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

Friday, July 11, 2003

Charlie Bennett

WS or WARP3? Whichever one you use will go a long way towards where he fits on your ballot.

Joey Numbaz (Scruff) Posted: July 11, 2003 at 03:59 PM | 38 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. Jeff M Posted: July 12, 2003 at 12:11 AM (#515514)
Bennett has never been on my ballot. I'm not his enemy and I'm not trying to talk anyone into removing him from their ballot. This is just my thinking.

1. Most of the positives I see in Bennett are his talent at the catching position. WS would give him 4 gold gloves, which isn't THAT many considering he played 15 years. I give him a little more credit than that b/c I'm not sure anyone has figured out how to measure the catching position adequately.

2. If you are looking at the WARP numbers, you'll see that except for two years, most of Bennett's WARP number comes from fielding, not hitting. That's fine, except (i) I don't know how FRAR is derived and (ii) fielding isn't more than 1/3 of the game. WARP3 thinks Bennett is better than King Kelly overall, which in my mind discredits the WARP fielding ratings.

3. He didn't catch a tremendous number of games. Maybe he would have if the seasons were longer. Who knows? Maybe he would have if it weren't for the train accident. But still, he caught fewer than 1,000 games. That's about 6-7 modern day seasons. Somebody mentioned Bob Boone the other day. Boone caught 2,200 games. Yes, catching was tougher in Bennett's era, but ....

4. I normalize stats to a season in which the average team scores 750 runs. I don't want to slip into a traditional stats analysis, but when you do that, Bennett is a .231/.309/.350 man. That's Jim Sundberg territory.

5. For those freaking out about what I said in #4, I do show Bennett as being about 20% better than the league at the plate over his career (which is much better than Sundberg). Unfortunately, we're going to see lots of guys who are 20% better at the plate, and lots of them are not HOMers.

6. Bennett's Pennants Added are very very low as our candidates go.

All in all, a very good receiver and a good guy to have on the team, but not a HOMer in my mind.

Interesting fact that I haven't seen posted about Bennett (though I may have missed it): "The Ballplayers" mentions that Tiger Stadium was Bennett Park until 1912, named because of Bennett's popularity with the Detroit fans. "The Ballplayers" doesn't have much to say about his playing ability, though.
   2. Jeff M Posted: July 12, 2003 at 01:30 AM (#515516)
I understand your points Jason, and again, I didn't write the analysis for purposes of convincing anyone one way or another.

A couple of follow-ups: I'm not excluding Bennett solely because he didn't catch many games. But there is a difference between guys who play 2,000 games at catcher and who are 20% better hitters than the league average and guys who play less than 1,000 games at catcher and are 20% better than the league. It isn't Bennett's fault, but that's the case. Over his career, Bennett caught in about 57% of his team's games. In a 162-game season, that would be about 92 games a year, or around 1,350 for his 15-year career...which isn't bad, but it doesn't blow me away; and, we don't know whether Bennett would have played for 15 years if he caught 92 games a year. We just don't know.

As for the adjustments, I don't care much for OPS, particularly in these early years. I adjust all the numbers in a manner similar to the way Bill James adjusts Willie Davis in The New Historical Baseball Abstract. Whatever individual numbers that produces in a 750 run context are then evaluated on a Runs Created basis, and then adjusted for season length. I'll have to give some thought to your comment about the errors, but I think since I'm using the RC formulas and measuring against a league average RC, that concept is accounted for (though it would not be if I was just using OPS).
   3. MattB Posted: July 12, 2003 at 02:49 AM (#515517)
As a new-found FOCB (he just sort of creeped up my ballot like that), I don't think that the fact that he only caught 1000 games instead of 2000 is really a strike against him.

I mean, in 1905 he's fourth all the all-time games caught list, I think, behind Deacon McGuire (1611, eventually), Chief Zimmer (1239 games) and Jack Clements (1073). Remember, Buck Ewing only caught 636 games.

But Bennett's career was earlier than those who caught more, so his first few seasons had fewer games in them. Comparing him to a 2000 game catcher at this point is merely hypothetical. Bennett has caught the 4th most games ever, and is better offensively and defensively than those who have caught more. That is enough to at least get him on the ballot. How high he goes depends on what you think of his defense.
   4. Chris Cobb Posted: July 12, 2003 at 03:26 AM (#515518)
On Bennett's durability in comparison to the other full-time nineteenth-century catchers:

Charlie Bennett -- 57% of team games caught, 1878-1893
   5. Chris Cobb Posted: July 12, 2003 at 04:27 AM (#515519)
Now I really will finish the post. Jeff, I'm really not sure what you're doing to get the run-environment as RC sees it for Bennett's seasons. Here's my understanding of Jason Koral's comment on errors and its consequences.

In 1883, the National League scored 5.79 runs/game. Multiplied out to 162 games, that's a 936 run context. Bennett's stats would thus, it seems, need to be reduced to put them in a 750-run context.

