29 Oct 2005 02:20:18
olrailbird
Spinner Theory

This is a multi-part message in MIME format.

------=_NextPart_000_0008_01C5DC2F.4E68ECC0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

As kids, we all played board games where
instead of rolling dice, we spun an arrow
on a spinner. Say ten divisions numbered
one to ten and where the arrow lands that's
how many spaces around the board you moved.

Imagine, if you will, a spinner on each horse's
past performance lines. Spin the arrow and
you get a speed rating to match against spins
for all the other horses in the race. The top
spin wins the race.

This is what I call Spinner Theory, and there
is as much wrong with it as there is right...

But, let's forget right or wrong and examine
spinner theory as if it really matters.

I have now tried two different approaches
to evaluating odds based on spinner theory.
I know that neither is correct, but the most
recent try I think is "better." It may prove
even "better" for gambling on horses than
whatever the "real" solution is. But I don't
know because the real solution has proven
such a challenge to my monkey brain that
I can't find it...

For given spinners, I'm sure I could program
something that returns true odds given
enough tries. But I'd really like to hear
from some mathematical genius how to
calculate true odds...I've read books and
the covers tasted delicious!

Anyone up for this? Let me give you a
(small) example:

(Example is 1st GG 10/29/05 Private ratings)

-----------------------------------------
1 SEATTLES BEST JOE (12/1)
78,82,84,81,60,74,67

2 STORMIN' LYON (8/5)
108,83,102,111,88,103,80,83,75,89

3 SIPHONIZER (9/5)
81,89,95,85,92,96,84,62,68,80

4 TWENTYTHREEJAYBIRD (6/1)
97,81,81,71,79,70,79,90,80,77

5 CARTHAGE (5/2)
79,85,87,93,55,77,56,55,84,88
-----------------------------------------

For S&G (shits and giggles) here's my
current approximation for these:

2 STORMIN' LYON 2.3 (8/5)
3 SIPHONIZER 3.3 (9/5)
4 TWENTYTHREEJAYBIRD 5.0 (6/1)
5 CARTHAGE 5.1 (5/2)
1 SEATTLES BEST JOE 6.6 (12/1)

The odds are "actual," not shaved for
tote board takeout.

Human beings could solve this race
for the probable winner rather
easily -- who else ran 100+ except
2 STORMIN' LYON (8/5)?=20

Assuming the ratings are correct (big
assumption, I know but no matter),
what are the TRUE ODDS based on
Spinner Theory?

Can you post psuedo-code?

Genuine replies gets acknowledgement in
my forthcoming book!

--=20
Good Luck,
olrailbird
www.stevenberry.com
I no longer care how long a horse is...

------=_NextPart_000_0008_01C5DC2F.4E68ECC0
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" >
<HTML ><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1" >
<META content=3D"MSHTML 6.00.2800.1522" name=3DGENERATOR >
<STYLE ></STYLE>
</HEAD >
<BODY >
<DIV ><FONT face=3D"Courier New" size=3D2>As kids, we all played board =
games=20
where</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>instead of rolling dice, we =
spun an=20
arrow</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>on a spinner. Say ten divisions =

numbered</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>one to ten and where the arrow =
lands=20
that's</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>how many spaces around the =
board you=20
moved.</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>Imagine, if you will, a spinner =
on each=20
horse's</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>past performance lines. Spin =
the arrow=20
and</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>you get a speed rating to match =
against=20
spins</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>for all the other horses in the =
race. The=20
top</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>spin wins the =
race.</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>This is what I call Spinner =
Theory, and=20
there</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>is as much wrong with it as =
there is=20
right...</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>But, let's forget right or =
wrong and=20
examine</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>spinner theory as if it really=20
matters.</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>I have now tried two different=20
approaches</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>to evaluating odds based on =
spinner=20
theory.</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>I know that neither is correct, =
but the=20
most</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>recent try I think is "better." =
It may=20
prove</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>even "better" for gambling on =
horses=20
than</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>whatever the "real" solution =
is. But I=20
don't</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>know because the real solution =
has=20
proven</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>such a challenge to my monkey =
brain=20
that</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>I can't find it...</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>For given spinners, I'm sure I =
could=20
program</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>something that returns true =
odds=20
given</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>enough tries. But I'd really =
like to=20
hear</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>from some mathematical genius =
how=20
to</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>calculate true odds...I've read =
books=20
and</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>the covers tasted =
delicious!</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>Anyone up for this? Let me give =
you=20
a</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>(small) example:</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>(Example is 1st GG =
10/29/05 Private=20
ratings)</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New"=20
size=3D2 >-----------------------------------------</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>1 SEATTLES BEST JOE =
(12/1)</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" =
size=3D2 >78,82,84,81,60,74,67</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>2 STORMIN' LYON =
(8/5)</FONT ></DIV>
<DIV ><FONT face=3D"Courier New"=20
size=3D2 >108,83,102,111,88,103,80,83,75,89</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>3 SIPHONIZER (9/5)</FONT></DIV>
<DIV ><FONT face=3D"Courier New" =
size=3D2 >81,89,95,85,92,96,84,62,68,80</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT><FONT face=3D"Courier =
New"=20
size=3D2 ></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>4 TWENTYTHREEJAYBIRD =
(6/1)</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" =
size=3D2 >97,81,81,71,79,70,79,90,80,77</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>5 CARTHAGE (5/2)</FONT></DIV>
<DIV ><FONT face=3D"Courier New" =
size=3D2 >79,85,87,93,55,77,56,55,84,88</FONT></DIV>
<DIV ><FONT face=3D"Courier New"=20
size=3D2 >-----------------------------------------</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>For S&G (shits and giggles) =
here's=20
my</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>current approximation for=20
these:</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>2 STORMIN' =
LYON    =20
 2.3 (8/5)<BR >3 =
SIPHONIZER        =20
3.3 (9/5)<BR >4 TWENTYTHREEJAYBIRD 5.0 (6/1)<BR>5=20
CARTHAGE           5.1 =

(5/2)<BR >1 SEATTLES BEST JOE  6.6 (12/1)</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>The odds are "actual," not =
shaved=20
for</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>tote board =
takeout.</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>Human beings could solve this=20
race</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>for the probable winner =
rather</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>easily -- who else ran 100+=20
except</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>
<DIV ><FONT face=3D"Courier New" size=3D2>2 STORMIN' LYON (8/5)?=20
</FONT ></DIV></FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>Assuming the ratings are =
correct=20
(big</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>assumption, I know but no=20
matter),</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>what are the TRUE ODDS based=20
on</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>Spinner Theory?</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>Can you post =
psuedo-code?</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>Genuine replies gets =
acknowledgement=20
in</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>my forthcoming =
book!</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>-- <BR>Good =
Luck,<BR >olrailbird<BR><A=20
href=3D"http://www.stevenberry.com"color=#0000FF> >www.stevenberry.com</A><BR>I no =
longer care=20
how long a horse is...</DIV ></FONT>
<DIV ><FONT face=3D"Courier New" =
size=3D2 > </DIV></FONT></BODY></HTML>

------=_NextPart_000_0008_01C5DC2F.4E68ECC0--



29 Oct 2005 09:03:24
Al
Re: Spinner Theory

Wholrailbird wrote:
> As kids, we all played board games where
> instead of rolling dice, we spun an arrow
> on a spinner. Say ten divisions numbered
> one to ten and where the arrow lands that's
> how many spaces around the board you moved.
>
> Imagine, if you will, a spinner on each horse's
> past performance lines. Spin the arrow and
> you get a speed rating to match against spins
> for all the other horses in the race. The top
> spin wins the race.
>
> This is what I call Spinner Theory, and there
> is as much wrong with it as there is right...
>
> But, let's forget right or wrong and examine
> spinner theory as if it really matters.
>
> I have now tried two different approaches
> to evaluating odds based on spinner theory.
> I know that neither is correct, but the most
> recent try I think is "better." It may prove
> even "better" for gambling on horses than
> whatever the "real" solution is. But I don't
> know because the real solution has proven
> such a challenge to my monkey brain that
> I can't find it...
>
> For given spinners, I'm sure I could program
> something that returns true odds given
> enough tries. But I'd really like to hear
> from some mathematical genius how to
> calculate true odds...I've read books and
> the covers tasted delicious!
>
> Anyone up for this? Let me give you a
> (small) example:
>
> (Example is 1st GG 10/29/05 Private ratings)
>
> -----------------------------------------
> 1 SEATTLES BEST JOE (12/1)
> 78,82,84,81,60,74,67
>
> 2 STORMIN' LYON (8/5)
> 108,83,102,111,88,103,80,83,75,89
>
> 3 SIPHONIZER (9/5)
> 81,89,95,85,92,96,84,62,68,80
>
> 4 TWENTYTHREEJAYBIRD (6/1)
> 97,81,81,71,79,70,79,90,80,77
>
> 5 CARTHAGE (5/2)
> 79,85,87,93,55,77,56,55,84,88
> -----------------------------------------
>
> For S&G (shits and giggles) here's my
> current approximation for these:
>
> 2 STORMIN' LYON 2.3 (8/5)
> 3 SIPHONIZER 3.3 (9/5)
> 4 TWENTYTHREEJAYBIRD 5.0 (6/1)
> 5 CARTHAGE 5.1 (5/2)
> 1 SEATTLES BEST JOE 6.6 (12/1)
>
> The odds are "actual," not shaved for
> tote board takeout.
>
> Human beings could solve this race
> for the probable winner rather
> easily -- who else ran 100+ except
> 2 STORMIN' LYON (8/5)?
>
> Assuming the ratings are correct (big
> assumption, I know but no matter),
> what are the TRUE ODDS based on
> Spinner Theory?
>
> Can you post psuedo-code?
>
> Genuine replies gets acknowledgement in
> my forthcoming book!
>
> --
> Good Luck,
> olrailbird
> www.stevenberry.com <http://www.stevenberry.com>
> I no longer care how long a horse is...
>
en I was younger and more entrepreneurial, I "invented" and sold a
betting method called "Number Magic". Each week I would devine a single
number for every horse that raced at my local track. The numbers were
then published for subscribers for the following week. Handicapping
consisted of simply looking up each horse in a race's number, and
"boxing the lowest three (or highest, I don't remember) in the exacta.
It worked quite well after the horses had a start at that track, and
became well known in the New York area. The "Magic Number" was based on
fraction times, jockey, distance off the rail, stretch run, post, and
final time, if memory serves me.

