h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

If you ever get a coin-tossing to come up 998 out of 1000, even with a fast computer simulation, I'll eat my shorts.

Jules: Sorry to Quip....but it is the nature of this beast. "So many variables ... so little time". Show me these variables i.e. weight of said coin, mechanism to flip said coin consistently, (with out human parameters)
Personal. I suspect that that the human biological brain can not hold/comprehend all the cosmic/ universal variables... but we do try. The humility is the acceptance, of the simple statement of"I don't know", and listening and perhaps learning.

>>If you ever get a coin-tossing to come up 998 out of 1000, even with a fast computer simulation, I'll eat my shorts.<<

Look you suggested it as an example not me.

Anyway..bed.

You can't do it, though. It's the proper scale of the aberration in my opinion. There are seven dates, and it's confined to being an American mathematician who knows a little history.

You can't do it, though.



I disagree. my bed is just over there and the computer is going off - now.

<>

and how many people share the same birthday world wide??????

you only have 7 days a week so lots of people being born on a tuesday is not a coincidence....or is it

Sorry--eight dates, not seven. Dates A, B, C, D, E, F, G and H are linked by A, B and C being a triad (a naturally construed triple) of historical dates; date D being my date of birth which precedes by exactly 10 days, one month, and 399 days (forming a span of 400) E, F and G, which are the birth dates of my three closest same-generation relatives (placing me at the center) and the nineteenth, nineteenth and twentieth anniversaries of A, B and C. D as a number American style produces four times date A in a very simple rearrangement and also four times that same re-ordering of H when D is written as a number European style, H being the date of the seemingly unrelated historical event of Garry Kasparov being beaten by a computer, computer analysis of chess openings being that which took up a majority of my time in recent years, and chess having been taught to me by the cousin who was slightly less close to me than the other three--older, never slept with, don't know birthday.

By the way, the Moon is in the same phase generally on the same dates each nineteen years, with some variation (I'm sure) related to things not being that precise and there being leap years. The Baha'i calendar uses nineteen days for each of nineteen months with four or five intercalary days per year.

Taff: I don't know seasonal variability factors in volved in computing the answer to your (IRRELEVANT) question, but about 6.8 billion divided by 365.25. Since it is irrelevant, calculate that on your own calculator. This is about a coincidence involving not one but eight dates.

The weirdness thing in here is 7061964=4*1765491 and 6071964=4*1517991.

Just because you disagree with Julzes' numerology - doesn't mean you have to be rude and unpleasant to him.

FWIW, the chance of tossing 998 heads in a run of 1000 is around 5 times 10 to the power of -296 (i.e. 0.000000000....5) with 296 0's after the point. In other words, if you tossed one coin every second for the lifetime of the universe and you ever saw a sequence of 1000 tosses in which 998 are heads, then there's something weird going on and it's very unlikely that it's an unbiased coin.

Thank you, Dogster.

dogster

what do you make of julzes 'evidence' that evolution on earth is being manipulated by undetectable god-aliens with advanced technology, and julzes has been chosen as some sort of vessel of revelation due to some coincidence in his birthday and some of his relations and a computer winning a chess game???

And my birthday coincidence alone won't quite be that extreme, no matter how it's analyzed, but it will be extreme enough for most scientific questions where probabilities are invoked to investigate a strange causation. There are other coincidences (that both relate and don't relate to this specific one) like the fact that a specific calculator I happen to have been using gives a strongly biased sequence of final digits to the sequence of differences of the fourth roots of 17797577777.777...7 and 17797577777, where ".777...7" indicates from one up to some large number of sevens after the decimal point. The twentieth value in the sequence of differences is what I call my 'Weird Arithmetical Leap', where, in an attempt to gain confirmation that 1779 and 1914 in (365.25)^4 are actually dates, I got an affirmative result that the only 1 is followed by 776 and later noted that the preceding digits 462 also give something of a date coincidence in that 1264 is both 1964-700 and 1776-512 (512=2^9).

Taff: I'm impressed that you come close to asking a reasonable question fairly close to representing what I'm saying accurately.

<>

is it only one calculator that does this or do other calculators give different answers for the same calculation????

It's the 1998 version of the calculator that was updated to something else this year or some other intervening year. All calculators are subject to error, deending on the type of calculation. I think part of the reason I chose the difference I did was because this is the type of calculation bound to give an error of some significance. I tweaked the way I used the calculator using some numerical analysis--the branch of maths related to the question--to determine that the last seven digits are off. Maybe someone using big number calculations will want to e-mail the accurate numbers to me after finding out what my e-mail address is. [There is a way to find it, but I'm not going to be specific for anyone to do it.] I don't care that much, because the calculator served its purpose for now, but maybe there is something in the longer accurate version, too, for later. ...4621776... is in the accurate part of the calculation in question, by the way.

Gotta go.

>>FWIW, the chance of tossing 998 heads in a run of 1000 is around 5 times 10 to the power of -296 (i.e. 0.000000000....5) with 296 0's after the point. In other words, if you tossed one coin every second for the lifetime of the universe and you ever saw a sequence of 1000 tosses in which 998 are heads, then there's something weird going on and it's very unlikely that it's an unbiased coin.<<

Ah I see.

Well in the reply I typed out but didn't end up posting last night I did half heatedly speculate on checking I wasn't standing under a large magnet or using a biased coin.

But I ended up not including those thoughts on the basis it was Julze's example and I was at that point just imagining extra variables (large magnets) so I deleted it

However in the case of 'something weird going on' I accept that and my thoughts on magnets and biased coins I think shows I was at least considering casual mechanisms that could account for such weirdness.

However, I think the exchange across 21853 - 21855 shows we are outside of what we might call statistical clustering of odd things (like a 998/1000 coin toss) and into a wholly unjustified level of weird.

For instance against Julze's conviction that evolution is manipulated, we have to weight in the balance up all the evidence of the non-random survival of random varying replicants which shows that it isn't.

The coincidences Julzes thinks he's spotted are I suspect not significant in this case, as I accept now (I'm a bit more awake) coincidence in the coin toss example would be.

I do find his argument dubious and I'm highly sceptical, but I think thus far I've avoided being rude.

Hi Julzes,

>If I were not subject to a history of mental illness [...] I'd probably think it was my place in life to rule the world one day.

Mmm.

>Is it parsimonious to assume that smokers get lung cancer more often than other people because we don't have a large enough sample

With medical tests, we calculate how large a sample we would need for the result to be statistically significant (defined as less than 1 in 20 of happening by chance, though that is an arbitrary figure). I am still waiting for your response to my suggestion that you calculate how likely your results (or ones just like them) are by chance.

>the cousin who was slightly less close to me than the other three--older, never slept with

So that's how it is in their family
- Ferris Beuller's Day Off

Gif