However, if you apply the basic RC formula [(H+BB)TB/PA] to the 1883 National League, it sees scoring at a rate of only 3.83 runs/game, a 620-run context. Basically, a third of the scoring is created by factors that the basic RC formula can't see -- errors, passed balls, baserunning. If you use James's period-specific formula, which I don't have, your discrepancy will be less, but it will still be significant for the 1880s game. If Bennett is playing in a context where _batting_ events on their own create only 3.83 runs/game, then his numbers should be significantly increased when translated into a modern 750-run context. This seems to me to be the appropriate basis for the translation.

Here's what I get from the two translations
   6. KJOK Posted: July 12, 2003 at 06:13 AM (#515520)
I'd have to disagree that the case for Bennett is just for defense, like Mazeroski or even Brooks Robinson. Comparing Bennett to Campanella:

EQA
   7. Jeff M Posted: July 12, 2003 at 01:56 PM (#515521)
Oh boy.

First, I understand the point about the errors. I think the era-specific RC formula adjusts for at least a part of this, because it weights the other events differently. Still, I may need to re-examine whether the translations work for this era. Maybe it would be better to judge the eras by earned runs. I'll work on that, and I appreciate the input (thanks for bringing it up Jason).

I'm not sure that's going to change the rankings I currently have, because if there's a translation error, it applies fairly equally to everyone we are evaluating. It would, however, affect comparisons to players in the eras where errors were lower. I'm not going to get into whether Bennett was as good a hitter as Campy, except to say I don't think he was.

BTW, I'm not excluding Bennett from the ballot based on the Sundberg comment, so maybe I shouldn't have made it.

Second, I'm not actually comparing Bennett to players who caught more. I'm saying that Bennett was a pretty good hitter but not as good a hitter as most of the other guys on my ballot. He gets a defensive bump for being a quality catcher, but since he caught fewer than 1,000 games, it is a smaller defensive bump than if he caught 2,000 games. Doesn't that make sense? If on a scale of 1 to 10 his catching ability is a 10, and he did it for 1,000 games, is that as good as catching (or even playing 2b) at the 10 level for 2,000 games? No. I'm not saying he wasn't durable. I'm just saying that I see his hitting as above-average but below-HOMer, and therefore there's got to be a lot of defense to push him on the ballot. He was an excellent defensive catcher for a fair number of games. It just doesn't push him on my ballot.

Third (and just an extension of the second point), Bennett's amazing defensive contribution was only applied in 57% of his team's games. Again, I'm not criticizing him for catching at that rate, because he obviously caught at the same rate as his contemporaries. I'm not requiring him to catch more frequently than the era standard. But it seems to me we hear the words "great defensive catcher" and we get starry-eyed. How much he contributes to a team is still related to how much he catches. He didn't really catch all that much, so I'm not going to take the "great defensive catcher" too far. I believe he was a great defensive catcher but that he probably didn't add much more to his teams defensively than an above-average defensive catcher would have added in more games.

Fourth, I don't have positional quotas. I'm comparing him to the other eligible candidates. I think he's better than McGuire, Robinson and Clements. I haven't evaluated Zimmer yet. That doesn't make him better than Harry Stovey, Hardy Richardson, Bid McPhee, et al, in my opinion. If he had the opportunity to sustain his outstanding catching for another 250+ games (and actually did so), that might make him equal to or better than those other guys, because his defensive contribution would obviously be higher. But he didn't.

Finally (and I mean finally -- I'm interested in the arguments everyone is making, but I'm on the verge of being tagged as an EOCB and I'm not actually campaigning to keep him out of the HOM) -- I think WARP overvalues the catching contribution. We don't really know how to calculate a catcher's defensive prowess based on available numbers. Bill James takes a crack at it in WS, but admits that it's a fishing expedition. WARP makes Bennett a better player than King Kelly based on those same available numbers, which most people agree don't tell us a lot about how good a catcher is. A lot of a catcher's value is intangible. And like all fielders, some of their value is related to durability -- playing well at a demanding position for a significant number of games.

Comparing another position/player, we have a pretty good handle on how to evaluate 2b fielding numbers. McPhee was an outstanding defender by all accounts. He played 2b for 2,100+ games. McPhee was a different hitter than Bennett, but they both probably were about 20% better than the average hitter. For Bennett to be ranked near McPhee, I would have to accept that 954 catching games (at a high level) is as good as 2,100+ 2b games (played at a high level). I can't do that.
   8. John (You Can Call Me Grandma) Murphy Posted: July 12, 2003 at 03:27 PM (#515522)
I'm not going to get into whether Bennett was as good a hitter as Campy, except to say I don't think he was.

Agreed. Bennett appears to be Campy because of the inferior competition the former played with. If you take into account standard deviation, the comparison between the two falls short. Campanella was definitely the better hitter.

I'm saying that Bennett was a pretty good hitter but not as good a hitter as most of the other guys on my ballot.

