One More Question Of Bobnob And Statheads
Christina
what would the true value of a double, triple, homerun be? I am just
annoyed lately of slugging percentage, it seems that a triple is very
similar to a double, is it really worth 150% of a double? Is it
really a 1:2:3:4 ratio for total bases, if not what do you think is
more represantive, maybe we could come up with a better version of
slugging percentage?
Jeff Schwab
It may be interesting to keep count, over the course of several seasons,
the number of singles/walks, doubles, and triples, across the American
League. One could then track what percentage of runners starting at
each base end up making it home. We can count a home run as always
being worth exactly one run (since any runs scored by players already on
base will be counted separately), and each other base hit can then be
valued as some "percentage of a run," defined by the probability that
the runner will make it home.
For example, suppose that for 60% of triples, the runner ends up getting
home, but the same is true for only 40% of doubles. A triple would then
be valued at 0.600 of a run, and a double at 0.400. A triple is worth
150% of a double with these hypothetical numbers, but that's just
because I pulled them out of thin air; given enough data, you could make
precise, meaningful comparisons.
Are these data available to the public? It wouldn't be hard to put the
stat together, if you had the data in a bunch of flat text files.
NK
wrote:
Regarding a double vs a triple. A quick Google search indicates
that, in 2007, runners on first scored 44% of the time on a double.
And I don't think many runners that were already on base get thrown
out at the plate on a triple. I think a homer might be undervalued,
and a single overvalued... but I don't have anything to back that up
other than my gut feeling.
--
NK
McDuck
One would need to take into account the number of outs at the time of
the hit. A HR is not a whole lot better than a triple if there are no
outs (I'd guess the runner on 3rd scores 80% of the time). But with
two outs, a HR is worth a LOT more than a triple. Similarly, a double
and a triple are a lot closer in value with no outs than with one out,
and closer to equal with two outs than with one out.
Gil Gamesh
"McDuck" <wallyDELE...@comcast.net> wrote in message
news:h88du4d2nds5sr4em...@4ax.com...
> One would need to take into account the number of outs at the time of
> the hit. A HR is not a whole lot better than a triple if there are no
> outs (I'd guess the runner on 3rd scores 80% of the time). But with
> two outs, a HR is worth a LOT more than a triple. Similarly, a double
> and a triple are a lot closer in value with no outs than with one out,
> and closer to equal with two outs than with one out.
That's all well & good but who's on first?
NK
<wallyDELE...@comcast.net> wrote:
>One would need to take into account the number of outs at the time of
>the hit. A HR is not a whole lot better than a triple if there are no
>outs (I'd guess the runner on 3rd scores 80% of the time). But with
>two outs, a HR is worth a LOT more than a triple. Similarly, a double
>and a triple are a lot closer in value with no outs than with one out,
>and closer to equal with two outs than with one out.
You're only looking at the runs scored potential. You have to factor
in the RBI potential as well.
--
NK
Giovanni Wassen
After a triple you have the chance to score on a squeeze-bunt, passed ball,
infield hit, sacrifice fly, after a double you (normally) can't. So, yes, a
triple has more value.
--
Gio
http://www.watkijkikoptv.info
http://myanimelist.net/profile/extatix
http://watkijkikoptv.info/animeblog
Ralph Kramden
> Let us say that a single has the value of 1. If this is the case,
> what would the true value of a double, triple, homerun be?
What you're thinking of is known in Sabermetrics as "Linear Weights",
invented by Pete Palmer*. Here's a good ref:
http://www.tangotiger.net/wiki/index.php?title=Linear_Weights
The run-scoring contribution of particular batters' events are
determined by linear regression. Here's what Tango comes up with in
that reference:
Single: .49, Double: .61, Triple: 1.14, HR 1.5, Walk: .33, SB: .14
*Tango notes that a similar analysis was done as far back as 1917, by
F.C.Lane:
http://www.tangotiger.net/fclane/
Jeff Schwab
How so? If you drive in a run, that guy was already on base, so the run
will count toward the value of the runner's original base hit.
Jeff Schwab
That's one of a large number of important subtleties. For example, one
also could note the score at the time the runner comes home. Not all
runs are created equal.
I think this was a lot of fans' beef with A-Rod a couple of years ago:
That he was hitting well on paper, but the runs were coming when they
were irrelevant. Runs surely could be weighted according to "likely
relevance."
bi...@ix.netcom.com
Thank you Ralph this is EXACTLY what I was looking for!!! So this is
similar to the linear weights I suppose? My husband has a book of
Lane, it is very interesting, he was way ahead of his time with the
numbers!
Ok let me see, if we move up a single to 1, we can have a double 1.24,
triple 2.33, homerun 3.06. I wonder how the slugging would look with
that? It might be interesting, maybe some who hit alot of doubles
like Boggs and Reed would be lowered a little? Also we could make a
walk 0.67 and go by plate appearances, maybe create a new version of
OPS that is more accurate? I wish I had time for this today, maybe
another day? Not sure for the steals, maybe they should be a seperate
category along with defense?
Ralph Kramden
> On Apr 16, 9:54 am, Ralph Kramden <szs.weba...@gmail.com> wrote:
>
> > On Apr 15, 7:57 pm, Christina <bi...@ix.netcom.com> wrote:
>
> > > Let us say that a single has the value of 1. If this is the case,
> > > what would the true value of a double, triple, homerun be?