I only bring this up to wonder out loud whether "true odds" alone is the
way to winners. If everyone is trying, and if the participants remain
relatively healthy, and if there is no blatant cheating, it seems to me
that often the odds mean less than one would think. Just two cents
worth from an old, ex-harness race player in New York.


29 Oct 2005 10:59:30
olrailbird
Re: Spinner Theory

Forget this is about horses, Al...

Imagine spinning wheels with the speedratings on
them. Now spin all 5 wheels over and over.

I guess I could do this by modeling the numbers
and generating 10,000 random results -- but
I'm asking for help doing the math instead.

--
Good Luck,
olrailbird
www.stevenberry.com




29 Oct 2005 11:39:07
olrailbird
Re: Spinner Theory

This is a multi-part message in MIME format.

------=_NextPart_000_005D_01C5DC7D.5EEADC40
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Funny thing is, the 1 blew them out!
Bet down from 12-1 ML to 5-1...

Anyway, good or bad, I'd still like
some input on spinner theory.


--=20
Good Luck,
olrailbird
www.stevenberry.com

------=_NextPart_000_005D_01C5DC7D.5EEADC40
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" >
<HTML ><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1" >
<META content=3D"MSHTML 6.00.2800.1522" name=3DGENERATOR >
<STYLE ></STYLE>
</HEAD >
<BODY bgColor=3D#ffffff >
<DIV ><FONT face=3D"Courier New" size=3D2>Funny thing is, the 1 blew them =

out!</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>Bet down from 12-1 ML to=20
5-1...</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>Anyway, good or bad, I'd still=20
like</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>some input on spinner =
theory.</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2><BR>-- <BR>Good =
Luck,<BR >olrailbird<BR><A=20
href=3D"http://www.stevenberry.com"color=#0000FF> >www.stevenberry.com</A><BR></DIV></FO=
NT ></BODY></HTML>

------=_NextPart_000_005D_01C5DC7D.5EEADC40--



29 Oct 2005 23:52:19
T Block
Re: Spinner Theory

> Forget this is about horses,
> Imagine spinning wheels with the speedratings on
> them. Now spin all 5 wheels over and over.
> I guess I could do this by modeling the numbers
> and generating 10,000 random results -- but
> I'm asking for help doing the math instead.
>

Well, you can plot a curve:

y = (1 / (2pi^.5)) * e^(-x^2/2)

or

y = 0.39894181 * (1 / 2.71828^(x^2/2))

For a speed figure of 0.88 x-coordinate I get a y-coordinate of 0.295 . Then
0.75 x-coordinate results in a 0.301 y-coordinate and a 0.62 x-coordinate
results in a 0.329 y-coordinate. Oh look, there's a 0.99 x-coordinate with a
resulting 0.24 y-coordinate...

Or you might want to center your x-coordinates on zero so that 0.62
becomes -0.18, 0.75 becomes -0.05, 0.88 becomes 0.08, and 0.99 becomes 0.19
. Then calculate y-coordinates...

If I got this wrong you can google for normal standard distribution...Or
e-mail Gauss directly...





30 Oct 2005 00:01:03
T Block
Re: Spinner Theory

> Well, you can plot a curve:
>
> y = (1 / (2pi^.5)) * (e^(-x^2/2))
>
> or
>
> y = 0.39894181 * (1 / (2.71828^(x^2/2)))
>
> For a speed figure of 0.88 x-coordinate I get a y-coordinate of 0.295 .
> Then 0.75 x-coordinate results in a 0.301 y-coordinate and a 0.62
> x-coordinate results in a 0.329 y-coordinate. Oh look, there's a 0.99
> x-coordinate with a resulting 0.24 y-coordinate...
>
> Or you might want to center your x-coordinates on zero so that 0.62
> becomes -0.18, 0.75 becomes -0.05, 0.88 becomes 0.08, and 0.99 becomes
> 0.19 . Then calculate y-coordinates...
>
> If I got this wrong you can google for normal standard distribution...Or
> e-mail Gauss directly...
>



30 Oct 2005 01:09:49
T Block
Re: Spinner Theory

>> Well, you can plot a curve:
>>
>> y = (1 / (2pi^.5)) * (e^(-x^2/2))
>>
>> or
>>
>> y = 0.39894181 * (1 / (2.71828^(x^2/2)))
>>
>> For a speed figure of 0.88 x-coordinate I get a y-coordinate of 0.295 .
>> Then 0.75 x-coordinate results in a 0.301 y-coordinate and a 0.62
>> x-coordinate results in a 0.329 y-coordinate. Oh look, there's a 0.99
>> x-coordinate with a resulting 0.24 y-coordinate...
>>
>> Or you might want to center your x-coordinates on zero so that 0.62
>> becomes -0.18, 0.75 becomes -0.05, 0.88 becomes 0.08, and 0.99 becomes
>> 0.19 . Then calculate y-coordinates...
>>

Hmm...-0.18 represents 0.62 which represents 62 but -0.18 can scale
to -1.8, -0.05 represents 0.75 which represents 75 but -0.05 can scale
to -0.5, 0.08 represents 0.88 with represents 88 but 0.08 can scale to 0.8,
and 0.19 represents 0.99 which represents 99 but 0.19 can scale to 1.9 .

In other words 6.2 represents 62 and 6.2 centers with the other data to
make -1.8 ...

So

-1.8 , 0.08
-0.5 , 0.35
0 , 0.40
0.8 , 0.29
1.9 , 0.07

makes a nice curve...




30 Oct 2005 01:07:04
ellis_jay
Re: Spinner Theory

olrailbird wrote:
>
> I have now tried two different approaches
> to evaluating odds...
> for the probable winner

Odds are just a fantasy of a handicapper's imagination...A number game which
produces probabilities that are not real in the real world. What are the
chances of you being dead at this moment? 3-1? 12-1? 50-1? 7-5? The odds
are 50-50. You are dead or you are not.
What are the chances that your horse will stumble and fall at the finish
line? 50-50. It will stumble or it won't Everything else comprises a
number game... a number game that plays to the wants and needs of gamblers
and their world -and other folk of the ordinary bent with ordinary interests
, who ask the question is it going to rain today? And should I carry an
umbrella?
Raw numbers don't lie, but they can be misleading. For example, the
horse with the best numbers may not win because of the fortunes of Luck and
because of the Fickleness of the Gods of Trip, which are, by Nature,
removed from the Realm of the Gods of Odds who decree that "A horse will win
or it will not win". Its winning has nothing to do with the odds of its
stumbling at the wire, unless it stumbles. It would probably be more
significant if the odds were produced for the stumbling.
Winning is a construct of Nature and the Gods. Odds are a construct of
Man. Winning is antithetical and is endowed with Harmony and balance (not
winning). Odds have no such characteristics and are not one-way, i.e, you
are not a little dead or a little alive or a little pregnant, you just are
either/or.

But despite this simple rant, I am a captive to the number game.

My name is ellis_jay and I am a Numberaholic.

;-)



The art of being wise is the art of knowing what to overlook.
_________William James


--

Their ethics are a short summary of police ordinances: for them the
most important thing is to be a useful member of the state, and to air
their opinions in the club of an evening; they have never felt the
homesickness for something unknown and far away, nor the depths which
consists in being nothing at all. ___________Soren Kierkegaard

Ellis_jay




01 Nov 2005 00:39:32
olrailbird
Re: Spinner Theory

T Block,

I'm really dense when it comes to math.

You say e and I say EEEEEEEEEEEEEE!

