Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site fisher.UUCP
Path: utzoo!decvax!genrad!panda!talcott!harvard!seismo!lll-crg!dual!qantel!ihnp4!mhuxn!mhuxr!ulysses!allegra!princeton!astrovax!fisher!david
From: david@fisher.UUCP (David Rubin)
Newsgroups: net.sport.baseball
Subject: Re: NL catchers, statistics, baseball philosophy
Message-ID: <744@fisher.UUCP>
Date: Mon, 12-Aug-85 12:33:44 EDT
Article-I.D.: fisher.744
Posted: Mon Aug 12 12:33:44 1985
Date-Received: Thu, 15-Aug-85 01:17:10 EDT
References: <408@philabs.UUCP>
Distribution: na
Organization: Princeton University.Mathematics
Lines: 181

[Sorry, but I was too contentious to just let it go!]

>.............You are a statistician, and love all sorts of
> wonderfully contorted statistics. I couldn't care less about on base
> percentage, slugging average, homeruns per at bat, etc. What I like to 
> see is a winning team, which is constructed from winning team players.

In regard to what YOU are interested: How are we to know what makes a
winning team?  And, more important to the issue at hand, how are we to
know to what an extent an INDIVIDUAL contributes to victory?  What
makes a player a "winning team player" save for your seal of approval?

> The question was, which was the better pick to start THIS all-star game...

That was the original question, but in your first posting, you
declared Pena to be the better player.  Not the better defensive
player (which would have been too indeterminate a matter for me to
bother responding), not the better player from April till June of 1985
(which I'd dispute, but with far less evidence), but just the better
player.  Your response to my first response did nothing to limit your
claims to defense or 1985; in fact, you challenged me to back up what
I was saying with some stats.  Well, you got 'em. ;-)

> But most importantly of all, we completely disagree on how to measure the
> players, so any argument from your statistical point of view is irrelevant
> to me, just as my argument will be meaningless to you.

I am willing to discuss the merits of various forms of measurement.  It 
appears to me that you are not seriously interested in how we measure
things, so long as they are comfortable for you.  You probably even
wrote your Congressman denouncing the metric system...:-).  I
suspect that had my "contorted" statistics (SA, OB, PO, A, et. al.)
supported your belief system, you would have rushed to embrace them,
much as you rushed to embrace the much more "basic" statistics of BA
and NGGR84 (Number of Gold Gloves Received in 1984).

> When constructing your all-star team, I gladly give you your type
> of player: Parker, Blyleven, etc. These statistical marvels are losers.
> They are not team players. They are concerned with their own numbers,

I'm not that familiar with Blyleven, but I'll vouch for Parker as a
team player.  He's welcome on "my" team.

Most people who apply words such as "character", "clutch", "team player",
etc., are doing so to cloak the discrepancy between their prejudices 
towards certain players and the performance of those players on the field.
The player who grouses but performs brilliantly helps his team win; the
player who lives the life of a saint while producing marginally helps
his team be happy.  Performance on the field, not in the locker room,
is what makes a winning team.  Moreover, the two archetypes I present
are exceedingly rare; almost all baseball players want to play, want
to win, and will do their best to do both.  There wasn't a man on the
NL all-star squad (nor the AL squad, as far I know) who pursues
their "numbers" to the detriment of their teammates.

> ....You can put together a team which will lead the league in all sorts of
> statistical categories, and I will put together a team to win the World
> Series.

If I pick first, you don't stand a chance (well, not much of one).
My team will excel at scoring and preventing runs.  Yours will excel
at "giving 100%".  Their 100% just won't be enough.

>For an example of such a team, look at the Mets. (Don't say this
> is due to Carter. They were the same last year - nearly won the division
> with terrible stats - they are often outhit when winning, and were outscored
> over the course of the season.)

The Mets of 1984 were, in some sense, "lucky": they won 90 games while
being outscored (being outhit seems irrelevant here, as *I* am the one
who has been pointing up the weakness of the link between hits and
runs).  They were able to do this because their bullpen performed awesomely
last season; they thus won a disproportionate share of the close games.

