Take on Griffey
Jack Cooney
>Going into today's game, he's hitting a robust .194. He seems to be
>drawing a lot of walks and knocking in some runs (but I know most people
>in this group don't think RBIs are a real stat). Unless he goes on a
>tear, he'll end the month hitting around the Mendoza line. Do you think
>he'll get it back up to his normal average by season's end, or will he
>hit around .230? I bring it up because I notice a couple people in
>other groups talking about how he's still being productive. Well, he's
>got Dave Kingman stats and most people hated Kingman (not me, I loved
>the guy). What are your (flameless) opinions?
I don't have the numbers to back up this statement, but I'm sure if you
went back and looked at Griffey's career on a month-by-month basis, you'd
find there were months in which he hit .200, others in which he batted
.400, and a whole bunch in between. For his career, he's a .297 hitter.
Let's say Griffey bats .300 the rest of the way (which will probably
happen):
(.200 + .300 * 5 / 6) = .283
I imagine he's pressing just a little bit to impress th hometown
fans. Once he gets settled in, I'm sure he'll start hitting for a higher
average like he normally does.
Ben Grieve was pretty terrible for the first few months last year before
rebounding to bat around .270 for the season. Same goes for Trot Nixon.
Every player will go through slumps. When they happen in April they're
just more apparent.
Regards,
jc
P.S. I presume you play some sort of fantasy baseball. If you can convince
some schmoe to trade him to you cheap, go for it. Conversely, if you own
him, don't get suckered into dealing him.
--
"If you get fresh with me on the mound or do something to show me up,
I`ll drill your ass."
-- Pedro Martinez, Boston Red Sox
Doug Drinen
>
> Let's say Griffey bats .300 the rest of the way (which will probably
> happen):
>
> (.200 + .300 * 5 / 6) = .283
Actually, the Reds have played only 20 games, so 7 and 8 would be more
appropriate numbers to use than 5 and 6. Taking your exercise a step further,
I estimate that if Griffey plays 160 games and hits his career average from
here on out, he'll end up at 286/373/550 with 37 homers. Not quite what Reds'
fans would've hoped, but far from a disaster.
> Every player will go through slumps. When they happen in April they're
> just more apparent.
Absolutely correct.
Doug
Voros
> Going into today's game, he's hitting a robust .194. He seems to be
> drawing a lot of walks and knocking in some runs (but I know most people
> in this group don't think RBIs are a real stat). Unless he goes on a
> tear, he'll end the month hitting around the Mendoza line. Do you think
> he'll get it back up to his normal average by season's end, or will he
> hit around .230? I bring it up because I notice a couple people in
> other groups talking about how he's still being productive. Well, he's
> got Dave Kingman stats and most people hated Kingman (not me, I loved
> the guy). What are your (flameless) opinions?
Axiom #1: Anybody can hit just about anything in 60 At Bats.
--
Voros McCracken
vo...@daruma.co.jp
2000 Projections at:
http://www.enteract.com/~mccracke/project/index.html
Doug Drinen
>
> Axiom #1: Anybody can hit just about anything in 60 At Bats.
Just curious, Voros, why do you call this an Axiom? It's something that could
be verified by "experiment," so it seems more like a Law (in the sense that a
physicist might use the term) to me.
Am I missing something, or did you just think axiom sounded cooler?
Doug
Voros
> Voros wrote:
>>
>> Axiom #1: Anybody can hit just about anything in 60 At Bats.
> Just curious, Voros, why do you call this an Axiom? It's something that could
> be verified by "experiment," so it seems more like a Law (in the sense that a
> physicist might use the term) to me.
I can call it Voros' law if you like, but I'm not sure how to go about
naming laws after myself. I have done the experimentation on this and it
is correct (correlations at that level are mostly no higher than could be
expected by chance).
> Am I missing something, or did you just think axiom sounded cooler?
Didn't really know the difference except that you didn't really need to do
anything but use common sense for an Axiom.
Basil T
Voros <vo...@daruma.co.jp> wrote:
> Doug Drinen <dri...@mailhost.math.dartmouth.edu> wrote:
> > Voros wrote:
> >>
> >> Axiom #1: Anybody can hit just about anything in 60 At Bats.