But shouldn't you be comparing him to his peers at the position he played? Obviously, all teams have to play their games with a first baseman, shortstop, catcher, etc. The rules don't allow for two second baseman in a game. Therefore, if a catcher is hypothetically better 50% better than the replacement level at that position, while another first baseman is 50% better than the replacement level at his position, they're basically equal in value.

My point is, while Bennett may not have been as good as some of the outfielders and first baseman in terms of hitting, he may be equal or even more impressive when comparing him to the other guys that toiled at his spot.
   9. Chris Cobb Posted: July 12, 2003 at 05:35 PM (#515523)
Jeff wrote: I don't have positional quotas. I'm comparing him to the other eligible candidates. I think he's better than McGuire, Robinson and Clements. I haven't evaluated Zimmer yet. That doesn't make him better than Harry Stovey, Hardy Richardson, Bid McPhee, et al, in my opinion.

I think that the statistical record provides room for significant differences of opinion on where Charlie Bennett should be ranked on the ballot; I hope this debate is making the case clearer that he should be somewhere on the ballot. I myself place him right now below McPhee, Stovey, and Richardson, but above Williamson, Tiernan, and Thompson.

I'd appeciate seeing folks make the case for where to position him within this group. How can we accurately assess the value of a 19th-century catcher's defense? How heavily should we value peak vs. career in rating catchers in comparison to players at other positions? What is the case for or against a "quota" vote for Bennett as the best full-career catcher from the first 40 years of professional baseball, regardless of where exactly a numerical evaluation places him?
   10. OCF Posted: July 13, 2003 at 12:30 AM (#515524)
There are references on this thread to Bennett catching 57% of games. However, there is a peak that's different. Here's what I said on last year's thread. Recall that, although Jack Clements isn't going into the HoM, he was respectable enough to draw discussion.

"
   11. User unknown in local recipient table (Craig B) Posted: July 13, 2003 at 03:56 AM (#515525)
For Bennett to be ranked near McPhee, I would have to accept that 954 catching games (at a high level) is as good as 2,100+ 2b games (played at a high level). I can't do that.

Or, you could say that Bennett was limited by the nature of his position, and move him up on that basis. It would have been impossible to catch 2100 games. Deacon White caught 1600, and he had a career of Pete Rose proportions.

You don't *have* to do this, of course. But it's a way of giving catchers a fair shake if you want to.

But shouldn't you be comparing him to his peers at the position he played? Obviously, all teams have to play their games with a first baseman, shortstop, catcher, etc.

Of course they do, and clearly defensive value should be position-calibrated as everyone agrees. But nobody bats as a shortstop, a left fielder, or a first baseman. They are just hitters.
   12. User unknown in local recipient table (Craig B) Posted: July 13, 2003 at 03:59 AM (#515526)
Sorry, that should be Deacon McGuire, obviously.
   13. John (You Can Call Me Grandma) Murphy Posted: July 13, 2003 at 06:41 AM (#515527)
Of course they do, and clearly defensive value should be position-calibrated as everyone agrees. But nobody bats as a shortstop, a left fielder, or a first baseman. They are just hitters.

This is ostensibly true, but we know intuitively that a catcher who hits thirty home runs has more value than a first baseman who accomplishes the feat.

I'm not suggesting that we should do this, BTW. What we should do is compare each player's offensive and defensive contributions with everyone else who played at the same position for each season. If you do this, you should find that you have voted for a proportionate number of players for each position over the next hundred "years."
   14. Howie Menckel Posted: July 13, 2003 at 01:56 PM (#515528)
Without having a 'quota' per se, I would TEND to have at least one player per position and 2-3 pitchers. Occasionally, I won't, if the talent just isn't there. And for a position like 1B, sometimes we rightfully slam a bunch in the HOM, leaving a dearth of 'also-rans' at the spot.
   15. User unknown in local recipient table (Craig B) Posted: July 14, 2003 at 05:45 PM (#515530)
we know intuitively that a catcher who hits thirty home runs has more value than a first baseman who accomplishes the feat.

There's nothing intuitive about it. We know it from being told it time and time again.

It's not true.

The catcher is more valuable than the first baseman, because he's catcher and has (almost certainly) more defensive value.

The catcher is more "valuable" to his GM, because he's so much rarer a bird than the first baseman and he'll command top dollar.

But if, like me, you consider value to lie in winning ballgames, then the offensive performances (purely as offensive performances) of a catcher and a first baseman with absolutely identical offensive stats, are absolutely equal.
   16. John (You Can Call Me Grandma) Murphy Posted: July 14, 2003 at 06:33 PM (#515531)
Craig B:

We're going off on a tangent here. My original point was to compare the sum total of each player to his peers at the position he played for. That's my analytical approach. Therefore, no position is ignored at the expense of another.
   17. John (You Can Call Me Grandma) Murphy Posted: July 14, 2003 at 07:47 PM (#515533)
Joe, you summed it up pretty well for me.
   18. KJOK Posted: July 14, 2003 at 08:14 PM (#515534)
John Murphy said"

"I'm not going to get into whether Bennett was as good a hitter as Campy, except to say I don't think he was.