>
> > What you're thinking of is known in Sabermetrics as "Linear Weights",
> > invented by Pete Palmer*. Here's a good ref:http://www.tangotiger.net/wiki/index.php?title=Linear_Weights
>
> > The run-scoring contribution of particular batters' events are
> > determined by linear regression. Here's what Tango comes up with in
> > that reference:
> > Single: .49, Double: .61, Triple: 1.14, HR 1.5, Walk: .33, SB: .14
>
> > *Tango notes that a similar analysis was done as far back as 1917, by
> > F.C.Lane:http://www.tangotiger.net/fclane/
>
> Thank you Ralph this is EXACTLY what I was looking for!!! So this is
> similar to the linear weights I suppose?
> walk 0.67 and go by plate appearances, maybe create a new version of
> OPS that is more accurate? I wish I had time for this today, maybe
> another day? Not sure for the steals, maybe they should be a seperate
> category along with defense?
It sounds like Tango's site is for you. Go there and poke around and
you'll find a whole bunch of interesting stuff.
His favorite batter's evaluation tool is "wOBA" (weighted on-base
average) which he defines as:
(0.72xNIBB + 0.75xHBP + 0.90x1B + 0.92xRBOE + 1.24x2B + 1.56x3B +
1.95xHR) / PA
bi...@ix.netcom.com
>
> - Show quoted text -
Wow Ralph you are the BEST! This is exactly what I am looking for, I
LOVE it, it seems to be a better number than OPS even if it is not as
easy to figure out. Thank you thank you thank you!!!!
Ron Johnson
You're talking linear weights (in some form)
It actually gets pretty complicated because the cost of an
out is not a constant (the higher the offensive context
the more costly an out). Hence Pete Palmer's formula
achieves its accuracy by applying a "slope corrector"
after the initial results have been calculated.
Jim Furtado's Extrapolated runs is the best (relatively)
simple approach.
XR - Extrapolated Runs =
(.50 × 1B) +
(.72 × 2B) +
(1.04 × 3B) +
(1.44 × HR) +
(.34 × (HP+TBB-IBB)) +
(.25 × IBB)+
(.18 × SB) +
(-.32 × CS) +
(-.090 × (AB - H - K)) +
(-.098 x K)+
(-.37 × GIDP) +
(.37 x SF) +
(.04 × SH)
Ron Johnson
Steve Mann studied this literally decades ago. The resulting
formula (with help from Pete Palmer -- the guy who came up
with linear weights) is a mess but works reasonably well.
About 30 years ago Steve Mann broke down the various
offensive events into their run and rbi components.
Method for the few who care.
(((BB*.25)+(HBP*.29)+(1B*.51)+(2B*.82)+(3B*1.38)+(HR*2.63)+
(SB*.15)-(CS*.28))*.52)+(.008*(AB+BB+HBP))+((3*(AB+BB+HBP)/6200)*
(1000*(OBP-.330)))
As I said, a mess. I've probably got redundant brackets or one
missing.
Here's an example - Eric Davis 1987
(The second last column is run value -- IE a single is worth .28
runs scored and .23 rbi)
# value total runs rbi
BB 84 .25 21.00 20.16 .84 .24 .01
HB 1 .29 .29 .28 .01 .28 .01
1B 75 .51 38.25 21.00 17.25 .28 .23
2B 23 .82 18.86 8.51 10.35 .37 .45
3B 4 1.37 5.48 2.44 3.04 .61 .76
HR 37 2.63 97.31 37.00 60.31 1.00 1.63
SB 50 .15 7.50 7.50 .00 .15 .00
CS 6 -.28 -1.68 -1.68 .00 -.28 .00
OBA .399 18.76
PA 562 4.50
120.50 95.21 91.80 .51
Basically he contributed 95 runs and 92 rbi. His excellent OBP
gave other hitters a chance to contibute 19 net runs (ie runs
or rbi) and 4.5 runs for general health. (Mann found that he could
for a small portion of runs and used what he called a garbage
constant) Roughly 51% of his value was in run scoring. Nothing
terribly unusual and therfore no particular reason to think that
runs created would under-value him.
A player's actual RBI total is actually a function of power and
opportunity and can be radically different from the number
Mann's formula predicts. Similarly a player's run total is
profoundly influenced by the guys batting after him. Mann's
own research shows that a player will get about 10% more
combined runs and rbi than predicted if he bats 4th.
Batting order position profoundly influences the shape of
the counter stats too. A guy batting first will see his
runs scored go up and rbi down in a fairly predictable
manner.
Ron Johnson
If what you're looking for is a relatively simple, generally
effective evaluation tool, OBP*1.7+SLG works pretty well.
It's certainly good enough (so for that matter is OPS) for
partial season evaluation. There's a huge amount of noise
in a partial season's data.
And if you want precision, the best approach is to use
a number of good metrics. Most players will come out
quite similar in any decent metric. (Bear in mind that the
standard error for the best metrics is in the range of 5
runs for a full-time player)
The guys who come out different are the really interesting
cases. Runs created will predictably over-value the
very best (because it's a multiplicative method it
asserts than a home run hit by Frank Thomas is more
valuable than one hit by Joe Carter) and will predictably
under-value high OBP, low SLG players.
Other methods have issues with other types of players.
IOW the formula should be where the discussion starts.
Christina
This is great information, VERY interesting and exactly what I am
looking for, thanks Ron!!!
Jeff Schwab
> (((BB*.25)+(HBP*.29)+(1B*.51)+(2B*.82)+(3B*1.38)+(HR*2.63)+
> (SB*.15)-(CS*.28))*.52)+(.008*(AB+BB+HBP))+((3*(AB+BB+HBP)/6200)*
> (1000*(OBP-.330)))
>
> As I said, a mess. I've probably got redundant brackets or one
> missing.