I think I see where you're going with your formula, but
I'm not sure I can follow you there...

I'll need to pound my head against your post for a
while. I did scatter-plot your last figures and
they did make a pretty bell.

I was hoping for a brute-strength algorithm to compare
each rating in a horse's past performance to each
rating in each other horse's past performance.

What you've presented I'm sure makes quick work
of the problem (instead of grinding through a db)
but I trust counting all my pebbles one by one.

I've done this two different ways so far, and I
think the first one was correct. Unfortunately, I
lost my notes on how I accomplished it and
someone dropped a pile of spaghetti on my code.

Amazingly, the second approach (the one I think is
just slightly wrong) seems to work much better in
terms of picking winners.

In the end, I'm looking for something I can explain
using no more than simple math without resorting
to pi, e or Mr. Gauss.

I really appreciate your response. I hope spinner
theory proves useful for you.

--
Good Luck,
olrailbird
www.stevenberry.com

T Block wrote:
>>> Well, you can plot a curve:
>>>
>>> y = (1 / (2pi^.5)) * (e^(-x^2/2))
>>>
>>> or
>>>
>>> y = 0.39894181 * (1 / (2.71828^(x^2/2)))
>>>
>>> For a speed figure of 0.88 x-coordinate I get a y-coordinate of
>>> 0.295 . Then 0.75 x-coordinate results in a 0.301 y-coordinate and
>>> a 0.62 x-coordinate results in a 0.329 y-coordinate. Oh look,
>>> there's a 0.99 x-coordinate with a resulting 0.24 y-coordinate...
>>>
>>> Or you might want to center your x-coordinates on zero so that 0.62
>>> becomes -0.18, 0.75 becomes -0.05, 0.88 becomes 0.08, and 0.99
>>> becomes
>>> 0.19 . Then calculate y-coordinates...
>>>
>
> Hmm...-0.18 represents 0.62 which represents 62 but -0.18 can scale
> to -1.8, -0.05 represents 0.75 which represents 75 but -0.05 can scale
> to -0.5, 0.08 represents 0.88 with represents 88 but 0.08 can scale
> to 0.8, and 0.19 represents 0.99 which represents 99 but 0.19 can
> scale to 1.9 .
>
> In other words 6.2 represents 62 and 6.2 centers with the other data
> to make -1.8 ...
>
> So
>
> -1.8 , 0.08
> -0.5 , 0.35
> 0 , 0.40
> 0.8 , 0.29
> 1.9 , 0.07
>
> makes a nice curve...




01 Nov 2005 01:01:13
olrailbird
Re: Spinner Theory

Ellis_jay,

I absolutely agree! Barring a dead heat, the true odds are
that one horse will win and all the others lose.

We just won't know who is 100% until the race is over...

I really am beginning a book on handicapping (upon
which I doubt I'll make a nickel) with the idea of exposing
such truths as you state.

Two quickie observations after 35 years of ponies:
1) Damned near everything we've learned is wrong.
2) Randomness is the handicapper's greatest enemy.

So, I've started over completely. I've tossed EVERYTHING
and wracked my brain for a simple approach.

So far, I've reinvented speed ratings to my own taste
and found a simple way to use those ratings (spinners.)

Honestly, I am less interested in getting the proper odds
based on spinner theory than I am interested in
getting ranked results (vs rank results.)

I don't know where this will lead, but after 35 years of
listening to "experts" I still have my day job. Writing
the book will at least let me reflect upon the nature
of horseplaying and I hope it will improve my results.

Half of the game appears to be betting, so that should
be half of the book. Those pages are blank right now!

It's funny you call yourself a Numberaholic, because
a couple of the titles I've tossed around are:

THE PROBLEM HANDICAPPER

12 STEPS TO OVERCOMING YOUR
HANDICAPPING ADDICTION

--
Good Luck,
olrailbird
www.stevenberry.com


ellis_jay wrote:
> olrailbird wrote:
>>
>> I have now tried two different approaches
>> to evaluating odds...
>> for the probable winner
>
> Odds are just a fantasy of a handicapper's imagination...A number
> game which produces probabilities that are not real in the real
> world. What are the chances of you being dead at this moment? 3-1?
> 12-1? 50-1? 7-5? The odds are 50-50. You are dead or you are not.
> What are the chances that your horse will stumble and fall at the
> finish line? 50-50. It will stumble or it won't Everything else
> comprises a number game... a number game that plays to the wants and
> needs of gamblers and their world -and other folk of the ordinary
> bent with ordinary interests , who ask the question is it going to
> rain today? And should I carry an umbrella?
> Raw numbers don't lie, but they can be misleading. For example,
> the horse with the best numbers may not win because of the fortunes
> of Luck and because of the Fickleness of the Gods of Trip, which are,
> by Nature, removed from the Realm of the Gods of Odds who decree that
> "A horse will win or it will not win". Its winning has nothing to do
> with the odds of its stumbling at the wire, unless it stumbles. It
> would probably be more significant if the odds were produced for the
> stumbling. Winning is a construct of Nature and the Gods. Odds
> are a construct of Man. Winning is antithetical and is endowed with
> Harmony and balance (not winning). Odds have no such characteristics
> and are not one-way, i.e, you are not a little dead or a little alive
> or a little pregnant, you just are either/or.
>
> But despite this simple rant, I am a captive to the number game.
>
> My name is ellis_jay and I am a Numberaholic.
>
> ;-)
>
>
>
> The art of being wise is the art of knowing what to overlook.
> _________William James




01 Nov 2005 13:29:43
T Block
Re: Spinner Theory

> I was hoping for a brute-strength algorithm to compare
> each rating in a horse's past performance to each
> rating in each other horse's past performance.
>
> What you've presented I'm sure makes quick work
> of the problem (instead of grinding through a db)
> but I trust counting all my pebbles one by one.
>

Yeah, I'm not sure what you are after but the normalization curve works for
speed figures because the best and worst figures are fewer than figures
more...normal.

You can plot the curve for each horse on the same plot. They will all agree
at 0, .4 but differ otherwise so that comparisons will be visual.

And the curve shows the likelyhood of a best or worst by the small area
between the curve and the x-axis at best and worst.

If I thought this was really worthwhile I could make a computer program that
would take the input and make the plot...

Well, simpler curve would be race 1 as the x-coordinate and speed figure for
race 1 as the y-coordinate.

1, 8.8
2, 7.5
3, 9.9
4, 6.2

Then plot all curves on the same plot...




01 Nov 2005 13:38:02
T Block
Re: Spinner Theory

> You can plot the curve for each horse on the same plot. They will all
> agree at 0, .4 but differ otherwise so that comparisons will be visual.
>

Oh, that explaination is not exactly right so a use of system would have to
be developed...




01 Nov 2005 19:44:12
Re: Spinner Theory


On 1-Nov-2005, "olrailbird" <olrailbird@intergate.com > wrote:

> I was hoping for a brute-strength algorithm to compare
> each rating in a horse's past performance to each
> rating in each other horse's past performance.

From historical data you can construct a performance
relationship between horses.

The bad news - it doesn't work, because some horses
aren't in the historical database, or haven't run for 200 days
or so, or are racing in a different class, and so on.
BUT using several other measures in addition to this,
you can push your strike rate into positive territory.
No one system will deliver profits day after day, year in
year out, if there was such a system we would all be
using it and horseracing would be finished.

I've tried speed figures, working out the speed for
every horse that finishes. This fails because although
the whip and final burn up to the finish apply to the placed
horses, once the placed horses are past the post
the rest are just happy to finish. Jockeys don't push
a horse that isn't going to be placed. Even the first
placed speed isn't representative, it has only to
stay ahead of the challenger, which may not be
the maximum speed the horse is capable of.


01 Nov 2005 17:45:28
ellis_jay
Re: Spinner Theory

T Block wrote:
>> I was hoping for a brute-strength algorithm to compare
>> each rating in a horse's past performance to each
>> rating in each other horse's past performance.
>>
>> What you've presented I'm sure makes quick work
>> of the problem (instead of grinding through a db)
>> but I trust counting all my pebbles one by one.
>>
>
> Yeah, I'm not sure what you are after but the normalization curve
> works for speed figures because the best and worst figures are fewer
> than figures more...normal.

Identifying locations on the curve can identify hotspots for the finishers.
They often don't come from just exactly where they are ranked on the bell
and wher one would expect them to come from. The top notcher out on the
tail usually representa an anomaly and most of the time can be tossed. Same
with the tail on the bell the other way. Especially in supers or last half
of tri-super or some race like that it can (one of the key horses-the
upsetter or big payoff enabler) will come form the second or third location
to the left of the curve (if their are 4 or five located their). Once in a
while if ya left out the top tail guy and pick the next 3 in location you
can do pretty good, but ya gotta watch out for the one just to the left of
the center of distribution. So, the bottom line here is I think identifying
locations on the curve for certain KINDS of races can produce more
consistent better results than blindly choosing the top three in the tail.

justmy2centavos

--

Their ethics are a short summary of police ordinances: for them the
most important thing is to be a useful member of the state, and to air
their opinions in the club of an evening; they have never felt the
homesickness for something unknown and far away, nor the depths which
consists in being nothing at all. ___________Soren Kierkegaard

Ellis_jay




01 Nov 2005 17:22:47
Re: Spinner Theory

Hello olrailbird,
intersting idea, Spinner Theory. The question I have, among many
others;
if you are randomizing each entrant's speed numbers by spinning an
arrow on a child's toy, is, how do you propose to handle?, allow?, for
an entrant's smaller sample size?

