If this is true, doesn't it suggest that turning strikes into outs by catching foul balls is highly advantageous? Could comparisons of catchers catching pop-ups in foul ground in a park with much foul terriroty to catchers making putouts in foul ground in a park with little foul territory give insight into the value of catching foul pop-ups in general?

3) I recall reading in discussions of the extent to which pitchers influence outs on b.i.p. that skill in inducing pop-ups is one factor that enables some pitchers to get outs on balls in play at a higher rate than other pitchers. Could it be the case, then, that a defense needs to capitalize on what are generally easy put-out opportunities?
   49. Slivers of Maranville descends into chaos (SdeB) Posted: February 08, 2007 at 01:30 AM (#2293811)

That's a good question, could his being so slow afoot have cost him enough offense to make him a HOMer? I know that his GIDP's bring his RC and Eqa down quite a bit.

How would one answer this without watching game tape or timing him to first vs. other players?


According to James, he was so slow that opposing SS and 2B would play in the outfield. And he still hit .343.
   50. Paul Wendt Posted: February 10, 2007 at 06:08 PM (#2295264)
1) So the fact that 3B-men and 1B-men make a higher percentage of their plays on foul balls is part of what makes defense at these positions less important?

If true, yes, that is part.

2)
3)


Of course it helps the defense to turn strikes into outs and hurts to let potential outs be strikes. But it doesn't help and hurt nearly so much as does the conversion of balls hit in fair territory.
   51. Cblau Posted: June 21, 2007 at 10:01 PM (#2412481)
I was flipping through the SABR Record Book, and noticed that Lombardi is the only player on the career GIDP list to have played most of his career before WW2. As OCF wrote, that is partly because the statistic wasn't kept until 1933. But Lombardi immediately jumped into the lead, and had no trouble maintaining it. He retired in 1947 with 261; in second place was Joe Medwick with 202. It wasn't until 1970 that anyone (Hank Aaron) surpassed Lombardi's career total.
   52. Paul Wendt Posted: June 22, 2007 at 02:31 AM (#2412644)
2) . . . Could comparisons of catchers catching pop-ups in foul ground in a park with much foul territory to catchers making putouts in foul ground in a park with little foul territory give insight into the value of catching foul pop-ups in general?

I think it would be better, providing more analytical traction, to look at home and road records of large-foul-territory career catchers such as Terry Steinbach in Oakland --and the other basepath fielders make some plays in foul territory. I don't know how far back there is good data on fair/foul location of flies.

3) I recall reading in discussions of the extent to which pitchers influence outs on b.i.p. that skill in inducing pop-ups is one factor that enables some pitchers to get outs on balls in play at a higher rate than other pitchers.

That is plausible.
For ball in play data there is a bias. A foul fly is a ball in play only if caught, right?

Do the people who score fielding by "zone" distinguish fair and foul by using the foul lines as zone boundaries?

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
BDC
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Syndicate

Page rendered in 0.4250 seconds
49 querie(s) executed