Agreed. Bennett appears to be Campy because of the inferior competition the former played with. If you take into account standard deviation, the comparison between the two falls short. Campanella was definitely the better hitter."

John, I don't see where standard deviation matters in this case. Bennett was just as valuable offensively in his time and league as Campy was in his time. Why would the standard deviation really matter unless you were actually trying to decide to elect EITHER Bennett or Campy, which we're not. We're trying to elect Bennett or his peers that played in the same era!
   19. Jeff M Posted: July 15, 2003 at 12:20 AM (#515535)
Joe:

First, I'm used to getting harsh treatment. This isn't a group that minces words. Sometimes it's constructive; sometimes it isn't. Anyway, I've corrected the translations using earned runs, and Bennett improves, but so does everyone else b/c they are all playing in roughly the same environment. This correction will help for comparison with later players, but it doesn't help him right now.

Second, I NEVER said Bennett was a below average hitter. My post, in fact, says he was 20% better than the league. The Sundberg comparison was tongue in cheek. In the original post, I immediately negated the Sundberg comment by noting that Bennett was a far better hitter. I meant the comparison lightly, but everyone has latched on to it as if it is the lynchpin of why he isn't on my ballot.

Third, I compare everything to league averages. Bennett hit 20% better than the league (which isn't all that great as HOMer standards go) and was an excellent catcher for a modest number of games. He's just barely off my ballot. A little more hitting or some more defensive contribution (in this case more games caught) would better allow him to compete with some of the offensive powerhouses on the ballot. How many runs did he save as a catcher? No one knows.

Theoretically, for him to make my ballot, his runs saved as a catcher would have to make up the difference between his +20% hitting performance and the >+40% hitting performances of the other guys on the ballot. Maybe he belongs, but I haven't heard anyone try to address how defense propels him ahead of guys who quantifiably accounted for a lot more runs. You know why? Because a catcher's defensive ability can't be quantified based on the numbers we have. What I hear instead is that positions should be equally represented and that he is the best catcher of the current choices. I agree he is the best catcher of the current choices. Why does that make him better than the other guys on the ballot right now? BECAUSE he's a catcher? There has to be a better argument than that.

I watched an Atlanta game yesterday with this issue in mind. How many OPPORTUNITIES did Javy Lopez have to save runs? Virtually none, other than the way he called the game (to the extent he has any input, which I question). Yes, it's a different game now. Yes, it's just one game. No, Javy is not as talented on defense as Charlie. But watch games with this concept in mind. If you take away the intangibles of catching, why is it such an important defensive position? The physical act of catching a pitch is not THAT hard. It's only noticeable when someone is really bad. Most fielding of bunts is relatively routine. There are very few close plays at the plate. That leaves us primarily with the ability to throw out runners, which is so tied to how well or poorly a pitcher holds the runner on, that it is hard to know what to attribute to the catcher.

So again, what are you guys seeing in Bennett's defensive STATS (since we don't know much about the intangibles) that makes you so sure that a +20% hitter is equivalent to a defense-neutral +40% hitter, keeping in mind that offense dominates the game?
   20. Chris Cobb Posted: July 15, 2003 at 02:11 AM (#515536)
Jeff, the view I and others have of Bennett's defense is not just based on highly processed stats whose origins are unclear. Bennett's defense was discussed pretty extensively in comparison to Buck Ewing's prior to the 1902 ballot -- I think some of that info was reposted on the catchers' positional threads, which have recently disappeared into the archive.

Here's the key data, assembled originally by jimd:

<i>Following are their catcher's defensive stats (from their BP player cards).

GC. AdjGC PO... A... E.. DP. PB.
   21. John (You Can Call Me Grandma) Murphy Posted: July 15, 2003 at 07:08 AM (#515537)
John Murphy said"

"I'm not going to get into whether Bennett was as good a hitter as Campy, except to say I don't think he was.


I'm here to let you know that John Murphy didn't say that, though he did agree with it. :-)

John, I don't see where standard deviation matters in this case. Bennett was just as valuable offensively in his time and league as Campy was in his time. Why would the standard deviation really matter unless you were actually trying to decide to elect EITHER Bennett or Campy, which we're not. We're trying to elect Bennett or his peers that played in the same era!

I must project myself wrong in my posts, because I tend to get too many exclamation points thrown at me. :-)

Seriously, I agree with you entirely. I'm on record for saying the same thing many times before. For right now, he's Campy. If we were rating the greatest 100 catchers, he wouldn't be.
   22. John (You Can Call Me Grandma) Murphy Posted: July 15, 2003 at 05:18 PM (#515538)
So again, what are you guys seeing in Bennett's defensive STATS (since we don't know much about the intangibles) that makes you so sure that a +20% hitter is equivalent to a defense-neutral +40% hitter, keeping in mind that offense dominates the game?