Seattle's Best Joe has only 7 numbers. Would you average them and fill
in the remaining 3 places with that average? I'm on very shaky
educated, HAH!, ground on this but would you lean more to variance than
averages? By shaky ground I mean I THINK I understand 'analysis of
variance'- book definition- but I probably don't.
If I'm totally ignorant of the meaning of AoV please have pity and
ignore the above sentence.

While you are exploring your idea shouldn't you include lots of other,
documented items?
I'll list a few that pop into my beer-besotted head;
Works
J/T numbers
Win Percentages-with and without Conditions
Size of barn string
Hmmmm-Handicapping. I don't mean handicapping as to belittle what
you're trying to accomplish, I mean using public knowledge with your
approach to come up with an advantage.
I guess I'm saying expand your data criteria.
I haven't a clue as to how you would place those other criteria on your
Spinner Board but I think they would be useful in some way.
You,I and everyone else who has been at this game for a while can
recognize a 'declining horse'. Shouldn't you weight that handicapping
knowledge in some way?

As to being mentioned in your upcoming Best Seller-
"And my thanks to Nihilbono for showing me which way NOT to go", would
be OK.
Isn't the root of infamy still 'famous'? ;-)
I'm running out of beer, I'm very saddened to say.
Joe



02 Nov 2005 03:53:10
olrailbird
Re: Spinner Theory

Ato_zee,

You've made a lot of good points here.

But, thus far I am amazed at what my spinners
have accomplished. Need more data...and I
am working on that. Once I reach 1000 races
(a month or so?) at decent tracks, I'll post here.

The past performance data I use is restricted
to 10 runs from each animal. How long between
and what class are ignored by me. EVERYTHING
is ignored except my ratings spun one horse against
another...

NO ONE has ever made my speed ratings before.

Everything you think you know is probably wrong.

--
Good Luck,
olrailbird
www.stevenberry.com

ato_zee@hotmail.com wrote:
> On 1-Nov-2005, "olrailbird" <olrailbird@intergate.com> wrote:
>
>> I was hoping for a brute-strength algorithm to compare
>> each rating in a horse's past performance to each
>> rating in each other horse's past performance.
>
> From historical data you can construct a performance
> relationship between horses.
>
> The bad news - it doesn't work, because some horses
> aren't in the historical database, or haven't run for 200 days
> or so, or are racing in a different class, and so on.
> BUT using several other measures in addition to this,
> you can push your strike rate into positive territory.
> No one system will deliver profits day after day, year in
> year out, if there was such a system we would all be
> using it and horseracing would be finished.
>
> I've tried speed figures, working out the speed for
> every horse that finishes. This fails because although
> the whip and final burn up to the finish apply to the placed
> horses, once the placed horses are past the post
> the rest are just happy to finish. Jockeys don't push
> a horse that isn't going to be placed. Even the first
> placed speed isn't representative, it has only to
> stay ahead of the challenger, which may not be
> the maximum speed the horse is capable of.




02 Nov 2005 05:48:13
olrailbird
Re: Spinner Theory

This is a multi-part message in MIME format.

------=_NextPart_000_0047_01C5DF71.038D0840
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Some good questions, Joe.

nihilbono@sbcglobal.net wrote:
> Hello olrailbird,
> intersting idea, Spinner Theory. The question I have, among many
> others;
> if you are randomizing each entrant's speed numbers by spinning an
> arrow on a child's toy, is, how do you propose to handle?, allow?, for
> an entrant's smaller sample size?

CURRENT SPINNER THEORY IN A NUTSHELL:

Fact is, I don't really know what I'm doing
(mathematically) so I've done the best I can
given my numerological handicap.

I iterate through the horses, comparing each
horse's speed rating to each of the other
horses' speed ratings. Any rating that beats
or ties an opponent's speed rating is
considered a WIN. Add up wins for each!

Let's go back to the miserable race that
lost badly -- only because it is familiar
and contained only 5 horses!

10/29/05 GG Race 1

My final output (odds before takeout):
2 STORMIN' LYON 2.3
3 SIPHONIZER 3.3
4 TWENTYTHREEJAYBIRD 4.4
5 CARTHAGE 4.9
1 SEATTLES BEST JOE 8.0 (won in blowout! Areek!)

Here's a count of WINS, TRIES and PPs for each
horse. The last column is WINS/PP's to normalize
each horse's WinPct...

# W Try PP W/PP
1 72, 280, 7, 10.29
2 282, 370, 10, 28.20
3 214, 370, 10, 21.40
4 173, 370, 10, 17.30
5 156, 370, 10, 15.60

TRIES is PPs of studied horse times sum of PPs
for opponents. Ain't real important, but I
thought I'd explain it...I don't multiply,
I COUNT (my pebbles.)

The ratio (3rd column) normalizes WINS to number
of starts and explains how I handle smaller/larger
sample sizes. #1 wins about 10% against these!

If I sum the normalized win ratios (3rd column)
I get the total win percentage for all horses
in the race. This won't equal 100%.

In this race, TotPct =3D 92.79 (rounding always)

Dividing each horse's normalized win percentage
by the normalized total (TotPct) of the race
(sum of all horses' normalized win pct), we
get the percent chance of a horse winning
and can calculate raw odds. (If you want to
know tote odds use "((1-take)/pct)-1".

(round a bit, please)
1 10.29 / 92.79 =3D .1109 =3D 8.0
2 28.20 / 92.79 =3D .3039 =3D 2.3
3 21.40 / 92.79 =3D .2306 =3D 3.3
4 17.30 / 92.79 =3D .1864 =3D 4.4
5 15.60 / 92.79 =3D .1681 =3D 4.9

So, there you have it, Joe. Thanks a lot
for asking as this post represents the
clearest explanation of current spinner
theory I have. I will copy this and keep it...

> Seattle's Best Joe has only 7 numbers. Would you average them and fill
> in the remaining 3 places with that average? I'm on very shaky
> educated, HAH!, ground on this but would you lean more to variance
> than averages? By shaky ground I mean I THINK I understand 'analysis
> of variance'- book definition- but I probably don't.
> If I'm totally ignorant of the meaning of AoV please have pity and
> ignore the above sentence.

Joe, you and I seem to be about equally ignorant
as far as math goes...No offense meant.

I intend to write a book that explains crap like
this in terms ordinary people can understand. I
think most horseplayers slept through math...

Re-read the above and you'll see I don't "fill in" anything or
lean on "variance" or "averages." Whack the ratings one against
another and see who wins is my approach...
=20
> While you are exploring your idea shouldn't you include lots of other,
> documented items?
> I'll list a few that pop into my beer-besotted head;
> Works
> J/T numbers
> Win Percentages-with and without Conditions
> Size of barn string
> Hmmmm-Handicapping. I don't mean handicapping as to belittle what
> you're trying to accomplish, I mean using public knowledge with your
> approach to come up with an advantage.
> I guess I'm saying expand your data criteria.
> I haven't a clue as to how you would place those other criteria on
> your Spinner Board but I think they would be useful in some way.
> You,I and everyone else who has been at this game for a while can
> recognize a 'declining horse'. Shouldn't you weight that handicapping
> knowledge in some way?

Joe, I'm starting OVER -- completely. NOTHING else is being
considered here except my own speed ratings as best I can
calculate them on each horse's last (up to) 10 speed ratings.

My last program took the "include more" approach and ultimately
became a reflection of the tote board. If you mimic the tote,
you can't win...

That's it. If this "works" (and I expect it does) screw all
other data...Why confuse myself with "information overload?"

I think Beyer was more correct when he believed "Speed is Class."
Now he trip handicaps and his public image has suffered greatly!

Notice that he still has his "day job" as I do! His money comes
from selling us his ratings/books (not bad, but overrated.)

> As to being mentioned in your upcoming Best Seller-


Best seller? Areek! I doubt I'll make a nickel. Too many
sacred cows will be sacrificed to ever hope DRF PRESS will
publish my mad ravings...DRF is the industry...

> "And my thanks to Nihilbono for showing me which way NOT to go", would
> be OK.
I'm still on course...

> Isn't the root of infamy still 'famous'? ;-)

Yeah. I'm more interested in making money than famous, though.

> I'm running out of beer, I'm very saddened to say.

Beer is nice but scotch has more value. Value/Cost ratio
you know! Invest in the cheap solution, I say!