It reminds me of Lisp.
S/B 954RR
"Christina" <bi...@ix.netcom.com> wrote in message
news:fvscu4lh0ckp06rp4...@4ax.com...
> Let us say that a single has the value of 1. If this is the case,
> what would the true value of a double, triple, homerun be? I am just
> annoyed lately of slugging percentage, it seems that a triple is very
> similar to a double, is it really worth 150% of a double?
At first glance I'd say it is. A sac fly or possibly a bunt brings home a
runner from third. It also definitely scores the run from first. It gives
the batter a chance to score on a past ball... Maybe it isn't but I sure
think it is.
S/B 954RR
> One would need to take into account the number of outs at the time of
> the hit. A HR is not a whole lot better than a triple if there are no
> outs (I'd guess the runner on 3rd scores 80% of the time).
How soon we forget. Chicks dig the long ball.
S/B 954RR
"Jeff Schwab" <je...@schwabcenter.com> wrote in message
news:wvOdnVIPwpFtRXrU...@giganews.com...
You calling Ron gay?
OceanView
I always thought an interesting stat would be "base runners advanced." A
single with the bases loaded is three, a double with two on is four (your
bases, plus one for each runner), a grand slam is seven. A GIDP is a -1.
It would mainly be a measure of clutch hitters rather than slugging,
though certainly a high SLG would have a high BRA (couldn't resist the
acronym!).
--
-------
I am short enough and ugly enough to succeed on my own. - Woody Allen
dano
"OceanView" <F...@chance.org> wrote in message
news:Xns9BEFEC59...@85.214.105.209...
This has got to be the most boring thread of the year.
Bob-Nob
> Christina wrote:
>> Let us say that a single has the value of 1. If this is the case,
>> what would the true value of a double, triple, homerun be? I am just
>> annoyed lately of slugging percentage, it seems that a triple is very
>> similar to a double, is it really worth 150% of a double? Is it
>> really a 1:2:3:4 ratio for total bases, if not what do you think is
>> more represantive, maybe we could come up with a better version of
>> slugging percentage?
> It may be interesting to keep count, over the course of
> several seasons, the number of singles/walks, doubles, and
> triples, across the American League. One could then track
> what percentage of runners starting at each base end up making
> it home.
I'm pretty sure you guys can find this information if you ask politely
over at either www.baseballthinkfactory.com or www.thehardballtimes.com
(or wherever Tango or MGL post now). This stuff has been studied -- I
don't have the numbers ready to hand, but there are people who do.
The sites I named are the places I'd look for that information.
<snip>
> Are these data available to the public? It wouldn't be hard to put
> the stat together, if you had the data in a bunch of flat text files.
Readily avilable? I don't think so, but I'm pretty sure it has
been posted to baseballthinkfactory at some point or other.
--Robert Machemer
--
Robert Paul Aubrey Machemer | For each time he falls, he shall
Amherst College, Math & Classics | rise again, and woe to the wicked!
Space Judge!!! | --Don Quixote (Man of La Mancha)
"He had everything a man could want: power, grace, and style..."
Ron Johnson
Thanks. Brings back old memories. (Old computer guy)
Ron Johnson
> Jeff Schwab <j...@schwabcenter.com> wrote innews:E_Odna9ZUfVq3nrU...@giganews.com:
>
>
>
>
>
> > NK wrote:
> >> On Wed, 15 Apr 2009 23:11:57 -0400, McDuck
> >> <wallyDELETEMEMcD...@comcast.net> wrote:
>
> >>> One would need to take into account the number of outs at the time of
> >>> the hit. A HR is not a whole lot better than a triple if there are no
> >>> outs (I'd guess the runner on 3rd scores 80% of the time). But with
> >>> two outs, a HR is worth a LOT more than a triple. Similarly, a double
> >>> and a triple are a lot closer in value with no outs than with one
> out,
> >>> and closer to equal with two outs than with one out.
>
> >> You're only looking at the runs scored potential. You have to factor
> >> in the RBI potential as well.
>
> > How so? If you drive in a run, that guy was already on base, so the
> run
> > will count toward the value of the runner's original base hit.
>
> I always thought an interesting stat would be "base runners advanced."
Done a few year back by James Tuttle. He found clear evidence
that batting order profoundly influenced the result.
A more complex version of this is called (among other names)
Run Probability Added.
Basic concept -- you know the expected runs scored for
every base/out situation, so you credit the batter
with the difference in game state before and after
the plate appearance (with adjustments for what happens
on the base)
And you can take the analysis one further step and
take the score into account. This gives you win
probability added or wpa.
Both rpa and wpa are subject to batting order issues.
Leverage Index (LI) is an attempt to adjust for this.
(WPA has other issues.)
You can now get wpa and li at baseball-reference.com
(as well as a more detailed explanation)
Jeff Schwab
> On Apr 16, 11:14 pm, OceanView <F...@chance.org> wrote:
>> I always thought an interesting stat would be "base runners advanced."
>
> Done a few year back by James Tuttle. He found clear evidence
> that batting order profoundly influenced the result.
If it didn't, we'd have to stop bickering about line-ups. :)
> You can now get wpa and li at baseball-reference.com
> (as well as a more detailed explanation)
What are the most trusted sources for raw data? Google turns up
multiple sources.
http://baseball1.com/content/view/57/82/
http://baseball-databank.org/
Jeff Schwab
> "McDuck" <wallyDELE...@comcast.net> wrote in message
> news:h88du4d2nds5sr4em...@4ax.com...