Because you have to play the game with catcher. If you can find a way to play it with two first baseman instead, please let us know. :-D

Seriously, let's do a hypothetical. If a first baseman is just average at his position (average equaling a 1400 OPS!), while a catcher is 100% better than average (average equaling a 400 OPS), how is the first baseman helping his team more. Yeah, the first baseman is creating over 300 runs a year, but the opposing team's is creating over 400. The catcher might only be creating a 100 runs a year, but that's a huge difference from everyone else at the position. The team here will be helped more by the catcher, not the first baseman.
   23. karlmagnus Posted: July 15, 2003 at 05:44 PM (#515540)
On this argument, the most valuable catcher is the one who manages to stagger through the most games as catcher. Clements, not Bennett.
   24. John (You Can Call Me Grandma) Murphy Posted: July 15, 2003 at 06:07 PM (#515541)
On this argument, the most valuable catcher is the one who manages to stagger through the most games as catcher. Clements, not Bennett.

I'm one of the few that have place Clements on a ballot, so I'm not unfriendly to him. However, Bennett was better than Jack when you compare them to their respective eras.

BTW, I think Chief Zimmer falls somewhere between those two.
   25. Cblau Posted: April 23, 2006 at 02:45 AM (#1983024)
I was doing some research on wild pitches and passed balls, and Bennett's numbers caught my eye.
His team had the fewest passed balls 10 of his 15 seasons. If you compare his rate of passed balls per game to the league average, you find he saved his teams 394 passed balls over his career. Of course, that's assuming his games were all complete games, which they weren't. His teams were 357 passed balls below average in total.


Some mind-boggling numbers- In 1878, Milwauee's 2 regular catchers (Bennett and Billy Holbert) had 42 passed balls in 56 games. Their other catchers had 57 passed balls in 14 games.

In 1887, Detroit's catchers were charged with 44 passed balls. Boston was second lowest with 90.
   26. Paul Wendt Posted: April 23, 2006 at 03:38 PM (#1983526)
No doubt there were great differences in catcher abilities to field pitches.
If you look at the Cincinnati Red Stockings tours in 1869 and 1870, playing more amateur than pro teams, I think you will find several games with something like 1 passed ball by Allison and 10 for the opposing catcher.

Something to look at:
When Bennett and Holbert were young, pro catchers were only beginning to play under the batter. I guess there was significant variation among them with youngsters playing under the batter more. If Bennett did play under the catcher more frequently than others (especially, given this data, his teammates with other primary positions), then his low PB rate is more remarkable. I'm sure that catcher positioning influenced abilities to throw out baserunners, to field pitches (passed balls), and to field flies, complexly.

Something to keep in mind:
PB data is probably incomplete. More on that if I find and recall.
   27. Paul Wendt Posted: February 24, 2008 at 08:33 PM (#2698702)
Recently I looked at playing time for Charlie Bennett and others in his league; that is, at full season equivalent games played at catcher. Here is something interesting.

Catcher tandem for eight seasons, Charlie Bennett and Charlie Ganzel
shares of team fielding games at catcher

Bennett Ganzel other
1886 D
.548 .357 Decker .111
1887 D
.354 .402 Briody .260
1888 D .545 .209 Wells .119
Sutcliffe .104
1889 B
.617 .293 Kelly .173
1890 B .634 .164 Hardie .187
1891 B
.536 .421 Kelly .079
1892 B
.230 .336 Kelly .474 (number one catcher)
1893 B.458 .305 Merritt .282 


* pennant winner
+ close but no cigar
   28. Paul Wendt Posted: February 24, 2008 at 08:37 PM (#2698706)
Catcher tandem for eight seasons, Charlie Bennett and Charlie Ganzel
shares of team fielding games at catcher

Bennett Ganzel other
1886 De+ .548 .357 - Decker .111
1887 De* .354 .402 - Briody .260
1888 Det .545 .209 - Wells .119, Sutcliffe .104
1889 Bo+ .617 .293 - Kelly .173
1890 Bos .634 .164 - Hardie .187
1891 Bo* .536 .421 - Kelly .079
1892 Bo* .230 .336 - Kelly .474 (number one catcher)
1893 Bo* .458 .305 - Merritt .282

* pennant winner
+ close but no cigar
   29. Paul Wendt Posted: February 24, 2008 at 08:56 PM (#2698733)
Rather than format anything more I'll summarize. Among all major league catchers here is Bennett's rank by FSE games at catcher.
(1878) 6 -- 7 3
(1882) 5 4 2 5 4 21 5 2 (usually 16 teams)
(1890) 9 14 22 9

In his first six seasons with Detroit, his FSE games (catcher games as share of team games) dropped
.833 .756 .713 .702 ; .574 .548 (1881-86).
But his rank among all mlb catchers decreased only once, with the big 1885 drop in share(;). As more games were played all catchers worked smaller shares of team games, generally dropping faster than Bennett's own share. Bennett's rank:
3rd (one league), 5, 4, 2 (three leagues), 5, 4