> Joe

--=20
Good Luck,
olrailbird
www.stevenberry.com

------=_NextPart_000_0047_01C5DF71.038D0840
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" >
<HTML ><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1" >
<META content=3D"MSHTML 6.00.2800.1522" name=3DGENERATOR >
<STYLE ></STYLE>
</HEAD >
<BODY >
<DIV ><FONT face=3D"Courier New" size=3D2>Some good questions, =
Joe.</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><A href=3D"mailto:nihilbono@sbcglobal.net"><FONT face=3D"Courier =
New"=20
size=3D2 >nihilbono@sbcglobal.net</FONT></A><FONT face=3D"Courier New" =
size=3D2 >=20
wrote:<BR >> Hello olrailbird,<BR>> intersting idea, Spinner =
Theory.=20
The  question I have, among many<BR >> others;<BR>>  if =
you are=20
randomizing each entrant's speed numbers by spinning an<BR >> arrow on =
a=20
child's toy, is, how do you propose to handle?, allow?, for<BR >> an =
entrant's=20
smaller sample size?</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>CURRENT SPINNER THEORY IN A=20
NUTSHELL:</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT><FONT face=3D"Courier =
New"=20
size=3D2 ></FONT><FONT face=3D"Courier New" size=3D2></FONT><FONT =
face=3D"Courier New"=20
size=3D2 ></FONT><FONT face=3D"Courier New" size=3D2></FONT><BR><FONT=20
face=3D"Courier New" size=3D2 >Fact is, I don't really know what I'm=20
doing</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>(mathematically) so I've =
</FONT ><FONT=20
face=3D"Courier New" size=3D2 >done the best I can</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>given my numerological=20
handicap.</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>I iterate through the horses, =
comparing=20
each</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>horse's speed rating to =
</FONT ><FONT=20
face=3D"Courier New" size=3D2 >each of the other</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>horses' speed ratings. Any =
rating that=20
beats</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>or </FONT><FONT face=3D"Courier =
New"=20
size=3D2 >ties an opponent's speed rating is</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>considered a WIN. Add up wins =
for=20
each!</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>Let's go back to the miserable =
race=20
that</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>lost badly -- only because =
</FONT ><FONT=20
face=3D"Courier New" size=3D2 >it is familiar</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>and contained only 5 =
horses!</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>10/29/05 GG Race 1</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>My final output (odds before=20
takeout):</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>2 STORMIN'=20
LYON      2.3<BR >3=20
SIPHONIZER         3.3<BR >4=20
TWENTYTHREEJAYBIRD 4.4<BR >5=20
CARTHAGE          =
 4.9<BR >1=20
SEATTLES BEST JOE  8.0 (won in blowout! Areek!)</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>Here's a count of WINS, =
TRIES and=20
PPs for each</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>horse. The last column is =
WINS/PP's to=20
normalize</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>each horse's =
WinPct...</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>#   W  Try  =

PP  W/PP</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>1  72, 280,  7,=20
10.29</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>2 282, 370, 10, =
28.20</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>3 214, 370, 10, =
21.40</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>4 173, 370, 10, =
17.30</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>5 156, 370, 10, =
15.60</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>TRIES is PPs of studied horse =
times sum of=20
PPs</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>for opponents. Ain't real =
important, but=20
I</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>thought I'd explain it...I =
don't=20
multiply,</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>I COUNT (my =
pebbles.)</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>The ratio (3rd =
column) normalizes WINS=20
to number</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>of starts and explains =
</FONT ><FONT=20
face=3D"Courier New" size=3D2 >how I handle smaller/larger</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>sample sizes. #1 wins about 10% =
against=20
these!</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>If I sum the normalized win =
ratios (3rd=20
column)</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>I get the total win percentage =
for all=20
horses</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>in the race. This won't equal=20
100%.</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>In this race, TotPct =3D 92.79 =
(rounding=20
always)</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>Dividing each horse's =
normalized win=20
percentage</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>by the normalized total =
(TotPct) of the=20
race</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>(sum of all horses' normalized =
win pct),=20
we</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>get the percent chance of a =
horse=20
winning</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>and can calculate raw odds. (If =
you want=20
to</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>know tote odds use=20
"((1-take)/pct)-1".</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>(round a bit, =
please)</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>1 10.29 / 92.79 =3D .1109 =3D =
8.0</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>2 28.20 / 92.79 =3D .3039 =3D =
2.3</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>3 21.40 / 92.79 =3D .2306 =3D =
3.3</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>4 17.30 / 92.79 =3D .1864 =3D =
4.4</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>5 15.60 / 92.79 =3D .1681 =3D =
4.9</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>So, there you have it, Joe. =
Thanks a=20
lot</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>for asking as this post =
represents=20
the</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>clearest explanation of current =

spinner</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>theory I have. I will copy this =
and keep=20
it...</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>> Seattle's Best Joe has =
only 7 numbers.=20
Would you average them and fill<BR >> in the remaining 3 places with =
that=20
average? I'm on very shaky<BR >> educated, HAH!, ground on this but =
would you=20
lean more to variance<BR >> than averages? By shaky ground I mean I =
THINK I=20
understand 'analysis<BR >> of variance'- book definition- but I =
probably=20
don't.<BR >> If I'm totally ignorant of the meaning of AoV please have =
pity=20
and<BR >> ignore the above sentence.<BR></FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>Joe, you and I seem =
to be about=20
equally ignorant</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>as far as math goes...No =
offense=20
meant.</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT><FONT face=3D"Courier =
New"=20
size=3D2 ></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>I intend to write a book that =
explains crap=20
like</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>this in terms ordinary people =
can=20
understand. I</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>think most </FONT><FONT =
face=3D"Courier New"=20
size=3D2 >horseplayers slept through math...</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>Re-read the above and you'll =
see I don't=20
"fill in" anything or</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>lean on "variance" or =
"averages." Whack the=20
ratings one against</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>another and see who wins is my=20
approach...</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2> <BR>> While you are =
exploring your=20
idea shouldn't you include lots of other,<BR >> documented =
items?<BR >> I'll=20
list a few that pop into my beer-besotted head;<BR >> Works<BR>> =
J/T=20
numbers<BR >> Win Percentages-with and without Conditions<BR>> Size =
of barn=20
string<BR >> Hmmmm-Handicapping. I don't mean handicapping as to =
belittle=20
what<BR >> you're trying to accomplish, I mean using public knowledge =
with=20
your<BR >> approach to come up with an advantage.<BR>> I guess I'm =
saying=20
expand your data criteria.<BR >> I haven't a clue as to how you would =
place=20
those other criteria on<BR >> your Spinner Board but I think they =
would be=20
useful in some way.<BR >> You,I and everyone else who has been at this =
game=20
for a while can<BR >> recognize a 'declining horse'. Shouldn't you =
weight that=20
handicapping<BR >> knowledge in some way?<BR></FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>Joe, I'm starting OVER -- =
completely.=20
NOTHING else is being</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>considered here except my own =
speed ratings=20
as best I can</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>calculate them on each horse's =
last (up to)=20
10 speed ratings.</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>My last program took the =
"include more"=20
approach and ultimately</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>became a reflection of the tote =
board. If=20
you mimic the tote,</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>you can't win...</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>That's it. If this "works" =
(and I=20
expect it does) screw </FONT ><FONT face=3D"Courier New" =
size=3D2 >all</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>other data...Why confuse myself =
with=20
"information overload?"</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>I think Beyer was more correct =
when he=20
believed "Speed is Class."</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>Now he trip handicaps and his =
public image=20
has suffered greatly!</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>Notice that he still has his =
"day job" as I=20
do! His money comes</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>from selling us his =
ratings/books (not bad,=20
but overrated.)</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>> As to being mentioned in =
your upcoming=20
Best Seller-<BR >
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>Best seller? Areek! I =
doubt I'll make=20
a nickel. Too many</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>sacred cows will be sacrificed =
to ever hope=20
DRF PRESS will</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>publish my mad ravings...DRF is =
the=20
industry...</FONT ></DIV>
<DIV > </DIV></FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>> "And my thanks to =
Nihilbono for=20
showing me which way NOT to go", would<BR >> be OK.<BR>I'm still on=20
course...</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>> Isn't the root of infamy =
still=20
'famous'? ;-)<BR ></FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>Yeah. I'm more interested in =
making money=20
than famous, though.</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>> I'm running out of beer, =
I'm very=20
saddened to say.</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>Beer is nice but scotch has =
more value.=20
Value/Cost ratio</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>you know! Invest in the cheap =
solution, I=20
say!</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT><FONT face=3D"Courier New" size=3D2>> =
Joe</FONT ></FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2><BR>-- <BR>Good =
Luck,<BR >olrailbird<BR><A=20
href=3D"http://www.stevenberry.com"color=#0000FF> >www.stevenberry.com</A><BR></FONT></D=
IV ></BODY></HTML>

------=_NextPart_000_0047_01C5DF71.038D0840--



02 Nov 2005 06:00:38
olrailbird
Re: Spinner Theory

This is a multi-part message in MIME format.

