Add new comment

This is a telling statement, and probably explains why any sort of debate here is pointless: "Your case seems to be "if it si not 100% proven, theobservation has no merit."
 
You're using a metric that you don't understand as a proxy for something you don't know to back up something you think you've seen visually in an extremely small sample, and what's hilarious is that you think this process has something to do with sabermetrics.  
You think I'm trying to prove or disprove something; I'm not.   All I'm saying is that you have no idea what his batspeed is based on the evidence here.  If your case is based more on your observations of Langerhans, SAY SO.   In any event, most of his XBHs have been pulled, and BOTH of his game-winning HRs have been pulled *against 'power pitchers.'*
"thanks for assuming I was an idiot and talking down to me like a 4 year old.  Awesome."
If you KNOW how they work, and you know their limitations, we wouldn't be having this discussion.   You're not measuring what you think you're measuring, so you haven't 'proven' your case albeit shy of 100% proof, you have nothing.   It's really, really important to understand when you have data and when you have noise.   This is exactly equivalent to saying that some player looks good defensively in his first 20 games or so, and then using fielding percentage to 'confirm' this.   You don't have 'proof' in that case, you've got a tiny sample of a terrible metric appended to a small visual sample.   That's.... well, it's something, but it's clearly not sabermetrics.   The entire approach is different.   Look, there's no shame in saying that you don't know, or that it's something to watch for, or whatever.   But you're clinging to this thesis (and getting defensive about it) in the face of a ton of uncertainty and variable data.  You are searching for ways to 'prove' that your eyes are right, and that's fine, but it's not sabermetrics.
So what ARE the league average splits in your b-ref metric, and how far from average is Langerhans?  You haven't addressed the Branyan counterexample either.   I still don't think it measures what you want to measure, but you could *possibly* try to show that Langerhans is further from average after you normalize BABIP or something. 
 
Yes, hittrackeronline only measures HRs, so it is not a perfect metric either.   I have no problem admitting when I don't have all the evidence.   However, it's at least measuring actual batspeed, and not a proxy of a proxy.   As we all know, the best thing to do here would be to check his splits with pitch fx data.   Then we'd have something.  
"When you have somet kind of substantive evidence that runs counter to the two proxies presented here, let me know."
I can't prove a negative, and in fact I'm not trying to prove anything.   I'm just saying that you don't have any basis for your claim.   Your claim may even be right, but neither of us know that.   When you understand this, then we may be able to have a civil discussion.