DaveFleming
C'mon Kaiser: you just compared Mickey Cochrane to Phil Rizzuto, Ed Stankey, and Dick Groat. You think I'm taking things too far?

It was just a thought experiment: if we were able to combine Mickey Cochrane's countable contributions on the field with the stuff we don't have a good feel for (all the defensive stuff a catcher does), and the stuff we can't count (his work as a manager, his calling of pitches, how effectively he ran practices, etc.), and if we combined Babe Ruth's (amazing) countable contributions to the stuff we can't count (the fact that he could be a pain in the ass), which guy comes out on top?

It's a conversation to have at the bar: there's no need to fret that the progress of Western Civilization has taken a regretful leap backwards, just because someone had the gumption to speculate on something unknown and unknowable. Breathe a little, will ya?
9:08 PM Jan 18th

KaiserD2
Well, a lot of people would argue that in one way or another, catchers do unmeasurable things that allow pitchers to get more people out, quicker. Or, conversely, to make it harder for them to get people out. The most obvious thing that could be, I think, would be to have a good, or bad, relationshio with the umpires. We now have the data that would allow us to see if there are more bad calls with certain catchers behind the plate than others. This also would take in the "framing" issue. Of course, what I'm saying here is that we might indeed be able to [i]measure[i] this additional contribution that a catcher could make. More generally, I think it would be extremely difficult to evaluate the overall contribution a catcher makes to pitching effectiveness, unless perhaps you had a team where two guys split the catching about evenly. But even a sample size of half a season, in my opinion, wouldn't be all that reliable.

I do think that the point of sabermetrics is to measure things. Now another conclusion of my book is that the role of luck in baseball is underrated. Luck in the form of missing or exceeding your Pythagorean projection over a season averages about 4 games a year--that is, the same as having, or not having, one more superstar on your team. But rather than look for mystical or emotional factors to explain those deviations, I'd rather just look at them as luck.

If we can't measure it, I wouldn't cast a vote for the Hall of Fame based on it--no. I would pose this question: can you think of any player, catcher or other, who has a great reputation for his "intangible" contributions who played mostly for losing teams? And if the answer is no, doesn't that tell you something? There are plenty of great hitters and pitchers who played mostly for losing teams (although the pitchers almost never get the recognition they deserve.) I will say this: in terms of what we can measure, the Hall of Fame standards for catchers have to be a lot lower than for anyone else to get any of them in at all. (At least if you're looking at how many extra games teams won because they had them.) I did not try, in my book, to say where that line should be. The Hall of Fame is not the main issue in the book, although I could not resist pointing out a number of people whose selection was, to say the last, dubious, and several others--much to my surprise--who have been kept out despite being overqualified.

Fire away!

David K


7:03 AM Jan 18th

KaiserD2
I am very dubious about the premise of this article, although there is one very sound and interesting conclusion embedded in the data.

When you say that Posada--or Tony Lazzeri, or Phil Rizzuto, or Bernie Williams--should be in the Hall of Fame because their teams were so successful, what you are really saying, I think, is that they belong in the Hall because their teammates were so good. Now quite a few people are in the Hall for that reason, including a great many pitchers. I have been giving a presentation this year comparing five pairs of pitchers. In each case, the worse pitcher is in the Hall, because he happened to play for excellent teams, improving his won-loss record.

Meanwhile, the outcome of any particular game has such a high random component to it that I don't think that comparing the winning percentages of catchers--the percentages of their teams in the games they played in--is meaningful either.

However, it is not surprising that the best center fielders show better winning percentages than the best catchers. That is because the best centerfielders are much more valuable players than the best catchers. Willie Mays had 16 seasons with 4 or more wins above average. That, as it turns out, is the definition of how good you have to be to be the best player on a pennant-winning team. Johnny Bench holds the record for number of >4 WAA seasons by a catcher--with 3. (This is all based on my own modifications to the baseball-reference.com calculations which will be explained in my book. They don't include position adjustments.) The reason, clearly, is that catching is so physically demanding that even great hitters at that position don't stay great very long. Jorge Posada had 1 such season, in 2003. He certainly helped the Yankees but that was the only year in which one could argue that he was a critical factor in winning the pennant. They needed better players than he to do that, and they had them.

Incidentally, regarding position adjustments--very few teams, as I said, ever won a pennant without at least one 4 WAA player (even now it's relatively rare to reach the playoffs without one), and no team whose best player was a catcher with 2-3 WAA ever won a pennant.

There is another interesting issue relating to catchers that I don't think has gotten enough attention. On defense there is very little that a catcher can do to save runs. He can, in high stolen-base eras, create significant numbers of extra outs by throwing runners out. But with respect to catching popups and fielding bunts, there aren't enough chances in the course of the season for superiority to prove out. Now sophisticated analysis may prove that some guys are better at "framing" than others and earn some strikeouts that way. But I'm not aware that that has happened yet.