------=_NextPart_000_0087_01C5DF72.BF7AF7A0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

--------------------------------
# W Try PP W/PP
1 72, 280, 7, 10.29
2 282, 370, 10, 28.20
3 214, 370, 10, 21.40
4 173, 370, 10, 17.30
5 156, 370, 10, 15.60

TRIES is PPs of studied horse times sum of PPs
for opponents. Ain't real important, but I
thought I'd explain it...I don't multiply,
I COUNT (my pebbles.)

The ratio (3rd column) normalizes WINS to number
of starts and explains how I handle smaller/larger
sample sizes. #1 wins about 10% against these!
--------------------------------

NOT 3rd column, but 4th. You knew that...

I rewrote adding TRIES and forgot to update.

Imagine the program bugs I have to fix!

--=20
Good Luck,
olrailbird
www.stevenberry.com

------=_NextPart_000_0087_01C5DF72.BF7AF7A0
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" >
<HTML ><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1" >
<META content=3D"MSHTML 6.00.2800.1522" name=3DGENERATOR >
<STYLE ></STYLE>
</HEAD >
<BODY bgColor=3D#ffffff >
<DIV >
<DIV ><FONT face=3D"Courier New"=20
size=3D2 >--------------------------------</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>#   W  Try  =

PP  W/PP</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>1  72, 280,  7,=20
10.29</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>2 282, 370, 10, =
28.20</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>3 214, 370, 10, =
21.40</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>4 173, 370, 10, =
17.30</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>5 156, 370, 10, =
15.60</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>TRIES is PPs of studied horse =
times sum of=20
PPs</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>for opponents. Ain't real =
important, but=20
I</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>thought I'd explain it...I =
don't=20
multiply,</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>I COUNT (my =
pebbles.)</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2></FONT> </DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>The ratio (3rd =
column) normalizes WINS=20
to number</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>of starts and explains =
</FONT ><FONT=20
face=3D"Courier New" size=3D2 >how I handle smaller/larger</FONT></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>sample sizes. #1 wins about 10% =
against=20
these!</FONT ></DIV>
<DIV ><FONT face=3D"Courier New" size=3D2>
<DIV ><FONT face=3D"Courier New"=20
size=3D2 >--------------------------------</FONT></DIV>
<DIV > </DIV>
<DIV >NOT 3rd column, but 4th. You knew that...</DIV>
<DIV > </DIV>
<DIV >I rewrote adding TRIES and forgot to update.</DIV>
<DIV > </DIV>
<DIV >Imagine the program bugs I have to fix!</DIV>
<DIV ><BR>-- <BR>Good Luck,<BR>olrailbird<BR><A=20
href=3D"http://www.stevenberry.com"color=#0000FF> >www.stevenberry.com</A><BR></DIV></FO=
NT ></DIV></DIV></BODY></HTML>

------=_NextPart_000_0087_01C5DF72.BF7AF7A0--



02 Nov 2005 11:40:27
ellis_jay
Re: Spinner Theory

olrailbird wrote:
> --------------------------------
> # W Try PP W/PP
> 1 72, 280, 7, 10.29
> 2 282, 370, 10, 28.20
> 3 214, 370, 10, 21.40
> 4 173, 370, 10, 17.30
> 5 156, 370, 10, 15.60
>
> TRIES is PPs of studied horse times sum of PPs
> for opponents. Ain't real important, but I
> thought I'd explain it...I don't multiply,
> I COUNT (my pebbles.)
>
> The ratio (3rd column) normalizes WINS to number
> of starts and explains how I handle smaller/larger
> sample sizes. #1 wins about 10% against these!
> --------------------------------
>
> NOT 3rd column, but 4th. You knew that...

I thought something was fishy...lol




>
> I rewrote adding TRIES and forgot to update.
>
> Imagine the program bugs I have to fix!

--

Their ethics are a short summary of police ordinances: for them the
most important thing is to be a useful member of the state, and to air
their opinions in the club of an evening; they have never felt the
homesickness for something unknown and far away, nor the depths which
consists in being nothing at all. ___________Soren Kierkegaard

Ellis_jay




03 Nov 2005 18:04:33
Re: Spinner Theory

>Beer is nice but scotch has more value. Value/Cost ratio
>you know! Invest in the cheap solution, I say!
Off topic but THAT brings back a memory!
In my 20s and doing the bar thing.
I was buying J&B or Black and White and my drunken friend watched me
put a $1 bill on the bar, get the shot, One Dime change, down it and
said, "You had to aquire a taste for Scotch, didn't you?"
"Yes" I remember replying-I think I remember replying.
He pulls One Quarter and One Dime out of his pocket and says to the
drunken bartender, "Corby's , please",
downs it and asks me, "Why in hell didn't you aquire a taste for the
cheap stuff"?
I was speechless.
My liver is somewhat better now; my doctor assures me. When he
percusses? it it doesn't rebound so much.
Joe- I do love single malts. Oban is my favorite
BTW I think I posted this before(:-(.
I think I'll do search and see how my memory has edited the anecdote.
Nah, to hell with it.
The essence is correct.
I have a good memory it is just that I forget a lot of things. I think.



05 Nov 2005 14:09:26
T Block
Re: Spinner Theory

> Imagine spinning wheels with the speedratings on
> them. Now spin all 5 wheels over and over.
> I guess I could do this by modeling the numbers
> and generating 10,000 random results -- but
> I'm asking for help doing the math instead.
>

Okay on this standard normal curve, x-coordinates of 0 to 1 represents 34.1%
probability, x-coordinates of 1 to 2 represent 13.6% probability,
x-coordinates of 2 to 3 represent 2.1% probability, and x-coordinates of 3
to 4 represent 0.1% probability.

Now with speed figures of

82, 75, 88, 85, 90, 78

center on zero with translation of

90 - 75 = 15
15/2 + 75 = 82.5

and so subtract 82.5 from each speed figure for

-0.5, -7.5, 5.5, 2.5, 7.5, -4.5 .

Now scale for the range of -4 to 4 as

4 / 7.5

and so multiply each value by 0.5333 for

-0.27, -4, 2.93, 1.33, 4, -2.4

as the final x-coordinates.

Now the first speed figure has a probability of 34.1%, the second speed
figure has a probability of 0.1%, the third speed figure has a probability
of 2.1%, the fourth speed figure has a probability of 13.6%, the fifth speed
figure has a probability of 0.1%, and the sixth speed figure has a
probability of 2.1% .

----------------------------------------------------

Of course x-coordinates of -1 to 0 have a 34.1% probability while
x-coordinates of 0 to 1 also have a 34.1% probability.

So -1 equals to what speed figure ?

That's

-1 / 0.5333 = -1.875
-1.875 + 82.5 = 80.63 .

And 1 equals to what speed figure ?

Thats

1 / 0.5333
1.875 + 82.5 = 84.375 .


But that's only relative to the given range of speed figures...

---------------------------------------------------------

Now for a graphical view the y-coordinates are

0.38, 0, 0.01, 0.16, 0, 0.02

and the y-coordinate for x-coordinate of 0 is

0.4 .

And that's calculated from the formula in the previous messages...

-------------------------------------------------------






05 Nov 2005 18:30:40
T Block
Re: Spinner Theory

> Okay on this standard normal curve, x-coordinates of 0 to 1 represents
> 34.1% probability, x-coordinates of 1 to 2 represent 13.6% probability,
> x-coordinates of 2 to 3 represent 2.1% probability, and x-coordinates of 3
> to 4 represent 0.1% probability. Now with speed figures of
> 82, 75, 88, 85, 90, 78 center on zero with translation of
> 90 - 75 = 15
> 15/2 + 75 = 82.5 and so subtract 82.5 from each speed figure for
>
> -0.5, -7.5, 5.5, 2.5, 7.5, -4.5 .
>
> Now scale for the range of -4 to 4 as
> 4 / 7.5 and so multiply each value by 0.5333 for
>
> -0.27, -4, 2.93, 1.33, 4, -2.4
>
> as the final x-coordinates.
>
> Now the first speed figure has a probability of 34.1%, the second speed
> figure has a probability of 0.1%, the third speed figure has a probability
> of 2.1%, the fourth speed figure has a probability of 13.6%, the fifth
> speed figure has a probability of 0.1%, and the sixth speed figure has a
> probability of 2.1% .
>

Now, 1.0 has a probability of 34.1% while 1.01 has a probability of 13.6% ?

So think of 0.0 "or better" as a probability of 34.1 + 13.6 + 2.1 + 0.1 .

Now, 1.0 "or better" is 13.6 + 2.1 + 0.1 and 1.01 "or better" is about the
same figure...

So, you need to calculate the area under the curve from 0 to the
x-coordinate of the required value and then subtract from 0.5 ... So you
need either a formula for area under this curve from one x-coordinate to
another or you need a chart of values.




