[time-nuts] are any time-nuts also random-nuts?

[Dr. David Kirkby]

J. Forster wrote:
> Many years ago, when the State Lottern Numbers Game started up here, I
> noticed that doubles were coming up a lot...  like 4662. I was utterly
> convinced they were overrepresented.
> 
> However, being cheap and sane, I decided to test the theory before
> spending $$. I collected ALL the numbers for a year or more and did
> statistical tests, including FFTs on each digit wheel.
> 
> Bottom line, the numbers were random. It was my PERCEPTION of a pattern
> that was tricking me.
> 
> All in, I spent $1 for a ticket to read.
> 
> YMMV,
> 
> -John

There's a guy in my local pub, that believe he has a way to increase his chances 
of winning the National Lottery in the UK. There are 49 balls numbered 1-49, and 
6 balls are chosen at 'random'.

What he has discovered, with the aid of a spreadsheet, is that when the 6 
winning numbers are announced, they usually sum to a number somewhere in the 
range 130 to 170. Very rarely is the sum very low or very high. So he reckons 
that if he picks numbers 6 numbers, and ensures they sum to something in the 
range 130 to 170, he has more chances of winning than if he picks numbers with a 
low or a high sum.

Of course, the minimum they can sum to is 1+2+3+4+5+6 = 21, and the maximum is 
44+45+46+47+48+49 = 279.

My wife thought I was going to have an argument with him once over this, so I 
decided to let him believe that his strategy maximises his chances of winning.

I'm sure I buy him far more beers than he ever buys me, so I guess his stratergy 
has not yet netted him the jackpot. I believe in the UK, the exptected return on 
a ?1.00 ticket is ?0.14, which is why I never do the lottery myself.


Dave