Here is the study's author's comments from another forum on some of the feedback he had received. The italics are the quotes he is responding to.
Elan Fudd wrote:Hey Everyone,
I apologize for not having an available version of my study for people to read. I wasn't prepared for this level of interest. I have a 80% finished write-up that I did this past summer immediately after finishing the research. I am going to work on finishing that write-up and putting it up on the web (after I make a web-site) I hope to have that for you guys sometime next week. Be warned though, my write up is on the long side and gets very technical at points, so unless you're pretty hard-core you may not care to read it. I hope to make a shorter version sometime this summmer, but no promises on that one. In the meantime I'll do some quick responses to some of the postings so far.
"Color me skeptical until I see the data, but Penn student Elan Fuld claims that Frank Duffy, Eddie Murray and Luis Gomez are the greatest clutch hitters of all time."
My study doesn't say that they are the "greatest clutch hitters" they just happen to be some of the ones that I have the most evidence that they posess SOME clutch-hitting ability. (An interesting side-note is that one of the other ones that I have the most evidence for is Bill Buckner. Go figure. At least I'm not saying he is a clutch fielder, right?)
"Oh wait - I think I can guess: in clutch circumstances Gomez's OPS+ rose from its normal mark of 40 to a stratospheric . . . 70 or 80.
I wonder who was the worst clutch hitter of the bunch?"
Actually, what you said about Gomez hits on something important. Being a clutch hitter doesn't require being good in important situations (at least according to the way I looked at it), it requires that you exhibit a pattern of doing better as the situation gets more important. So if Barry Bonds was a choke hitter you might still want him up in that key spot, because even performing somewhat worse than usual still makes him preferable to anyone else on the Giants.
If I were in the Major leagues and struck out everytime in unimportant situations, but as situations got important I somehow managed to hit a bloop single or draw a walk 1 time in 100, that would make me a clutch hitter. You still wouldn't want me on your team. Elan in the clutch, may be better than Elan most of the time, but nonetheless worse than your average single-A utility infielder.
As far as who was the worst clutch hitter (I call them choke hitters) the number of choke hitters who appeared are about the number we would expect to see by chance if there were indeed no true clutch or choke hitters in MLB. So out of the ones who appeared choke, most of them probably were not. That said, it is interesting that the hitters who appeared decisively choke didn't have very illustrious careers. You may not have heard of them. The two who stood out the most clearly as choke were: Jeff Stone and Joe Strain.
"what would eventually be nice is to see a study with play by play, using the expected runs scored matrix on each and every at bat and keep that as a players 'runs produced' total.
other more complicated way to measure 'clutch' would be to use that players on 'personal' run expected matrix and then identify all 'clutch' at bats and see if he is better or worse than what would be expected in those situation.
of course there is a lot of data that would be useful if we could just look at play by play data. (does retrosheet have pbp for most games?) "
I did actually use play-by-play retrosheet data, and I did use probabilities to construct a measure of the play-by-play performance, however, 'runs produced' was not the approach I took. When I post my paper you can see the gory details, for now I have to limit my responses somewhat due to time constraints.
And then we'll watch the feeding frenzy and he will hate this site forever. "
Constructive criticism and debate are always welcome, I won't hate this site for reading my work critically, just no personal attacks please.
"What, no Jeter? This so-called study is obviously rigged!!!! "
My data was retrosheet 1974-1992 so Jeter wasn't in the sample. Retrosheet only goes up to 1992 and before 1974 it is missing games. Although as a Sox fan I really don't like Jeter. If I was doing a study on recent data and Jeter came out clutch, I'd say so.
"the player would start to think this is very important and start doing something that makes him hit better.
Is this a good thing? Wouldn't you want a guy to do his best all the time? "
Maybe its not a good thing. Or maybe the guy has a highest peak level of concentration/awareness that exceeds his highest sustatainable level. I don't know. I don't examine potential causes of clutch hitting. Just whether it exists and who are clutch (or choke) hitters.
OK That's all for now folks. Thank you all for showing so much interest at my work, even the skepticism is appreciated. As an ardent Sox fan, skepticism was my first reaction when Bill Buckner's name came up near the top of the list of clutch hitters. (Upon seeing that, I triple checked my work and it wasn't a mistake)
Hope you all have a good weekend!