data.3news-bydate.train.rec.sport.baseball.105034 Maven / Gradle / Ivy
From: [email protected] (Rudy Wade)
Subject: Re: Bases loaded walk gives Reds win in 12
Organization: Home of the Brave
Lines: 113
In article <[email protected]> [email protected] (Edward [Ted] Fischer) writes:
>In article [email protected] (Mark Singer) writes:
quoted for future reference...
>shown to be consistent in the past. Overall performance, L/R splits,
>even matching hitting/pitching styles. All of these will give me some
>advantage if used properly.
>was predicting the future, the next AB. He was predicting that
>Sabo was more likely to get a hit than Samuel.
He believed one of two things: 1) Sabo was more likely than Samuel
to get a hit OR 2) that more good would be done in the long run
by leaving Sabo in regardless of the expectation of the actual outcome.
In fact you don't know what Perez had in mind when he left Sabo in.
>By supporting the swap, you are predicting the opposite.
In fact, I don't think anyone is claiming that they can predict
the future, or any particular future event. But we can believe
that certain trends are due to a cause (whether or not we have
identified the cause) and therefore will continue.
>>And I am not dismissing your work. I'm stating my opinion.
>>You are saying that your work renders any opinion to the contrary
>>invalid, so that the retention of that opinion is some kind of insult
>>to your work.
>Well, yes. You are aware of its existance. You claim to be incapable
>of understanding it (though I suspect you are simply unwilling). Yet
>you rather forcefully state the opposite. You don't seem to think the
>work is worth reading (yet you obviously feel the topic is important).
>I'd say this is insulting.
I think seeing insults in other people's opinions is kind of silly.
After all, Mark didn't call you a total idiot, or call your work
stupid, he simply stated that it didn't change his opinion on the
subject.
>>I did not say that it is a consistent skill.
>> I have said that it is an indicator of performance under a
>> certain set of circumstances.
>RIGHT! This is the beef. It has not proven to be an indicator of
>future performance under *any* circumstances. At least none that
>we've been able to come up with. If you know of some where it *is* an
>indicator of future performance, please let us in on your secret.
If player A hits better on Tuesdays and always has, and Mark believes
that it might be an indication that he will hit on Tuesday better next
season as well, would you respond the same way? At some point you
might admit that all variable might not be known to you (who knows
what this guy does every Monday night? Maybe he sees his sports
shrink on Tuesday mornings, or has his Vitamin B shots Monday nights?)
But this is something that's true of one guy only. It doesn't
mean that there will be a meaningful correlation for the entire
league by days of the week, nor that there should be. But it doesn't
mean we can't make predictions based on that for that particular
player.
>>I believe that by
>>season's end that Chris Sabo's batting average in clutch situations
>>will be significantly lower than his batting average in non-clutch
>>situations. I can't prove that it will happen, so I guess we'll
>>just have to wait and see.
>
>Is this simply a prediction for Chris Sabo for this year? Or is this
>a prediction for *all* batters who have, over the past few years, hit
>(xxx amount?) worse in the clutch than overall?
I only see a prediction for one player here. I don't see anything
about ALL batters.
>>Here's another one for you. In 1989 - 1991 Joe Carter's batting
>>average in clutch situations was significantly below his batting
>>average in non-clutch situations each year. I presume you think
>>this is random.
>I'm not going to get into case analysis. Sure, you can find somebody
>who hit poorly from '89-'91 and then hit poorly in '92 as well. You
>can also find those who hit poorly from '89-'91 and then hit *well* in
>'92.
That's exactly what Mark is trying to do though. Find hitters
that have these correlation and ask whether we can make predictions
for these hitters based on their past performance.
>>major league ballplayers. It just makes me think that in 1993
>>Joe Carter's batting average in the clutch is not going to be
>>as high as his ba in non-clutch situations.
>Gambler's fallacy. Unless there is reason to expect consistency, a
Huh? What does gambler's fallacy have to do with anything?
Whether you can know the reason for correlation or not, you can't
deny that it has existent, you can only make an argument that
you don't think it is likely to continue to exist because you
can't see a reason for it to exist.
>run proves nothing. Can you give us a reason to expect clutch BA to
>correlate from one year to the next? I've seen a detailed study of
Someone posted recently on why negative clutch would correlate.
The argument along the lines of star player with L/R splits will
always see an adverse condition in late innings of a close game
(i.e. opposing manager will always bring in a AH pitcher to
face him, where his manager will not pull him for a PH)
>I don't see how you can deny it.) As for "total idiot"? Yes. If you
>prove yourself unwilling to even *consider* evidence that might
>suggest that you are wrong, I would say the term fits nicely.
>So tell me? Does the term fit? Or do you have an open mind?
What about you? If the shoe fits, will you wear it with an open mind?
Rudy