>
> > Just curious, Voros, why do you call this an Axiom? It's something
that could
> > be verified by "experiment," so it seems more like a Law (in the
sense that a
> > physicist might use the term) to me.
>
> I can call it Voros' law if you like, but I'm not sure how to go about
> naming laws after myself. I have done the experimentation on this and
it
> is correct (correlations at that level are mostly no higher than could
be
> expected by chance).
>
> > Am I missing something, or did you just think axiom sounded cooler?
>
> Didn't really know the difference except that you didn't really need
to do
> anything but use common sense for an Axiom.
That's what I thought you were saying. Perhaps they use the word
incorrectly, but I heard two different announcers doing two different
games use the phrase (I'm paraphrasing) "To quote one of baseball's old
axioms, 'If you have a runner at third w/ no one out, you need to try to
hit it to the right side.'" Strategy or grammar aside, this appears to
be one of the pantheon of baseball "axioms." In fact, I remember my dad
buying me a book with that very title when I was about 9 or 10.
Basically, it was a book on baseball fundamentals accepting as
"universal truths" (my Webster's definition).
Basil T
--
"And when Alexander saw the breadth of
his domain, he wept for there were no
more worlds to conquer. Benefits of a
classical education." -Hans from "Die Hard"
Sent via Deja.com http://www.deja.com/
Before you buy.
Doug Drinen
>
> Voros <vo...@daruma.co.jp> wrote:
> > Doug Drinen <dri...@mailhost.math.dartmouth.edu> wrote:
> > I can call it Voros' law if you like, but I'm not sure how to go about
> > naming laws after myself.
The same way you go about naming Axioms after yourself.
> > I have done the experimentation on this and
> > it is correct (correlations at that level are mostly no higher than could
> > be expected by chance).
Presto, you're perfectly justified calling it Voros' Law.
> > > Am I missing something, or did you just think axiom sounded cooler?
> >
> > Didn't really know the difference except that you didn't really need
> > to do anything but use common sense for an Axiom.
If you call something an axiom, you're asking us to accept it without proof,
based *solely* on common sense. If you call it a law, then you're claiming
you've verified it and that we could too if we wanted to. You can call it an
axiom, but you don't have to, and generally speaking, you want as few axioms
as possible.
Doug
Roger Moore
>Doug Drinen <dri...@mailhost.math.dartmouth.edu> wrote:
>> Voros wrote:
>>>
>>> Axiom #1: Anybody can hit just about anything in 60 At Bats.
>> Just curious, Voros, why do you call this an Axiom? It's something that could
>> be verified by "experiment," so it seems more like a Law (in the sense that a
>> physicist might use the term) to me.
>I can call it Voros' law if you like, but I'm not sure how to go about
>naming laws after myself. I have done the experimentation on this and it
>is correct (correlations at that level are mostly no higher than could be
>expected by chance).
Actually, I think it sounds more like a Conjecture to me than either an
Axiom, Law, or Theorem. A Conjecture is something that presumably can be
proven or demonstrated, but which hasn't been proven or demonstrated
yet. I think that "anyone can hit anything over 60 AB" fits into that
category pretty well.
In any case, I think that Voros's Conjecture is pretty clearly wrong,
albeit in an interesting, correctable way. The problem is that there's
much more room for hitting anything _worse_ than one could normally hit,
but there's some kind of ceiling for players hitting dramatically better
than their previous performance. A good case of this is hitting for
power. A slugger like Griffey could very easily have a 60 AB homer
drought, where he completely failed to hit any dingers. A total
non-slugger like Walt Weiss, though, isn't going to suddenly go through a
60 AB stretch were he hits 10 homers. The interesting thing about this is
that it means that sudden surges in success are more likely real than
equally dramatic drops in success. If Griffey goes without a homer for 60
AB, it just means that he's had a bad 60 AB. If Walt Weiss hits 10 homers
in 60 AB, though, it probably means that he's made a genuine improvement
to his ability to hit for power.
--
Raj (r...@alumni.caltech.edu)
Master of Meaningless Trivia (626) 585-0144
What if there were no hypothetical questions?