>
>> One would need to take into account the number of outs at the time of
>> the hit. A HR is not a whole lot better than a triple if there are no
>> outs (I'd guess the runner on 3rd scores 80% of the time).
>
> How soon we forget. Chicks dig the long ball.
Good stuff. :)
Dano
Says the guy that just stated that RBI stats are meaningless. OK.
Dano
>
Let me just add...I'm sorry to offend all you "statheads". Maybe I
shouldn't have given my personal opinion of your obsession with numbers and
statistics...whatever floats your boat I suppose. I just feel like this
gets carried to a truly laughable extreme when you not only come up with the
most obscure and complex formulas to support your arguments, but then
declare that the simple and obvious...not to mention time honored stats like
RBI are overrated or even irrelevant.
As for politics? I think it's clear by the responses and lengths of the
political threads that they are by far more interesting and participated in,
than your stat discussions to most here. I very rarely start any of those
OT, political threads, but I will exercise my right to speak out in response
where I wish. I think your problem with me is much more about my politics.
That's your right as well to disagree with me. ANY discussion would be
boring and worthless if we all agreed, all the time. Feel free to ignore if
you don't like it. Or state your own damn opinion. If you were to block
all of us who participate in the political threads, you wouldn't have many
left to discuss anything with.
PETER SHORTS
> je.s.t...@hehxduhmp.org wrote:
where do you get off saying that it's ok for you to say that this
baseball discussion is boring, but it's not ok for him to say the same
about political discussions in a fucking baseball group. do us all a
favor and shoot yourself.
Dano
> Dano <janea...@yahoo.com> wrote:
>>> It's better than 99.99999% of the political crap you all spew.
>> Says the guy that just stated that RBI stats are meaningless. OK.
>
>
> Except for the most macro of scales, where it tends to simply track
> with "good hitter anyways", RBI *is* meaningless because all it says
> is "i got a hit with someone in scoring position". The same hitter
> could hit exactly the same in two different seasons, if one time
> people were on base he'd have RBI and if the other he didn't he'd
> have no RBI. HOw is that at *all* a useful metric for a batter?
So all the good RBI hitters are merely a byproduct of where they hit in a
batting order? Really? Explain why some hitters accumulate more RBI year
after year...from one team to another. Are the good production guys put in
prime position to deliver by accident? In other words...which came first?
Chicken or egg? If you bat Jason Varitek fourth, will he deliver RBI like
Manny or Youk? Of course having runners on base is part of any metric...but
some hitters produce more than others no matter where they hit. That's WHY
they are put in positions to bat after good OBP guys. If they don't
produce, they won't bat in that slot for very long. But just because there
are other relevant factors...statistics that explain the reasons...does NOT
negate the importance of a very basic stat like RBI.
Dano
>> Says the guy that just stated that RBI stats are meaningless. OK.
>
mario in victoria
> je.s...@hehxduhmp.org wrote:
>> dano <janea...@yahoo.com> wrote:
>>> This has got to be the most boring thread of the year.
>> It's better than 99.99999% of the political crap you all spew.
>
> Let me just add...I'm sorry to offend all you "statheads". Maybe I
> shouldn't have given my personal opinion of your obsession with numbers and
> statistics...whatever floats your boat I suppose. I just feel like this
> gets carried to a truly laughable extreme when you not only come up with the
> most obscure and complex formulas to support your arguments, but then
> declare that the simple and obvious...not to mention time honored stats like
> RBI are overrated or even irrelevant.
But that's part of the beauty of baseball. The game is so constructed
that it lends itself very well to the 'playing with statistics' models.
No other sport is like that. Baseball probably has the most complex
rules, traditions and situational occurrences of any other sport. I
don't necessarily do it, but I find the fact that the 'stats' can be
measured, compared, diluted, re-funneled and gone over time and again
extremely interesting and just plain fun.
> As for politics? I think it's clear by the responses and lengths of the
> political threads that they are by far more interesting and participated in,
> than your stat discussions to most here. I very rarely start any of those
> OT, political threads, but I will exercise my right to speak out in response
> where I wish. I think your problem with me is much more about my politics.
> That's your right as well to disagree with me. ANY discussion would be
> boring and worthless if we all agreed, all the time. Feel free to ignore if
> you don't like it. Or state your own damn opinion. If you were to block
> all of us who participate in the political threads, you wouldn't have many
> left to discuss anything with.
I'm surely an 'offender' here, and I'm on your side, Dano, but it's a
pity that so many off-topic discussions turn into name-calling and such.
Ah, well.
mario in victoria
--
both can be fun
Ron Johnson
> Ron Johnson wrote:
> > On Apr 16, 11:14 pm, OceanView <F...@chance.org> wrote:
> >> I always thought an interesting stat would be "base runners advanced."
>
> > Done a few year back by James Tuttle. He found clear evidence
> > that batting order profoundly influenced the result.
>
> If it didn't, we'd have to stop bickering about line-ups. :)
>
> > You can now get wpa and li at baseball-reference.com
> > (as well as a more detailed explanation)
>
> What are the most trusted sources for raw data? Google turns up
> multiple sources.
www.retrosheet.org is the root source for all play by play data.
Complete seasons going back to the mid-50s with other stuff
(like the Dodger data files with pitch counts going back to
1946) available by request to Dave Smith (the guy who runs
Retrosheet)
Baseball-reference.com has tools like PI (play index)
that allow you to do all manner of searches on the
processed raw data.