The Mets of 1985 have a mediocre bullpen, but because their starters
have done the job and their offense has picked up tremendously, they
are now outscoring their opponents to the degree one would expect of a
first place team.  If Sisk and Orosco were performing as they did in
1984, the result would be wonderful to behold, indeed.  You're right
that the Mets of 1985 are performing nearly as well (actually, you're
wrong, as they are above .600 this season, while ending last season at
.556, but I thought I'd be generous) as in 1984; you are wrong to
conclude that Carter makes little difference.  The Mets haven't just
gained Carter since last season; they lost Brooks and their previously
super bullpen, and they have gained ground.  Carter appears to be very
valuable indeed, if you really want to look at things this way.

> We could go on indefinitely with statistical arguments ("Why do you
> subtract HR's? "Because otherwise the run is counted twice."

A misleading answer.  If a man singles another home, the run is counted
twice: once as a run scored for the man who crosses the plate, another
as an rbi for the batter.  Why treat a homerun differently?

>..............................................."Why do
> you center so much on R and RBI?" "Because I feel runs are all that
> ultimately count on offense...

Runs are all that count, and this is all well and good when discussing
TEAMS: how do we then free the individual from the burden/aid of his
teammates when judging his contribution?  When discussing individuals,
it is more to the point to discuss individual accomplishment.

>.....................................................	There is
> no way that you are going to convince me that the Gold Glove was awarded
> arbitrarily, 

If you accept the opinion of "authority" as final, why do you bother
discussing such things on your own?  Why do you assume that awards are
never given in error?  How else, if not by prejudice or coin flip, is
an award given when two contenders possess equal virtue? Why does
"Gold Glove" == best at position (unless fielding is the only thing
that influences winning)?

>or that your baseball knowledge is superior to their collective knowledge.

How dare I, eh?  How dare any of us, then!  Dissent as sacrilege!
Baseball transformed from art (critics may dispute) and science
(analysts may dispute) to religion!  I think Paul's going to
excommunicate me. :-)

More seriously, the weight of the argument should be counted more than
the weight of its supporters.  Polling is not the path to truth.

>...................Nor will you convince me that the statistics you
> cite are not lineup-dependent (check Madlock's stats batting third vs.
> batting sixth, or before and after Stargell's retirement.) 

I'd just LOVE to discuss them.  Post or mail those stats if you think
they refute me.  That's what discussion is all about, I thought.  I
wonder whether the differences you refer to are due to the things you
say, or are due to you retaining information about runs and rbi's
(which would be affected), or to other factors (I seem to recall 
Madlock having some serious injury problems since Stargell's retirement).

>......................................................The bottom
> line is that ALL statistics are lineup-dependent, e.g., on-base percentage
> increases when batting in front of a good player. 

They may or may not.  However, the difference in OB caused by such a
position in the batting order is certainly far, far, far less than the
difference caused in runs scored.  I do not claim perfection, merely
improvement.

>...........................................After all, no player
> plays baseball all by himself. This is a team game.

True.  But if we still want to evaluate individuals (and we do!), we
must, to the extent possible, ignore the rest of the team!  If it were
a team game (period), you would have never plucked Pena from the
"anonymity" of the Pirate roster to discuss his PERSONAL virtues.

> Furthermore, you say that the Pirate lineup was not that much worse than 
> the Montreal lineup in 83 and 84 because the two teams finished about 
> the same. But OF COURSE, the Pirates did this due to pitching, and the Expos 
> did this due to hitting! Nice feat of legerdemain yourself.

Untrue.  The Pirates outscored the Expos.  I guess you didN'T read
those articles I posted.

>................................................the Pirate
> lineup sucked last year and this year, as anyone can see, whereas the 
> Expo lineup (offense!) was not bad at all.

Saying so doesn't make it so.  The Pirate lineup did "suck" early last
year, but over the course of the entire season, it outperformed many 
other lineups, including the Expos! (This does not show that the Pirate
lineup didn't "suck", only that if it did, so did the Expo lineup.)

> So let us put this argument to rest. You will post your beloved statistics
> and I will ignore them, because I am not, at heart, a statistics fan - 
> I am a baseball fan.

On the contrary, you are quite willing to USE whatever statistics
support your beliefs, regardless of their merits and impervious to
suggestions they don't mean what you say they mean.  I am no less a
baseball fan because I desire to understand the game better.  Are you
more of a fan for spurning information?

					David Rubin
			{allegra|astrovax|princeton}!fisher!david