Doug Drinen
>
> Voros <vo...@daruma.co.jp> writes:
>
> >Doug Drinen <dri...@mailhost.math.dartmouth.edu> wrote:
> >> Voros wrote:
> >>>
> >>> Axiom #1: Anybody can hit just about anything in 60 At Bats.
>
> >> Just curious, Voros, why do you call this an Axiom? It's something that could
> >> be verified by "experiment," so it seems more like a Law (in the sense that a
> >> physicist might use the term) to me.
>
> >I can call it Voros' law if you like, but I'm not sure how to go about
> >naming laws after myself. I have done the experimentation on this and it
> >is correct (correlations at that level are mostly no higher than could be
> >expected by chance).
>
> Actually, I think it sounds more like a Conjecture to me than either an
> Axiom, Law, or Theorem. A Conjecture is something that presumably can be
> proven or demonstrated, but which hasn't been proven or demonstrated
> yet.
But he just said above, "I have done the experimentation on this and it is
correct."
> In any case, I think that Voros's Conjecture is pretty clearly wrong,
> The problem is that there's
> much more room for hitting anything _worse_ than one could normally hit,
> but there's some kind of ceiling for players hitting dramatically better
> than their previous performance. A good case of this is hitting for
> power. A slugger like Griffey could very easily have a 60 AB homer
> drought, where he completely failed to hit any dingers. A total
> non-slugger like Walt Weiss, though, isn't going to suddenly go through a
> 60 AB stretch were he hits 10 homers.
Probably true, but he can go through 60 AB stretches where his overall
offensive production is equivalent to Griffey's production during a 10 HR in
60 AB stretch. Maybe.
I interpret Voros' Law to mean something like "Anyone can hit within 300
points of his "true" OPS in a 60 AB stretch, or some such. He probably should
at some point say precisely what Voros' Law is, in the technical sense. Then,
having done that, he could revert to using the informal statement: anyone can
hit anything in 60 ABs.
How 'bout it, Voros, could you explain exactly what Voros' Law is?
> The interesting thing about this is
> that it means that sudden surges in success are more likely real than
> equally dramatic drops in success. If Griffey goes without a homer for 60
> AB, it just means that he's had a bad 60 AB. If Walt Weiss hits 10 homers
> in 60 AB, though, it probably means that he's made a genuine improvement
> to his ability to hit for power.
Now *that's* a conjecture.
Doug
Voros
> Basil T wrote:
>>
>> Voros <vo...@daruma.co.jp> wrote:
>> > Doug Drinen <dri...@mailhost.math.dartmouth.edu> wrote:
>> > I can call it Voros' law if you like, but I'm not sure how to go about
>> > naming laws after myself.
> The same way you go about naming Axioms after yourself.
>> > I have done the experimentation on this and
>> > it is correct (correlations at that level are mostly no higher than could
>> > be expected by chance).
> Presto, you're perfectly justified calling it Voros' Law.
>> > > Am I missing something, or did you just think axiom sounded cooler?
>> >
>> > Didn't really know the difference except that you didn't really need
>> > to do anything but use common sense for an Axiom.
> If you call something an axiom, you're asking us to accept it without proof,
> based *solely* on common sense. If you call it a law, then you're claiming
> you've verified it and that we could too if we wanted to. You can call it an
> axiom, but you don't have to, and generally speaking, you want as few axioms
> as possible.
Alright, then I guess it's "Voros' Law."
VOROS' LAW
================
Anyone can hit just about anything in 60 At Bats.
For reference points:
The late Jose Oliva.
Manny Ramirez' first Cup of Coffee in the bigs.
Ken Griffey Jr., 2000
David Bell, 1999
Max Venable, 1989
Todd Haney, 1993?
Shane Spencer, 1998
Voros
>>Doug Drinen <dri...@mailhost.math.dartmouth.edu> wrote:
>>> Voros wrote:
>>>>
>>>> Axiom #1: Anybody can hit just about anything in 60 At Bats.
>>> Just curious, Voros, why do you call this an Axiom? It's something that could
>>> be verified by "experiment," so it seems more like a Law (in the sense that a
>>> physicist might use the term) to me.