And there's Chadwick -- still in beta -- that allows to
to run your own stuff. It's a sourceforge project.
And since the retrosheet data files are in a well documented
format you probably wouldn't have much trouble in building
your own parser (which is the way guys like Tom Ruane
work with the data)
Ron Johnson
I know you can get the base/out state tables at billjamesonline.net
Yeah, pay site, but it's dirt cheap.
Here's an old base/out table -- values won't be too different today.
Bases Outs
0 1 2
empty 0.57 0.31 0.12
1st 0.97 0.60 0.27
2nd 1.18 0.73 0.33
1st, 2nd 1.63 1.01 0.48
3rd 1.52 1.00 0.41
1st, 3rd 1.92 1.24 0.52
2nd, 3rd 2.05 1.50 0.64
1st, 2nd, 3rd 2.54 1.70 0.82
In case it's not clear these are the average number of
runs scored in an inning from the various base out
situations (innings where 3 outs weren't recorded not
included)
And the answer to a slightly different question:
This one shows the probability of scoring at least one run
given the 24 bases/out situations:
Bases Outs
0 1 2
empty .30 .18 .08
1st .45 .29 .15
2nd .64 .42 .22
1st, 2nd .66 .44 .24
3rd .86 .67 .28
1st, 3rd .88 .66 .29
2nd, 3rd .85 .71 .29
1st, 2nd, 3rd .89 .69 .34
(2000 season data)
Ron Johnson
> Dano <janeandd...@yahoo.com> wrote:
> > So all the good RBI hitters are merely a byproduct of where they hit in a
> > batting order? Really? Explain why some hitters accumulate more RBI year
>
> couple of weeks ago. Feel free to go back and look it up in the archives.
>
> Long story short: You're wrong. RBI is largely meaningless (except that
> alltime RBI leaders also tend to have been alltime hitting greats and
> vice versa, IE a very macro level) for a per-player measurement.
No. Not meaningless. Tom Ruane did a very detailed study of rbi
(available in the research area at Retrosheet. Anybody
interested in this stuff is encouraged to look at the stuff
under http://www.retrosheet.org/Research/Research.htm)
I did a much simpler one and came up with an rbi estimator that
works very well over time and pretty well in any given season.
ABROB*(SLG*1.09-BA*.66) (ABROB is AB with runners on base)
And yes. Two players, equal SLG, same opportunities, normal
distribution of runners the guy with the *lower* BA will
tend to drive in more runs. Partially because there are
a lot more PAs with runners on first and partially because
a single with a RISP is very far from a guaranteed RBI.
Runner on second will rarely score on an infield single
and will only score on about 60% of the singles to left
or right. (Surprised its that low? I was.)
And I've check out the big errors. The primary source is an
unusual distribution of baserunners. It's obviously easier
to drive Ellsbury in from third than Ortiz from first.
Over time this tends to even out, but in any given year
it can be pretty variable.
At the career level there's about a 98% correlation between the
estimates and the actual rbi -- with only one fairly big
miss (Joe Carter). (And Arne Olsen ran another study on
what he called rbi vultures. Long and short of it is
that Carter's teams didn't score more runs than you'd
expect. He just happened to play on teams that had
players consistently good at reaching, reaching scoring
position and running the bases. From Brett Butler to
Paul Molitor and Roberto Alomar)
Anyhow it's not accurate to say rbi carries no signal.
It is however accurate to say that rbis are an extremely
predictable function of power (most simply represented
by SLG) and opportunity.
And opportunity is not under the batter's control. Why
give him any credit?
Dano
>> the political threads that they are by far more interesting and
>> participated in, than your stat discussions to most here. I very
>> rarely start any of those OT, political threads, but I will exercise
>> my right to speak out in response where I wish. I think your
>> problem with me is much more about my politics.
>
> least
> it falls under the "alt.sports.baseball", unlike the political
> garbage.
>
I never said your stuff is off topic. I just gave my opinion of these
boring ass, never ending stat discussions. Have at it as you wish.
Wow...is your skin thin or what?
> I have no idea what side you fall on in the political stuff as I
> almost never read them, and generally stop reading any thraed that
> goes down
> that route.
>
As I generally ignore these stat related threads. Sorry I brought it up.
I'll simply do as I usually do and ignore this bullcrap as much as possible.
Sheesh.
>> you don't like it. Or state your own damn opinion. If you were to
>> block all of us who participate in the political threads, you
>> wouldn't have many left to discuss anything with.
>
> Most of you have been blocked, a few of you occasionaly say
> non-stupid things about baseball so get unblocked from time to time.
Right back at ya. With you...it's simply been your sunny disposition that
gets to me. I usually just ignore you the old fashioned, low tech way.
Whatever...
Dano
> Dano <janea...@yahoo.com> wrote:
>> So all the good RBI hitters are merely a byproduct of where they hit
>> in a batting order? Really? Explain why some hitters accumulate
>> more RBI year
>
> a couple of weeks ago. Feel free to go back and look it up in the
> archives.
>
> Long story short: You're wrong. RBI is largely meaningless (except
> that alltime RBI leaders also tend to have been alltime hitting
> greats and vice versa, IE a very macro level) for a per-player
> measurement.
Yeah, yeah...says you...
Ralph Kramden
> But just because there
> are other relevant factors...statistics that explain the reasons...does NOT
> negate the importance of a very basic stat like RBI.
rm, is that you?
http://groups.google.com/group/rec.sport.baseball/msg/3c8f369ad798d156?hl=en
Dano
> Dano <janea...@yahoo.com> wrote:
>> I never said your stuff is off topic. I just gave my opinion of
>> these boring ass, never ending stat discussions. Have at it as you
>> wish.