Bennett missed part of the 1887 championship season in Detroit (FSE .354, rank 21st in two leagues). But he worked behind the plate in 10 of 15 world series games, Ganzel 7, Sutcliffe 1.
   30. Paul Wendt Posted: February 24, 2008 at 09:05 PM (#2698741)
In 1892 Mike Kelly was the number one catcher, Bennett number three. They were both long in the tooth. In the championship series with Cleveland the three catchers worked two games each. Bennett scored 2 runs (2 for 7), Ganzel 1 (4 for 8), Kelly none (oh for 8).
   31. Paul Wendt Posted: May 11, 2008 at 11:40 PM (#2777048)
Joe Dimino in today's ballot covering HOM Catchers:
15. Charlie Bennett - Serious mashing from 1881-1888. His career OPS+ sat at 135 after his age 33 season. He was purchased for an estimated $30K after the 1888 season, so his value in his time was pretty much unquestionably that of a superstar.

Joe,
Do you know a source for this? It's a surprising number because Kelly and Clarkson were purchased for $10000 each after the 1886 and 1887 seasons, I have read time and again. Here is a new one from John Clarkson at Wikipedia.
>>
On April 3, 1888, the White Stockings sold Clarkson to the Boston Beaneaters for $10,000 -- a huge sum at the time. Clarkson followed teammate King Kelly, who had been sold from Chicago to Boston the previous year. Boston had paid $10,000 apiece for Kelly and Clarkson, and they became known as the “$20,000 Battery.”
<<

This on Clarkson and Charlie Bennett is from the same article.
>>
After the 1893 season, Clarkson went on a hunting trip with his close friend Charlie Bennett, who had been his catcher from 1888-1890. Bennett got off the train in Wellsville, Kansas and when he tried to reboard, Bennett slipped and fell under the train’s wheels. Bennett lost both of his legs in the accident. Clarkson witnessed the incident, and it was said to have severely affected his already unstable nature.
<<

Charlie Bennett at wikipedia does not mention Boston's acquisition. It does say this.
>>
After the 1893 season, Bennett went hunting with pitcher John Clarkson. Bennett got off the train in Wellsville, Kansas to speak to an acquaintance. When he tried to reboard, Bennett slipped and fell under the train's wheels. Bennett lost both legs in the accident. He was fitted with artificial limbs but his baseball career was over.

After his injury, Bennett moved to Detroit, where he operated a cigar store.
<<

So Bennett operated a cigar store in Detroit, Clarkson in Bay City.

>>
Detroit fans held a day in his honor, and he was given a wheelbarrow full of silver dollars. When a new ballpark was opened in Detroit in 1896, it was named Bennett Park in his honor. Bennett caught the first pitch at Bennett Park in 1896. It became a Detroit tradition for Bennett to catch the first pitch in Detroit, an honor that Bennett continued for every home opener through 1926.
<<

Bennett must have been the catcher in about half of Clarkson's Boston games. This may be one of them or it may be fiction. (from
>>
A famous story survives from Clarkson's Boston years, an example of disdain or comedy, depending on one's view. Late one game, in a gathering darkness, Clarkson pitched a lemon to the plate, called a strike by umpire Jack Kerins. When shown, by the catcher, that he had declared a citrus a strike, Kerins called the game.
<<
(referenced by wikipedia. "The Top 100 Cubs of All Time - #20 John Clarkson" at bleedcubbieblue.com)
   32. Cblau Posted: May 12, 2008 at 12:41 AM (#2777129)
He perhaps took it from Baseball-Reference.com. It's an error. Bennett was sold together with Deacon White, Charlie Ganzel, Hardy Richardson, and Dan Brouthers for an estimated $30,000, according to David Ball. White refused to report and was sold to Pittsburgh. (Retrosheet's Web site has it right.)
   33. bjhanke Posted: May 15, 2008 at 11:12 PM (#2782647)
You can get a start looking at Charlie Bennett's defense if you look up the stats in the old Spalding guides. They printed catcher defensive stats every year. I pulled a couple out at random - they don't vary much from year to year in terms of who does what. For the 1881 season (the 82 guide), Bennett leads the league in fielding percentage, which was probably considered the primary stat, since it's the one they order the catcher defensive numbers by. He leads it by 53 points. In putouts per game, he's at 6.23. There is one other guy over 6, but he played only 34 games. There are a few fives, but most of the league is in the 4s. Basically Bennett laps the field. In assists per game, Bennett has a pedestrian 1.27. This was not a good year for his throwing arm. The league leader is Buck Ewing, in his first full season, at 2.06. Pop Snyder is second at 1.81. In other words, Ewing laps the field in this stat. In passed balls per game, Bennett is in at .57, leading the league. Holbert has .68, and then Ewing and a couple of others are about even, a little over .7. So Bennett leads the league in three out of four categories, lapping the field in all three.