>>I can call it Voros' law if you like, but I'm not sure how to go about
>>naming laws after myself. I have done the experimentation on this and it
>>is correct (correlations at that level are mostly no higher than could be
>>expected by chance).
> Actually, I think it sounds more like a Conjecture to me than either an
> Axiom, Law, or Theorem. A Conjecture is something that presumably can be
> proven or demonstrated, but which hasn't been proven or demonstrated
> yet. I think that "anyone can hit anything over 60 AB" fits into that
> category pretty well.
> In any case, I think that Voros's Conjecture is pretty clearly wrong,
> albeit in an interesting, correctable way. The problem is that there's
> much more room for hitting anything _worse_ than one could normally hit,
> but there's some kind of ceiling for players hitting dramatically better
> than their previous performance. A good case of this is hitting for
> power. A slugger like Griffey could very easily have a 60 AB homer
> drought, where he completely failed to hit any dingers. A total
> non-slugger like Walt Weiss, though, isn't going to suddenly go through a
> 60 AB stretch were he hits 10 homers. The interesting thing about this is
> that it means that sudden surges in success are more likely real than
> equally dramatic drops in success. If Griffey goes without a homer for 60
> AB, it just means that he's had a bad 60 AB. If Walt Weiss hits 10 homers
> in 60 AB, though, it probably means that he's made a genuine improvement
> to his ability to hit for power.
That's the more or less part. Obviously there are boundaries in
performance that are difficult to reach.
But in the case stated above, while it's not likely that Weiss would hit
10 Homers in 60 At Bats, if he does that still doesn't mean he's
established an astoundingly high new level of performance. If he hits 10
in 60 at bats, it's still reasonable to believe the possibility that he
might only hit five more all year. After David Bell's big surge last year,
he hit home runs at about the same rate he always had in his career before
the surge.
While statistically such a result is very significant, it would only be so
if the 60 at bats were representative of the type of at bats he'd have in
an entire season.
This is the other big problem with sample sizes. That not only is it
tought to make statistical inferrences from such a small sample, but that
60 at bats are very often highly concentrated in a few number of parks,
pitchers and weather conditions all of which could skew the results one
way or the other.
The fact remains that correlations from 60 AB samples with future
performance are VERY low and therefore what happens in those samples is
not particularly valuable in estimating future performance.
Voros
> Roger Moore wrote:
>>
>> Voros <vo...@daruma.co.jp> writes:
>>
>> >Doug Drinen <dri...@mailhost.math.dartmouth.edu> wrote:
>> >> Voros wrote:
>> >>>
>> >>> Axiom #1: Anybody can hit just about anything in 60 At Bats.
>>
>> >> Just curious, Voros, why do you call this an Axiom? It's something that could
>> >> be verified by "experiment," so it seems more like a Law (in the sense that a
>> >> physicist might use the term) to me.
>>
>> >I can call it Voros' law if you like, but I'm not sure how to go about
>> >naming laws after myself. I have done the experimentation on this and it
>> >is correct (correlations at that level are mostly no higher than could be
>> >expected by chance).
>>
>> Actually, I think it sounds more like a Conjecture to me than either an
>> Axiom, Law, or Theorem. A Conjecture is something that presumably can be
>> proven or demonstrated, but which hasn't been proven or demonstrated
>> yet.
> But he just said above, "I have done the experimentation on this and it is
> correct."
>> In any case, I think that Voros's Conjecture is pretty clearly wrong,
>> The problem is that there's
>> much more room for hitting anything _worse_ than one could normally hit,
>> but there's some kind of ceiling for players hitting dramatically better
>> than their previous performance. A good case of this is hitting for
>> power. A slugger like Griffey could very easily have a 60 AB homer
>> drought, where he completely failed to hit any dingers. A total
>> non-slugger like Walt Weiss, though, isn't going to suddenly go through a
>> 60 AB stretch were he hits 10 homers.
> Probably true, but he can go through 60 AB stretches where his overall
> offensive production is equivalent to Griffey's production during a 10 HR in
> 60 AB stretch. Maybe.