>
> pointing out that my complaints about the off-topicness of the
> political threads is somewhat hypocritical as *any* non-red sox
> discussion is technically off topic. I'd argue though that general
> baseball talk
> is far more relevant than political debate.
>
> As they say, alt.politics is thataway ...
Ever poke your head in there? I'd just as soon be stranded on a desert isle
with Rush and Hannity.
PETER SHORTS
so instead you drag it in here. way to go douchebag.
The Gnorkmeister
> Dano wrote:
> > je.s.t...@hehxduhmp.org wrote:
> >> dano <janeandd...@yahoo.com> wrote:
> >>> This has got to be the most boring thread of the year.
> >> It's better than 99.99999% of the political crap you all spew.
>
> > Let me just add...I'm sorry to offend all you "statheads". Maybe I
> > shouldn't have given my personal opinion of your obsession with numbers and
> > statistics...whatever floats your boat I suppose. I just feel like this
> > gets carried to a truly laughable extreme when you not only come up with the
> > most obscure and complex formulas to support your arguments, but then
> > declare that the simple and obvious...not to mention time honored stats like
> > RBI are overrated or even irrelevant.
>
> But that's part of the beauty of baseball. The game is so constructed
> that it lends itself very well to the 'playing with statistics' models.
>
> No other sport is like that. Baseball probably has the most complex
> rules, traditions and situational occurrences of any other sport. I
> don't necessarily do it, but I find the fact that the 'stats' can be
> measured, compared, diluted, re-funneled and gone over time and again
> extremely interesting and just plain fun.
Ah Mario, a rare boo boo by you. It should be "of any sport". Since
baseball is baseball it can't be both baseball and an other sport at
the same time as your writing implies.
>
> > As for politics? I think it's clear by the responses and lengths of the
> > political threads that they are by far more interesting and participated in,
> > than your stat discussions to most here. I very rarely start any of those
> > OT, political threads, but I will exercise my right to speak out in response
> > where I wish. I think your problem with me is much more about my politics.
> > That's your right as well to disagree with me. ANY discussion would be
> > boring and worthless if we all agreed, all the time. Feel free to ignore if
> > you don't like it. Or state your own damn opinion. If you were to block
> > all of us who participate in the political threads, you wouldn't have many
> > left to discuss anything with.
>
> I'm surely an 'offender' here, and I'm on your side, Dano, but it's a
> pity that so many off-topic discussions turn into name-calling and such.
No they don't, you miserable scumbag!
>
> Ah, well.
Exactly
>
> mario in victoria
gnork in Los Angeles
The Gnorkmeister
Very interesting. One thing in particular illustrates my contempt for
ML managers. With a man on 1st and no outs a team has a 45% chance of
scoring a run. With a man on 2nd and one out there is a 42% chance of
scoring. Yet fool after fool, year after year will call for sacrifice
bunts. Baseball managers going on a cruise together would create a
ship of fools.
Dano
Ron Johnson
>
> Very interesting. One thing in particular illustrates my contempt for
> ML managers. With a man on 1st and no outs a team has a 45% chance of
> scoring a run. With a man on 2nd and one out there is a 42% chance of
> scoring. Yet fool after fool, year after year will call for sacrifice
> bunts. Baseball managers going on a cruise together would create a
> ship of fools.
Here's the thing though. Base for an out is almost always a losing
play (though there are specific exceptions). It's also a silly
assumption to make.
This is a few years old, but I doubt the numbers are substantially
different. Here's how attempted sacs broke down in practice.
(This doesn't consider those who fail to get the bunt down
and end up hitting with two strikes on them. Nobody is very
successful in that situation)
Sac 71%
PO 10%
FC 7%
Hit 5%
Sac+E 3%
FCN 2%
DP 2%
(FCN is a fielder's choice with no outs recorded)
Hits are kind of skewed by a few players who reach a substantial
percentage of the time. It's almost never a bad idea to bunt
with a guy who's a good bunter and is fast. IE if base
for an out is the consolation prize and really bad results
are rare and he has some chance of reaching, bunting
will never cost. Not many of these guys around of course.
(Or to put it another way, Earl Weaver is the spokesman in
chief against bunting. And yet he bunted a lot with Belanger,
twice leading the league in sac. Good bunter, fast,
and not giving up all that much in terms of what Belanger
could give you hitting away. If he really needed to
hit away, he constructed his team to be able to pinch-hit
at will for Belanger)
And that's before considering the game theory aspect (which
Tango does deal with). The defense will have to position
themselves in a less that optimum way to deal with the
threat of the bunt. Don't threaten it and you give back
a bit.
It's generally a bad idea to bunt with a position player
who is neither fast enough to have a decent chance of
reaching and isn't a good bunter.
nate
> Ron Johnson <ron7...@gmail.com> wrote:
> > And opportunity is not under the batter's control. Why
> > give him any credit?
>
> prowess, RBI tells us absolutely nothing because it relies
> on opportunity.
i think we all agree that RBI by itself is not 100% a good indicator.
but you are saying that it has 0.0% value. i must disagree.
maybe you'd like to restate it's value between 0% and 100% ?
on the otherhand if we divide RBI by Possible RBI, then the value of
the stat goes up considerably.
or, to get back to the BRA, divide baserunners advanced by possiblt
base runners advance.
in this way, a Grand Slam is worth 10/10. a bases-empty HR is worth
4/4. Runners on 1st & 3rd with a homerun is worth 8/8.