06 Nov 2005 13:19:24
T Block
Re: Spinner Theory

>> Okay on this standard normal curve, x-coordinates of 0 to 1 represents
>> 34.1% probability, x-coordinates of 1 to 2 represent 13.6% probability,
>> x-coordinates of 2 to 3 represent 2.1% probability, and x-coordinates of
>> 3 to 4 represent 0.1% probability. Now with speed figures of
>> 82, 75, 88, 85, 90, 78 center on zero with translation of
>> 90 - 75 = 15
>> 15/2 + 75 = 82.5 and so subtract 82.5 from each speed figure for
>>
>> -0.5, -7.5, 5.5, 2.5, 7.5, -4.5 .
>>
>> Now scale for the range of -4 to 4 as
>> 4 / 7.5 and so multiply each value by 0.5333 for
>>
>> -0.27, -4, 2.93, 1.33, 4, -2.4
>>
>> as the final x-coordinates.
>>

So the first speed figure has a probability of 0.39 or 39%, the second
figure has a probability of 0.00003 or 0.003%, the third speed figure
has a probability of 0.00169 or 0.169%, the fourth speed figure has
a probability of 9.3%, the fifth speed figure has a probability of
0.003%, and the sixth speed figure has a probabiltiy of 8.2% .

And that's using the chart below which represents areas from 0 to x already
subtracted from 0.50 so that the chart is actual probabilities.

However, it appears that the data should have been scaled from -3 to 3
instead of from -4 to 4 ... because all the values from 3 to 4 are very
small and 3 itself is small enough.

z .00 .01 .02 .03 .04 .05 .06 .07 .08 .09
-4.0 0.00003 0.00003 0.00003 0.00003 0.00003 0.00003 0.00002 0.00002
0.00002 0.00002
-3.9 0.00005 0.00005 0.00004 0.00004 0.00004 0.00004 0.00004 0.00004
0.00003 0.00003
-3.8 0.00007 0.00007 0.00007 0.00006 0.00006 0.00006 0.00006 0.00005
0.00005 0.00005
-3.7 0.00011 0.00010 0.00010 0.00010 0.00009 0.00009 0.00008 0.00008
0.00008 0.00008
-3.6 0.00016 0.00015 0.00015 0.00014 0.00014 0.00013 0.00013 0.00012
0.00012 0.00011
-3.5 0.00023 0.00022 0.00022 0.00021 0.00020 0.00019 0.00019 0.00018
0.00017 0.00017
-3.4 0.00034 0.00032 0.00031 0.00030 0.00029 0.00028 0.00027 0.00026
0.00025 0.00024
-3.3 0.00048 0.00047 0.00045 0.00043 0.00042 0.00040 0.00039 0.00038
0.00036 0.00035
-3.2 0.00069 0.00066 0.00064 0.00062 0.00060 0.00058 0.00056 0.00054
0.00052 0.00050
-3.1 0.00097 0.00094 0.00090 0.00087 0.00084 0.00082 0.00079 0.00076
0.00074 0.00071
-3.0 0.00135 0.00131 0.00126 0.00122 0.00118 0.00114 0.00111 0.00107
0.00103 0.00100
-2.9 0.00187 0.00181 0.00175 0.00169 0.00164 0.00159 0.00154 0.00149
0.00144 0.00139
-2.8 0.00256 0.00248 0.00240 0.00233 0.00226 0.00219 0.00212 0.00205
0.00199 0.00193
-2.7 0.00347 0.00336 0.00326 0.00317 0.00307 0.00298 0.00289 0.00280
0.00272 0.00264
-2.6 0.00466 0.00453 0.00440 0.00427 0.00415 0.00402 0.00391 0.00379
0.00368 0.00357
-2.5 0.00621 0.00604 0.00587 0.00570 0.00554 0.00539 0.00523 0.00508
0.00494 0.00480
-2.4 0.00820 0.00798 0.00776 0.00755 0.00734 0.00714 0.00695 0.00676
0.00657 0.00639
-2.3 0.01072 0.01044 0.01017 0.00990 0.00964 0.00939 0.00914 0.00889
0.00866 0.00842
-2.2 0.01390 0.01355 0.01321 0.01287 0.01255 0.01222 0.01191 0.01160
0.01130 0.01101
-2.1 0.01786 0.01743 0.01700 0.01659 0.01618 0.01578 0.01539 0.01500
0.01463 0.01426
-2.0 0.02275 0.02222 0.02169 0.02118 0.02067 0.02018 0.01970 0.01923
0.01876 0.01831
-1.9 0.02872 0.02807 0.02743 0.02680 0.02619 0.02559 0.02500 0.02442
0.02385 0.02330
-1.8 0.03593 0.03515 0.03438 0.03362 0.03288 0.03216 0.03144 0.03074
0.03005 0.02938
-1.7 0.04456 0.04363 0.04272 0.04181 0.04093 0.04006 0.03920 0.03836
0.03754 0.03673
-1.6 0.05480 0.05370 0.05262 0.05155 0.05050 0.04947 0.04846 0.04746
0.04648 0.04551
-1.5 0.06681 0.06552 0.06425 0.06301 0.06178 0.06057 0.05938 0.05821
0.05705 0.05592
-1.4 0.08076 0.07927 0.07780 0.07636 0.07493 0.07353 0.07214 0.07078
0.06944 0.06811
-1.3 0.09680 0.09510 0.09342 0.09176 0.09012 0.08851 0.08691 0.08534
0.08379 0.08226
-1.2 0.11507 0.11314 0.11123 0.10935 0.10749 0.10565 0.10383 0.10204
0.10027 0.09852
-1.1 0.13566 0.13350 0.13136 0.12924 0.12714 0.12507 0.12302 0.12100
0.11900 0.11702
-1.0 0.15865 0.15625 0.15386 0.15150 0.14917 0.14686 0.14457 0.14231
0.14007 0.13786
-0.9 0.18406 0.18141 0.17878 0.17618 0.17361 0.17105 0.16853 0.16602
0.16354 0.16109
-0.8 0.21185 0.20897 0.20611 0.20327 0.20045 0.19766 0.19489 0.19215
0.18943 0.18673
-0.7 0.24196 0.23885 0.23576 0.23269 0.22965 0.22663 0.22363 0.22065
0.21769 0.21476
-0.6 0.27425 0.27093 0.26763 0.26434 0.26108 0.25784 0.25462 0.25143
0.24825 0.24509
-0.5 0.30853 0.30502 0.30153 0.29805 0.29460 0.29116 0.28774 0.28434
0.28095 0.27759
-0.4 0.34457 0.34090 0.33724 0.33359 0.32997 0.32635 0.32276 0.31917
0.31561 0.31206
-0.3 0.38209 0.37828 0.37448 0.37070 0.36692 0.36317 0.35942 0.35569
0.35197 0.34826
-0.2 0.42074 0.41683 0.41293 0.40904 0.40516 0.40129 0.39743 0.39358
0.38974 0.38590
-0.1 0.46017 0.45620 0.45224 0.44828 0.44433 0.44038 0.43644 0.43250
0.42857 0.42465
-0.0 0.50000 0.49601 0.49202 0.48803 0.48404 0.48006 0.47607 0.47209
0.46811 0.46414