> I interpret Voros' Law to mean something like "Anyone can hit within 300
> points of his "true" OPS in a 60 AB stretch, or some such. He probably should
> at some point say precisely what Voros' Law is, in the technical sense. Then,
> having done that, he could revert to using the informal statement: anyone can
> hit anything in 60 ABs.
> How 'bout it, Voros, could you explain exactly what Voros' Law is?
It's the weekend, so I guess I can dig up the data. There's a lengthy post
on it on Baseball Boards dot Com. Of course, I'm currently trying to look
up the URL and they're moving the site.
Basically I took a series of 50-100 AB Major league seasons in recent
history and compared them with minor league seasons of 300 ABs or more the
same year, to see which was a better indicator of future performance. The
discussion centered around Mark Quinn, and my position was that since
Quinn was a good hitter in AAA, his MLB cup of coffee did not indicate
anything that his AAA numbers didn't and therefore was not particularly
useful in assessing his future (which I said at the time looked good as a
hitter nevertheless).
Anyway as one might expect, the higher minor league sample correlated
better with the following season than the cup of coffee did. What was
amazing was how poorly the cup of coffee really correlated.
One key note was the following:
There were fifty players who had not played any real time in the majors
before the season (the actual def. was more precise but I forget), had at
least 300 ABs in the minors that year, had from 50-100 ABs in the majors
that year, and posted 200 ABs in the majors the next year.
The highest OPS of the 50 was Shane Spencer in 1998.
The lowest OPS of the 50 was Manny Alexander in 1996? or 1995? (around
.240! or something ridiculously low).
The following year Alexander had a higher OPS than Spencer did in 1999.
If, in a 60 At Bat sample, the highest OPS player is no guarantee to
outhit the lowest OPS player over the next couple hundred at bats, it
seems to me the value of a 60 at bat sample is limited to say the least.
>> The interesting thing about this is
>> that it means that sudden surges in success are more likely real than
>> equally dramatic drops in success. If Griffey goes without a homer for 60
>> AB, it just means that he's had a bad 60 AB. If Walt Weiss hits 10 homers
>> in 60 AB, though, it probably means that he's made a genuine improvement
>> to his ability to hit for power.
> Now *that's* a conjecture.
And of course some recent examples, like Spencer or Jose Oliva, tend to
disagree. Using something like Binomial Distributions, yes such a breakout
is VERY significant, but a 60 AB sample is different because the
probabilities could be skewed upward or downward severely by facing a
higher concentration of good or bad pitchers, (Voros has Deja Vu moment,
strange) good or bad hitters parks, or good or bad weather.
Not only is the difference between someone who's good at hitting the
target and someone who is bad relatively small (a peculiar aspect of
baseball) but the target is moved somewhat randomly after each throw to a
new location that may be easier or harder to hit. In 60 tries one might
get a higher concentration of "easy" targets than another.
Ron Johnson
>
>Alright, then I guess it's "Voros' Law."
>
>VOROS' LAW
>================
>Anyone can hit just about anything in 60 At Bats.
>
>For reference points:
>
>The late Jose Oliva.
>Manny Ramirez' first Cup of Coffee in the bigs.
>Ken Griffey Jr., 2000
>David Bell, 1999
>Max Venable, 1989
>Todd Haney, 1993?
>Shane Spencer, 1998
Brett Boone
July 1995 .367/.417/.755 107 PAs
August 1995 .152/.252/.212 112 PAs
--
RNJ
Voros
> Brett Boone
Most of Troy Glaus' 1999 looked like that.
Hopefully he can avoid another repeat of last May and solidify himself as
an All-Star candidate.
James Weisberg
Doug Drinen <dri...@mailhost.math.dartmouth.edu> wrote:
>If you call something an axiom, you're asking us to accept it without proof,
>based *solely* on common sense. If you call it a law, then you're claiming
>you've verified it and that we could too if we wanted to. You can call it an
>axiom, but you don't have to, and generally speaking, you want as few axioms
>as possible.
A minor nitpick. You want as *simple* as axioms as possible. That way
the "correctness" and/or acceptance of them is more likely. Increasing the
number of axioms while decreasing their complexity is a valuable tradeoff.
--
World's Greatest Living Poster