- nate
mario in victoria
> On Apr 17, 10:50 am, mario in victoria <mario5...@hotmail.com> wrote:
<snip>
>> But that's part of the beauty of baseball. The game is so constructed
>> that it lends itself very well to the 'playing with statistics' models.
>>
>> No other sport is like that. Baseball probably has the most complex
>> rules, traditions and situational occurrences of any other sport. I
>> don't necessarily do it, but I find the fact that the 'stats' can be
>> measured, compared, diluted, re-funneled and gone over time and again
>> extremely interesting and just plain fun.
>
> Ah Mario, a rare boo boo by you. It should be "of any sport". Since
> baseball is baseball it can't be both baseball and an other sport at
> the same time as your writing implies.
I think, at the most, I'm guilty of a redundancy. Baseball is a sport,
just as 'soccer' is a sport. Each of them are what they are: individual
kind sports, and obviously unique (unlike Squash, Raquetball, Tennis, etc).
So I'm redundant.
So I'm redundant.
<snip>
mario in victoria
--
again
McDuck
<ron...@gmail.com> wrote:
>
>It's generally a bad idea to bunt with a position player
>who is neither fast enough to have a decent chance of
>reaching and isn't a good bunter.
Yes. And even in the case of a poor hittter who is a good bunter and
is also fast, the value of the bunt depends in part on the game
situation. Needing one run late makes the bunt attractive, especially
if the pitcher is dominant. Early in a game, down 7 runs, the bunt
becomes less attractive. Just to give the extremes <g>.
I really do wonder why managers ever want a poor bunter to bunt ---
unless he is a really terrible hitter, like the typical pitcher in the
NL.
"Is Dimaggio a good bunter?"
"I don't know and I hope I never find out."
PaulH
>
> I know you can get the base/out state tables at billjamesonline.net
>
> Yeah, pay site, but it's dirt cheap.
>
> Here's an old base/out table -- values won't be too different today.
>
>
> Bases Outs
> 0 1 2
> empty 0.57 0.31 0.12
> 1st 0.97 0.60 0.27
> 2nd 1.18 0.73 0.33
> 1st, 2nd 1.63 1.01 0.48
> 3rd 1.52 1.00 0.41
> 1st, 3rd 1.92 1.24 0.52
> 2nd, 3rd 2.05 1.50 0.64
> 1st, 2nd, 3rd 2.54 1.70 0.82
>
> In case it's not clear these are the average number of
> runs scored in an inning from the various base out
> situations (innings where 3 outs weren't recorded not
> included)
>
Empirically-based run expectancy state tables for every year going back
to 1954 are also available free of charge on the Baseball Prospectus site:
<http://www.baseballprospectus.com/statistics/sortable/index.php?cid=204022>
--Paul
OceanView
news:74rhk3F...@mid.individual.net:
> OceanView <F...@chance.org> wrote:
>> I always thought an interesting stat would be "base runners
>> advanced." A
>
> That's about as stupid as RBI, for the same reasons that RBI is
> stupid.
>
No, but like most baseball stats, it depends. A lot of good hitters hit
better when a real threat is hitting behind them. Pitchers pitch better
with a better defense behind them. It's mosr clear-cu and measurable than
many other sports. What constitues a quality shot in hockey or half-a-
tackle in football?
--
-------
I am short enough and ugly enough to succeed on my own. - Woody Allen
OceanView
news:fvscu4lh0ckp06rp4...@4ax.com:
> Let us say that a single has the value of 1. If this is the case,
> what would the true value of a double, triple, homerun be? I am just
> annoyed lately of slugging percentage, it seems that a triple is very
> similar to a double, is it really worth 150% of a double? Is it
> really a 1:2:3:4 ratio for total bases, if not what do you think is
> more represantive, maybe we could come up with a better version of
> slugging percentage?
>
Bottom line: 72 percent of all statistics are irrelevant, 81 percent of
which are made-up on the spot, and 99 percent of baseball fans disagree 75
percent of the time.
Jeff Schwab
> 72 percent of all statistics are irrelevant, 81 percent of
> which are made-up on the spot, and 99 percent of baseball fans disagree 75
> percent of the time.
Research causes cancer in lab rats.
Christina
>Christina <bi...@ix.netcom.com> wrote in
>news:fvscu4lh0ckp06rp4...@4ax.com:
>
>> Let us say that a single has the value of 1. If this is the case,
>> what would the true value of a double, triple, homerun be? I am just
>> annoyed lately of slugging percentage, it seems that a triple is very
>> similar to a double, is it really worth 150% of a double? Is it
>> really a 1:2:3:4 ratio for total bases, if not what do you think is
>> more represantive, maybe we could come up with a better version of
>> slugging percentage?
>>
>
>Bottom line: 72 percent of all statistics are irrelevant, 81 percent of
>which are made-up on the spot, and 99 percent of baseball fans disagree 75
>percent of the time.
Ocean is this really true or a joke?
Dano
Of course it's true.
Throws like Mary
90% of statistics are true.
bi...@ix.netcom.com
>
> - Show quoted text -
But throws if a statistic is the numbers of things that happened isn't
it always true, it is just a question of how it is to be interpreted
or if it has value? For instance Throws let us say I make up a
statistic walk per triple ratio, a player has 50 walks and 5 triples,
I give him a 10.0. This is true, he has 10 walks per triple, it is a
silly satisitic but it is true none the less if that makes sesne?
wayback1918
>
> - Show quoted text -
Jason Bay is leading the AL in BB/3B with 13.0.......One ahead of
Marco Scutaro
Monster Zero
> Christina <bi...@ix.netcom.com> wrote innews:fvscu4lh0ckp06rp4...@4ax.com:
41% of people who say stats are 72% irrelevant are 19.7% more likely
to be killed or severely injured by 1% milk.
