# How will Timmyrocker attempt to lie his way out of this one?

"Mark Rafn" <dagon@xxxxxxxxx> wrote in message
news:l7mft2-ie6.ln1@xxxxxxxxxxxxxxxxxx
> >Mark Rafn wrote:
>>> Hardly. You're very likely to see many sets of 144 rolls that deviate a lot
>>> from the expectation of 24 7s.
>
> Timmyrocker <timmyrocker@xxxxxxx> wrote:
>>um... not by much actually. i have been impressed over and over by 144
>>rolls having almost exactly 24 sevens, rarely deviating from a range of
>>22-26 sevens.
>
> You haven't haven't counted precisely very often, or you've been surprisingly
> lucky in terms of seeing such a small variance in number of 7s in 144 rolls.
> You should see 22-26 about half the time. A little fudging and a little
> selective memory can turn this into "rarely deviating", I suppose...
>
> You should see 24, with a standard deviation somewhere around 4.5 (back of
> envelope, maybe it's a bit less). This means 90% of the time, it'll be within
> 7.2 rolls, 99% within 11.25. Looked at another way, there is a 1% chance for
> a
> set of 144 rolls to contain fewer than 13 or more than 35 7s.
>
> A range of 22-26 is about half a standard deviation, and just under half the
> time will a set of 144 rolls be in that range.
>
>>You will see an occasional high deviation, but i clearly wouldn't
>>characterize it as you did : "many sets of 144 rolls that deviate a
>>lot".
>
> Depends on your definition of "many", I guess. If someone is basing a theory
> on the idea that in 144 rolls, one is guaranteed to have 22-26 7s, the fact
> that over half the time they're wrong seems like a lot to me.
>
> If you're saying that it's often between 13 and 36, I'd say there's not a lot
> of deviation - only 1% or so of the time.

This presents Timmyrocker with a dilemma. He clearly does not understand the
Standard Normal Distribution and how it is inexorably linked to confirming that
trials are random and independent.

Timmyrocker, The Science Guy, is now confronted with reconciling his
manufactured extensive collection (100's ?) of "sets 144 roll dice records" with
his statistically unsupportable claims that 90+% of his records show one half of
a standard deviation or less in the number of sevens.

Timmyrocker made the same preposterous claim of low deviation distributions when
he previously manufactured his thousands of trials contained in sets of 100 coin
flip records.

Innumeracy is a great danger to those frauds who hope to "cook the books" to
support their preposterous assertions based in their intuitive misunderstanding
of the relation of averages to distribution and probability.

The only question that remains is ... How will Timmyrocker attempt to lie his
way out of this one?

--
Onward thru the fog,
Mason

TommyKnocker's Law - Modify your current bets based on prior results because it
all evens out.

.

• Follow-Ups:

## Relevant Pages

• Re: Why online poker is better for the money player.
... >> surprising deviation from the expected (expected in the mathematical ... >small sample size (five million rolls) ... and I got 16.66548% doubles. ... by re-seeding the rng fairly often. ...
(rec.games.backgammon)
• Re: How Many Sevens Should You Expect in Six Throws of the Dice?
... >22-26 sevens. ... lucky in terms of seeing such a small variance in number of 7s in 144 rolls. ... selective memory can turn this into "rarely deviating", ... A range of 22-26 is about half a standard deviation, ...
(rec.gambling.craps)
• Re: How Many Sevens Should You Expect in Six Throws of the Dice?
... You're very likely to see many sets of 144 rolls that deviate a lot ... 22-26 sevens. ... In order to later reflect on my gambling sessions, ... You will see an occasional high deviation, ...
(rec.gambling.craps)
• Re: Bill Reid, Kelly Criterion
... of using a distribution to assess risk/gain, ... started thinking about diversification. ... with a standard deviation of 0.6491%. ... the "law of large numbers", that is, "as the number of trials grows large, ...
(misc.invest.stocks)
• Re: Resistance to Solder Heat, practical values, SMD resistor
... and expensive to test enough resistors to have a clear idea of the ... distribution of resistance shifts after exposure to solder heat, ... or mean deviation of the product of the ... Gaussian distribution that came out of the production process. ...
(sci.electronics.design)