In 1886 (the 87 guide), looking only at the NL (the guide has separate stats for the AA), Bennett leads fielding percentage by 48 points over Clements, who leads the #4 guy (Ewing) by only 11 points. Bennett laps the field again. In putouts per game, he is tied for third, at 6.34, with Deacon McGuire. Clements leads with 6.63, and Ganzell is second at 6.52. In assists per game, Bennett is weak, at 1.25. The league leader is Kelly, at 2.04. Then there's Graves of St. Louis, who finished last in fielding percentage, but did have 1.95 assists per game. Then it's back to Buck Ewing, at 1.90. In passed balls per game, Bennett has .37. The next best guy is Briody, at .55. Then Clements at .63 and Ewing at .64. So Bennett laps the field in only two categories this time.

None of this, of course, is adjusted for anything. It's just the raw stats from the Spalding guides. Still, if you want to know why the people of the time thought that Charlie Bennett was a great defensive catcher, and that Buck Ewing was also excellent and had the cannon arm of the league, this is the evidence they were working with. Adjustments are up to us moderns. Clements also keeps popping up at #2 or #4 or #6 in a lot of categories. Snyder is another who looks good.
   34. DanG Posted: May 16, 2008 at 01:33 PM (#2783001)
Post #20 by Chris Cobb, from the Internet Archive:

Posted 10:11 p.m., July 14, 2003 (#23) - Chris Cobb
Jeff, the view I and others have of Bennett's defense is not just based on highly processed stats whose origins are unclear. Bennett's defense was discussed pretty extensively in comparison to Buck Ewing's prior to the 1902 ballot -- I think some of that info was reposted on the catchers' positional threads, which have recently disappeared into the archive.

Here's the key data, assembled originally by jimd:

Following are their catcher's defensive stats (from their BP player cards).

GC. AdjGC PO... A... E.. DP. PB.
636 614.8 3301 1017 322 078 360 Ewing
957 925.4 4969 1531 485 117 542 Ewing (normalized playing time)
954 925.4 5123 1048 379 114 352 Bennett

Ewing has 483 more Assists, involved in 3 more DP.
The cost is 106 more Errors and 190 more Passed Balls.

Bennett has 154 more PO. From the team data, I estimate that Ewing had 710 non-K putouts (1069 normalized). Bennett had an estimated 1674 non-K putouts. So the actual margin is more like 600 more PO, due to the strike-outs recorded by Ewing's pitching staffs (lot of Keefe and Welch there).

From the assists, it's obvious that Ewing has the arm, a gun resulting in 50% more assists than Bennett, which also motivated the tryouts at 3B. (This would have kept his bat in the lineup if he could handle it, and liked it. His 1882 move looks fine, but he's back at C in 83 for whatever the reason. The 1887 move is less successful; he looks average out there.)

However, Bennett seems to make up the difference in other areas; he was more sure-handed (errors and passed balls), and the putouts indicate that he was either more mobile (pops), quick-handed (back then, foul tips were outs when held onto), maybe a better game-caller (perhaps getting more strike-outs per game then the other catchers on his team), something else I'm overlooking, or some combination of the four.

You may say this is only comparative data among catchers, that it doesn't show why catchers are more valuable, but it shows quite a bit of that. Think about how many important plays these guys are making per game, and think about how hard it is to make those plays when you've been catching a hard-throwing pitcher from 50 feet _bare-handed_ for 150 pitches. The importance of nineteenth-century catchers is much more tangible than that of catchers in the 2003 game.
   35. DanG Posted: May 16, 2008 at 01:41 PM (#2783006)
Between posts 1 and 2 there was originally this posting:

Posted 8:51 p.m., July 11, 2003 (#2) - Jason Koral
I dont know how your doing the normalization. The 1880s were a low offensive environment; to the extent that lots of runs scored, its due to the enormous amount of errors. Thus, you cant properly normalize OPS on the basis of runs scored numbers.

Adjusted league OPS over Bennett's career is .667. Compare that to say the 1985 AL (lower than a 750 run environment) where league OPS was .732. You have to normalize Charlie UP, not down to get proper comparisons (assuming no time line adjustment). Normalized to the OBP/SLUG environment of 1985 American League, Charlie is very respectable .360/.440. Compare that to an AL catcher of the time - say Lance Parrish who was .313/.440. CB looks very good indeed

Its true that many non-HoM players are 20% than overall league offense over their careers, but how many were catchers for any significant time, much less excellent defensive catchers? None. Catchers above that offensive level include Berra, Bench, Cochrane, Dickey, Hartnett. They are not a lot over: mostly in the 22-28% range. Ted Simmons is even (but obviously not defensively equal). Those below include Fisk and Carter. You see the pattern: the very greatest catchers in history are above 20%; everyone else, including some truly excellent players, is below. If you have defense and staying power, 20% over is good enough for a catcher.