The Gnorkmeister
Citation?
Throws like Mary
Generally speaking, yes. You are referring to the broader definition
of a statistic: numerical data; observable data. Observation
statistics are generally true on their face, provided no one questions
the methods or the data itself (someone might question Bond's HR
totals, etc.) Hence my joke: 90% of statistics are true. Most
"statistics" disputed in this group are actually disputes over
inferences, not data.
The narrow definition of statistics is an inference made about a
population based on a sample population. (This gives rise to the
frequent refrain about "sample size." It is a fallacy if applied to
the broad definition above, absent some form of inductive inference,
such as a prediction.)
Does that make sense?
Pearly Soames
>
> A lot of good hitters hit
> better when a real threat is hitting behind them. Pitchers pitch better
> with a better defense behind them. It's mosr clear-cu and measurable than
> many other sports.
>
I don't think either of those two statements is clear-cut. If they've
been measured, I haven't seen the data to support them.
--
"I'll keep it short and sweet. Family, religion, friendship ... these
are the three demons you must slay if you wish to succeed in business."
Ron Johnson
> OceanView wrote:
>
> > A lot of good hitters hit
> > better when a real threat is hitting behind them. Pitchers pitch better
> > with a better defense behind them. It's mosr clear-cu and measurable than
> > many other sports.
>
> I don't think either of those two statements is clear-cut. If they've
> been measured, I haven't seen the data to support them.
This comes up so often that there's actually an FAQ on the matter.
http://www.baseball1.com/faqs/protection-faq.html
TO sum up, clear evidence of what David calls weak protection.
IBBs go up for an unprotected (by any definition) hitter.
No evidence of strong protection
Ron Johnson
>
> I really do wonder why managers ever want a poor bunter to bunt ---
> unless he is a really terrible hitter, like the typical pitcher in the
> NL.
I think it's as simple as the widely held belief that if you're in
the majors you should be able to bunt.
Ron Johnson
>
> Empirically-based run expectancy state tables for every year going back
> to 1954 are also available free of charge on the Baseball Prospectus site:
>
> <http://www.baseballprospectus.com/statistics/sortable/index.php?cid=2...>
Thanks Paul. Knew they were available, just couldn't remember where.
Dano
Widely held by WHO? Some knuckleheaded fans in a newsgroup or message
board? Judging by the widespread non-use of the bunt in todays
game...especially the AL...the widespread trend would seem to be AWAY from
bunting as a strategy.
Of all the skill sets that will get a player to the bigs...where do you
suppose bunting skills rank? You really think a lot of guys are bypassed
over THAT skillset? <g>
Dano
Widely held by WHO? Some knuckleheaded fans in a newsgroup or message
Ron Johnson
> Ron Johnson wrote:
> > On Apr 18, 2:22 am, McDuck <wallyDELETEMEMcD...@comcast.net> wrote:
>
> >> I really do wonder why managers ever want a poor bunter to bunt ---
> >> unless he is a really terrible hitter, like the typical pitcher in
> >> the NL.
>
> > I think it's as simple as the widely held belief that if you're in
> > the majors you should be able to bunt.
>
> Widely held by WHO? Some knuckleheaded fans in a newsgroup or message
> board?
I used to collect quotes of this nature. They're substantially
less common these days, but even so you still hear some
new managers talk about how everybody has to get on board
the little ball train.
> Judging by the widespread non-use of the bunt in todays
> game...especially the AL...the widespread trend would seem to be AWAY from
> bunting as a strategy.
There are managers in today's game who just don't bunt.
That's driving the overall numbers of bunts down.
But there are plenty of managers whose tactical decisions
are the same as the guys managing in 1972.
And they're not wrong ... provided ...
>
> Of all the skill sets that will get a player to the bigs...where do you
> suppose bunting skills rank?
Somewhere around zero. Anybody stupid enough to select on ability
to bunt gets what they deserve.
But there is a mindset that's not uncommon that says essentially
any major league player who wants to can get the bunt down.
And ignores the reality that bunting against major league
pitching is *hard*.
I only have detailed stats on bunt attempts for one year,
and in that year exactly one player (Tony Gwynn) got the
bunt down (at least a base for an out) on more
than half their attempts (Gwynn was over 75%). Doesn't
mean a bad bunt. This also includes balls fouled off.
But what this suggests is an awful lot of flailing
away after going 0-2 even among guys who bunt at
least semi-regularly.
> You really think a lot of guys are bypassed
> over THAT skillset? <g>
That would be my point too. One of the clearest changes in
the game starting in the late 80s is that the fast, low
power switch-hitters have been largely driven from the game.
(The number of switch-hitters is dramatically down from their
peak, and most of them are what Doug Drinen called "slappin
swifties")
An awful lot of good bunt candidates went with them.
John W Gintell
Still, you'd think that occasional bunting practice for everyone would be useful.
Dano
It would come at the expense of extra BP. So I disagree. For everyone that
is. Again...in the NL, I'm sure all pitchers go through bunting drills.
And I suppose for your quicker and lighter hitting players it wouldn't be a
bad idea either. Do you know for a fact that none of those guys ever work
on their bunting? I'd be very surprised if that's the case.