Many resources are needed to download a project. Please understand that we have to compensate our server costs. Thank you in advance. Project price only 1 $
You can buy this project and download/modify it how often you want.
From: [email protected] (Harold_Brooks)
Subject: Re: Bases loaded walk gives Reds win in 12
Organization: Glazier, Texas Urban Renewal Task Force
Lines: 101
In article [email protected] (Mark Singer) writes:
>In article <[email protected]> [email protected] writes:
>>In article [email protected], [email protected] (Mark Singer) writes:
>
>>>During spring training I made a similarly innoncent-looking comment
>>>about clutch hitting on this bb and the flames were flying. "no such
>>>thing as clutch hitting" they (the SDCN's) all screamed. I assumed
>>>they also meant there was no such thing as any kind of clutch performance,
>>>given their comments. I'm still licking the wounds, but I do have
>>>a rebuttal planned. :) (gotta do my homework, as they say)
>>
[some deletions]
>> Rather they showed, quite convincingly, that
>> past clutch performance has never been able to predict FUTURE
>> clutch performance.
>
>I'm sure *you* are convinced, Dave. Apparently so is Tony Perez.
>I, however, am not.
>
>
>I believe at the time of the Sabo plate appearance that Juan Samuel
>was on the bench available for pinch-hit duty. Over the past four
>seasons:
>
> Non-Clutch Clutch
> AB H BA AB H BA
>
>Sabo 1539 452 .294 259 59 .228
>Samuel 1564 383 .245 278 83 .299
>
>I do *not* claim that this is a complete statistical analysis that
>proves the existence of clutch hitting, or compelling testimony that
>Samuel would have gotten a base hit. I'm sure there is something wrong
>with the sample size, or that the basic assumption that clutch
>performance is random invalidates any conclusions from this limited
>application. Or something or other. I don't pretend to understand
>statistical analysis.
>
Well, we agree on the last part.:-) One of the basic things you need to
have in a statistic to be able to predict a player's performance on it
in the future is for there to be a correlation from year to year. A
player's batting average is correlated fairly well from year to year.
A player's ability to walk or infielder's Defensive Average are correlated
better. That is to say, given their past performance in those statistics
we can have a pretty good handle on how they'll do next year. Put in
some simple information about aging and you can do even better. One of
the basic problems with something like "clutch" batting average - overall
batting average is that the correlation from year to year is almost zero.
Adding to the sample size doesn't seem to help much. As a counterexample to
what you showed, consider the following two players from 1984-1987:
Non-Clutch Clutch
AB H BA AB H BA
Maldonado 1060 260 .245 254 78 .307
Lemon 1643 457 .278 256 57 .223
If you had had these two players in 1988, by your logic, in those "clutch"
situations, you'd bat Maldonado for Lemon in a blink of an eye. Well, in
1988, Maldonado hit .267 in "non-clutch" and .190 in "clutch", while Lemon
hit .254 in "non-clutch" and .313 in "clutch". Before you accuse me of
completely cooking the data, there were 96 players who had 25+ "clutch"
at bats every year from 1984-1988 (according to Elias). As a simple
measure of clutchness, let's just look at clutch BA-non-clutch BA. If you use
'84-'87 to predict '88 for those 96 players, you'd find that 27 of the
96 were below league average in that measure both in '84-87 and in '88,
26 were above league average in both periods and the other 43 were
above in one and below in the other. If you were just flipping coins,
you'd expect to get 24 above/below in both and 48 that switched. The
difference between the observed results and the coin flip experiment
is not statisically significant.
BTW, correlating players' _overall_ batting average from '84-'87 with
'88 gives a correlation coefficient of 0.59, which is significant at
something better than the 99.9% confidence level. Correlating their
(clutch-non-clutch BA) for the same period gives a correlation of
0.088, significant at no level of any interest.
>
>
>> Everyone would agree (I hope) that a grand
>> slam in the bottom of the ninth when your team is down by three
>> is a VERY clutch hit. There is just no way to predict who is more
>> likely to get the clutch hit based on past performance (hitting in
>> "clutch" situations).
>>
>
>If you say so, Dave. No way.
>
Actually, it's technically incorrect to say that we can't predict future
clutch performance. It's more correct to say that we can't predict
future clutch performance with any skill.
Harold
--
Harold Brooks [email protected]
National Severe Storms Laboratory (Norman, OK)
"I used to work for a brewery, too, but I didn't drink on the job."
-P. Bavasi on Dal Maxvill's view that Florida can win the NL East in '93