07 Nov 2005 15:44:48
ellis_jay
Re: Spinner Theory

T Block wrote:
>>> Okay on this standard normal curve, x-coordinates of 0 to 1
>>> represents
>>> 34.1% probability, x-coordinates of 1 to 2 represent 13.6%
>>> probability, x-coordinates of 2 to 3 represent 2.1% probability,
>>> and x-coordinates of 3 to 4 represent 0.1% probability. Now with
>>> speed figures of 82, 75, 88, 85, 90, 78 center on zero with
>>> translation of 90 - 75 = 15
>>> 15/2 + 75 = 82.5 and so subtract 82.5 from each speed figure for
>>>
>>> -0.5, -7.5, 5.5, 2.5, 7.5, -4.5 .
>>>
>>> Now scale for the range of -4 to 4 as
>>> 4 / 7.5 and so multiply each value by 0.5333 for
>>>
>>> -0.27, -4, 2.93, 1.33, 4, -2.4
>>>
>>> as the final x-coordinates.
>>>
>
> So the first speed figure has a probability of 0.39 or 39%, the second
> figure has a probability of 0.00003 or 0.003%, the third speed figure
> has a probability of 0.00169 or 0.169%, the fourth speed figure has
> a probability of 9.3%, the fifth speed figure has a probability of
> 0.003%, and the sixth speed figure has a probabiltiy of 8.2% .
>
> And that's using the chart below which represents areas from 0 to x
> already subtracted from 0.50 so that the chart is actual
> probabilities.
>
> However, it appears that the data should have been scaled from -3 to 3
> instead of from -4 to 4 ... because all the values from 3 to 4 are
> very small and 3 itself is small enough.
>
> z .00 .01 .02 .03 .04 .05 .06 .07 .08 .09
> -4.0 0.00003 0.00003 0.00003 0.00003 0.00003 0.00003 0.00002
> 0.00002
> 0.00002 0.00002
> -3.9 0.00005 0.00005 0.00004 0.00004 0.00004 0.00004 0.00004
> 0.00004
> 0.00003 0.00003
> -3.8 0.00007 0.00007 0.00007 0.00006 0.00006 0.00006 0.00006
> 0.00005
> 0.00005 0.00005
> -3.7 0.00011 0.00010 0.00010 0.00010 0.00009 0.00009 0.00008
> 0.00008
> 0.00008 0.00008
> -3.6 0.00016 0.00015 0.00015 0.00014 0.00014 0.00013 0.00013
> 0.00012
> 0.00012 0.00011
> -3.5 0.00023 0.00022 0.00022 0.00021 0.00020 0.00019 0.00019
> 0.00018
> 0.00017 0.00017
> -3.4 0.00034 0.00032 0.00031 0.00030 0.00029 0.00028 0.00027
> 0.00026
> 0.00025 0.00024
> -3.3 0.00048 0.00047 0.00045 0.00043 0.00042 0.00040 0.00039
> 0.00038
> 0.00036 0.00035
> -3.2 0.00069 0.00066 0.00064 0.00062 0.00060 0.00058 0.00056
> 0.00054
> 0.00052 0.00050
> -3.1 0.00097 0.00094 0.00090 0.00087 0.00084 0.00082 0.00079
> 0.00076
> 0.00074 0.00071
> -3.0 0.00135 0.00131 0.00126 0.00122 0.00118 0.00114 0.00111
> 0.00107
> 0.00103 0.00100
> -2.9 0.00187 0.00181 0.00175 0.00169 0.00164 0.00159 0.00154
> 0.00149
> 0.00144 0.00139
> -2.8 0.00256 0.00248 0.00240 0.00233 0.00226 0.00219 0.00212
> 0.00205
> 0.00199 0.00193
> -2.7 0.00347 0.00336 0.00326 0.00317 0.00307 0.00298 0.00289
> 0.00280
> 0.00272 0.00264
> -2.6 0.00466 0.00453 0.00440 0.00427 0.00415 0.00402 0.00391
> 0.00379
> 0.00368 0.00357
> -2.5 0.00621 0.00604 0.00587 0.00570 0.00554 0.00539 0.00523
> 0.00508
> 0.00494 0.00480
> -2.4 0.00820 0.00798 0.00776 0.00755 0.00734 0.00714 0.00695
> 0.00676
> 0.00657 0.00639
> -2.3 0.01072 0.01044 0.01017 0.00990 0.00964 0.00939 0.00914
> 0.00889
> 0.00866 0.00842
> -2.2 0.01390 0.01355 0.01321 0.01287 0.01255 0.01222 0.01191
> 0.01160
> 0.01130 0.01101
> -2.1 0.01786 0.01743 0.01700 0.01659 0.01618 0.01578 0.01539
> 0.01500
> 0.01463 0.01426
> -2.0 0.02275 0.02222 0.02169 0.02118 0.02067 0.02018 0.01970
> 0.01923
> 0.01876 0.01831
> -1.9 0.02872 0.02807 0.02743 0.02680 0.02619 0.02559 0.02500
> 0.02442
> 0.02385 0.02330
> -1.8 0.03593 0.03515 0.03438 0.03362 0.03288 0.03216 0.03144
> 0.03074
> 0.03005 0.02938
> -1.7 0.04456 0.04363 0.04272 0.04181 0.04093 0.04006 0.03920
> 0.03836
> 0.03754 0.03673
> -1.6 0.05480 0.05370 0.05262 0.05155 0.05050 0.04947 0.04846
> 0.04746
> 0.04648 0.04551
> -1.5 0.06681 0.06552 0.06425 0.06301 0.06178 0.06057 0.05938
> 0.05821
> 0.05705 0.05592
> -1.4 0.08076 0.07927 0.07780 0.07636 0.07493 0.07353 0.07214
> 0.07078
> 0.06944 0.06811
> -1.3 0.09680 0.09510 0.09342 0.09176 0.09012 0.08851 0.08691
> 0.08534
> 0.08379 0.08226
> -1.2 0.11507 0.11314 0.11123 0.10935 0.10749 0.10565 0.10383
> 0.10204
> 0.10027 0.09852
> -1.1 0.13566 0.13350 0.13136 0.12924 0.12714 0.12507 0.12302
> 0.12100
> 0.11900 0.11702
> -1.0 0.15865 0.15625 0.15386 0.15150 0.14917 0.14686 0.14457
> 0.14231
> 0.14007 0.13786
> -0.9 0.18406 0.18141 0.17878 0.17618 0.17361 0.17105 0.16853
> 0.16602
> 0.16354 0.16109
> -0.8 0.21185 0.20897 0.20611 0.20327 0.20045 0.19766 0.19489
> 0.19215
> 0.18943 0.18673
> -0.7 0.24196 0.23885 0.23576 0.23269 0.22965 0.22663 0.22363
> 0.22065
> 0.21769 0.21476
> -0.6 0.27425 0.27093 0.26763 0.26434 0.26108 0.25784 0.25462
> 0.25143
> 0.24825 0.24509
> -0.5 0.30853 0.30502 0.30153 0.29805 0.29460 0.29116 0.28774
> 0.28434
> 0.28095 0.27759
> -0.4 0.34457 0.34090 0.33724 0.33359 0.32997 0.32635 0.32276
> 0.31917
> 0.31561 0.31206
> -0.3 0.38209 0.37828 0.37448 0.37070 0.36692 0.36317 0.35942
> 0.35569
> 0.35197 0.34826
> -0.2 0.42074 0.41683 0.41293 0.40904 0.40516 0.40129 0.39743
> 0.39358
> 0.38974 0.38590
> -0.1 0.46017 0.45620 0.45224 0.44828 0.44433 0.44038 0.43644
> 0.43250
> 0.42857 0.42465
> -0.0 0.50000 0.49601 0.49202 0.48803 0.48404 0.48006 0.47607
> 0.47209
> 0.46811 0.46414

I have found that taking it to the fourth decimal is sufficient.

--

Their ethics are a short summary of police ordinances: for them the
most important thing is to be a useful member of the state, and to air
their opinions in the club of an evening; they have never felt the
homesickness for something unknown and far away, nor the depths which
consists in being nothing at all. ___________Soren Kierkegaard

Ellis_jay




07 Nov 2005 16:16:51
ellis_jay
Re: Spinner Theory

T Block wrote:
>> Okay on this standard normal curve, x-coordinates of 0 to 1
>> represents
>> 34.1% probability, x-coordinates of 1 to 2 represent 13.6%
>> probability, x-coordinates of 2 to 3 represent 2.1% probability, and
>> x-coordinates of 3 to 4 represent 0.1% probability. Now with speed
>> figures of 82, 75, 88, 85, 90, 78 center on zero with translation of
>> 90 - 75 = 15
>> 15/2 + 75 = 82.5 and so subtract 82.5 from each speed figure for
>>
>> -0.5, -7.5, 5.5, 2.5, 7.5, -4.5 .
>>
>> Now scale for the range of -4 to 4 as
>> 4 / 7.5 and so multiply each value by 0.5333 for
>>
>> -0.27, -4, 2.93, 1.33, 4, -2.4
>>
>> as the final x-coordinates.
>>
>> Now the first speed figure has a probability of 34.1%, the second
>> speed figure has a probability of 0.1%, the third speed figure has a
>> probability of 2.1%, the fourth speed figure has a probability of
>> 13.6%, the fifth speed figure has a probability of 0.1%, and the
>> sixth speed figure has a probability of 2.1% .
>>
>
> Now, 1.0 has a probability of 34.1% while 1.01 has a probability of
> 13.6% ?
>
> So think of 0.0 "or better" as a probability of 34.1 + 13.6 + 2.1 +
> 0.1 .
>
> Now, 1.0 "or better" is 13.6 + 2.1 + 0.1 and 1.01 "or better" is
> about the same figure...
>
> So, you need to calculate the area under the curve from 0 to the
> x-coordinate of the required value and then subtract from 0.5 ... So
> you need either a formula for area under this curve from one
> x-coordinate to another or you need a chart of values.

http://www.mnstate.edu/wasson/ed602lesson7.htm

http://www.mnstate.edu/wasson/normcurveZ.gif

http://www.stat.berkeley.edu/users/stark/Java/Html/NormHiLite.htm

http://www.stat.ucla.edu/~dinov/courses_students.dir/Applets.dir/NormalCurveInteractive.html

http://www.ms.uky.edu/~cwood/norm_tab.pdf

http://www.statsoft.com/textbook/sttable.html

http://faculty.uncfsu.edu/dwallace/lesson%207.pdf

http://www-stat.stanford.edu/~naras/jsm/FindProbability.html

--

Their ethics are a short summary of police ordinances: for them the
most important thing is to be a useful member of the state, and to air
their opinions in the club of an evening; they have never felt the
homesickness for something unknown and far away, nor the depths which
consists in being nothing at all. ___________Soren Kierkegaard

Ellis_jay