As for playing time, Bennett's 950 games caught is fully equal to a modern catching 2000+ games. It was simply impossible for a catcher of that era to catch that many games; the practical limit was around 1000. Bresnahan caught under 1000 despite an early start to his career. Chief Zimmer, who actually was the Bob Boone of the late 19th century, managed to exceed 1200. Wilbert Robinson - the Sundburg of that era - managed to get past 1300. But Bennett was a good deal older then those men . . . by the standards of his time, he was very durable.

Its legit to decide as an analytical matter not to make an adjustment for this fact; to judge Bennett's playing time by the standards of our time rather than his. There is a consquence to this though:it means the first real catcher to make the HoM will be either Cochrane (maybe Schang). Should the first 50 or so years of professional ball be bereft of a HoM player who played most his games at catcher? that doesnt seem right to me.
   36. DanG Posted: May 16, 2008 at 01:43 PM (#2783009)
Here's Chris Cobb's posts in #3 and 4 in their original form:

Posted 11:26 p.m., July 11, 2003 (#5) - Chris Cobb
On Bennett's durability in comparison to the other full-time nineteenth-century catchers:

Charlie Bennett -- 57% of team games caught, 1878-1893
Jack Clements -- 57% of team games caught, 1885-1898
Wilbert Robinson -- 57% of team games caught, 1886-1902
Chief Zimmer -- 57% of team games caught, 1888-1903
Deacon McGuire -- 60% of team games caught, 1890-1906

Bennett is as durable as any of his contemporaries who managed long careers as catchers.

On translating Bennettt's statistics into a 750-run environment:

Posted 12:27 a.m., July 12, 2003 (#6) - Chris Cobb
Now I really will finish the post. Jeff, I'm really not sure what you're doing to get the run-environment as RC sees it for Bennett's seasons. Here's my understanding of Jason Koral's comment on errors and its consequences.

In 1883, the National League scored 5.79 runs/game. Multiplied out to 162 games, that's a 936 run context. Bennett's stats would thus, it seems, need to be reduced to put them in a 750-run context.

However, if you apply the basic RC formula [(H+BB)TB/PA] to the 1883 National League, it sees scoring at a rate of only 3.83 runs/game, a 620-run context. Basically, a third of the scoring is created by factors that the basic RC formula can't see -- errors, passed balls, baserunning. If you use James's period-specific formula, which I don't have, your discrepancy will be less, but it will still be significant for the 1880s game. If Bennett is playing in a context where _batting_ events on their own create only 3.83 runs/game, then his numbers should be significantly increased when translated into a modern 750-run context. This seems to me to be the appropriate basis for the translation.

Here's what I get from the two translations
(also park-adjusted)

Bennett 1883 -- .305 BA, .350 OBP, .474 SLG
936 to 750 -- .277 BA, .321 OBP, .432 SLG
620 to 750 -- .328 BA, .375 OBP, .511 SLG

The first translation makes a little better than Jim Sundberg was in 1980, one of his better years, in a 730-run environment
Sundberg 1980 -- .273 BA, .353 OBP, .384 SLG

The second translation makes him into -- someone else.

Bennett's RC for 1883 are 61.62. Translated from 620 to 750 run environments, that becomes 76.38. Translating that from a 100 game season to 154, that becomes 117.62 RC in 142 games caught.

Roy Campanella in 1953, a 770-run environment:
.312 BA, .395 OBP, .611 SLG
123 RC in 144 Games

1883 is probably Bennett's best year, though he had a couple nearly as good; 1953 is probably Campy's best year in the NL, though he a couple of other great ones. Worked out this way, Bennett's offensive game and Campy's look pretty close in value.

Having done all this in order to see what the translation would produce, I now look back to see their OPS+ for each of the seasons in question:

Bennett, 1883 -- 152
Campanella, 1953 -- 155
(Sundberg, 1980 -- 105)

Certainly there are problems with OPS+ in the nineteenth century, and this one example may not be representative, but it does suggest that the statistic is still useful as a quick measure of value, since the long way around the barn tends to lead to about the same place :-) .
   37. DanG Posted: May 16, 2008 at 01:45 PM (#2783011)
Here's Chris Cobb's posts in #3 and 4


I meant to say #4 and 5
   38. Paul Wendt Posted: May 16, 2008 at 03:58 PM (#2783161)
35. DanG Posted: May 16, 2008 at 09:41 AM (#2783006)
Between posts 1 and 2 there was originally this posting:

Posted 8:51 p.m., July 11, 2003 (#2) - Jason Koral

36. DanG Posted: May 16, 2008 at 09:43 AM (#2783009)
Here's Chris Cobb's posts in #3 and 4 in their original form:

Posted 11:26 p.m., July 11, 2003 (#5) - Chris Cobb


Jason Koral and Chris Cobb 2003,
Good work!

Dan Greenia 2008,
Great work!

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
Jim Wisinski
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Syndicate

Page rendered in 0.7652 seconds
49 querie(s) executed