When a team has a "great pitching staff," we often assume that the catcher was a great receiver. But for most of the history of baseball--the steroid era is an exception in this respect--a team with a "great pitching staff" was usually a team with solid, slightly above average pitchers that played in a good pitchers park, and had great offense.

Catching is very hard, but catchers are very rarely a critical variable in a team's failure or success. And I definitely believe, as Bill James once wrote, that any catcher who shows real promise as a great hitter should be switched to some other position.

David Kaiser​
7:14 PM Jan 17th

DaveNJnews
Excellent piece. Thanks for giving me a different way at looking at something.
12:51 PM Jan 15th

OldBackstop
re: starting pitchers and Posada. I was thinking back, and I recall that there was a dustup with AJ Burnett, and Molina was named his personal catcher for at least awhile.

So I created the formula (M - P * C / T) >= 6, where T = team games, P = games started by the pitcher, C = games started by the catcher, and M = mutual starts by the pitcher and catcher.

It shows some unusual pairings that would have affected Posada's W-L record. For instance, in 2006, on a team that won 97 games, Cory Lidle got nine starts, and had the worst winning percentage and ERA for any Yankee with that many starts. But Sal Fasano started seven of those games. Jorge started 30 games for Mike Mussina, who only had seven losses.

I have researched Yankee personal catchers back to 1927, according to the above formula, calculated pitcher, catcher and battery starts, and compiled my work into a table. I then colored it with a soft teal, backdated it to 2011, and loaded it on to some site called:

www.mikemav.com/xbl/PC_NYA.htm

Thank me later.


1:55 PM Jan 14th

tangotiger
The cleanest way is to only look based on the starters, regardless of whether he came in as a sub.

You can do starter-as-C and starter-as-DH.

Alternatively, you can use win expectancy when he entered the game, and when he exited the game.
9:13 AM Jan 14th

chuck
Dave, perhaps I was the reader you referenced here. I did look at the team’s W-L record with and without Posada. But I did go through the game logs and only counted as “with” Posada those games in which he either played the whole game or had at least 5 innings.

All the other games- his partial games where he played the smaller portion, had an overall record of 70-115. So, yes, if one were to include that in Posada’s record the stats would be deceiving in creating a worse record for him.

But as it stands, those partial games get lumped into the OTHER catchers’ records; is that fair to them? Those non-Posada + mini-Posada Yankee games from 1997-2011 had a record of 474-289 (.621). If that 70-115 record is also not counted against the other catchers, then the purely non-Posada games would have this record:
404-174 (.699).
That seems kind of incredible, but it's what the math says.

From 1997 through 2011 New York went 1465-961 (.604) overall.
Now consider these splits...
911-574 (.613) in games with Posada playing the whole game.

BUT, in some of these, Posada was not catching at all, but was the DH.
In those games, the team went 86-74 (.538).
What to do with those? For purposes of this study, I'm throwing them out; he's playing the whole game... but not catching. That improves his record to:
825-500 (.623) when he caught the entire game.

080-098 (.449) in games with Posada playing the majority portion of the game.
070-115 (.378) in games with Posada appearing in minority portion of game.
leaves...
404-174 (.699) in games without Posada at all.
12:55 AM Jan 14th

tangotiger
Dave, excellent article. I responded here:
tangotiger.com/index.php/site/article/did-yanks-play-better-when-posada-did-not-play

If you can do that with all the catchers, it would be interesting.

Note a typo: 115 losses, not 155. The win% is correct.​
11:10 PM Jan 13th

flyingfish
Thanks, Dave. Anything that makes me think more carefully about baseball is good. :) Also, I appreciate your intellectual honestly in continuing to look at the problem after you thought you had the solution and then doing more analysis.
10:36 PM Jan 13th

MarisFan61
Dave: I realized the rationale (although I don't think it's very strong, because whatever positive reputation Beltran has, I don't think it's comparable to the others, not even Snider's although one could argue it isn't far behind) but......

All I said was that listing Beltran together with those others "seems more than a bit incongruous," and I think indeed it does.
9:31 PM Jan 13th

DaveFleming
In the original draft I put Beltran right after DiMaggio, just for you, Maris. I tossed in Speaker on the edits. They all have pretty solid reputations as winners, which was the context.
8:18 PM Jan 13th

mrbryan
Snider didn't really have the benefit of the 1963 Dodgers championship, since he spent that year playing for the Mets, who really weren't going to help anyone's winning percentage.
7:20 PM Jan 13th

shthar
thanks for solving the old 'bullpen catcher W-L%' conundrum.
3:06 PM Jan 13th