Hacker News Reader

The same, in theory, applies to sequences like 1 2 3 4 5 6 "because that will never happen", and numbers >31 (birth days).

In practice though, the number of n jackpot winners is significantly less likely than 1... and do you really care if you only win $10M instead of $20M?

View on HN

Except 1,2,3,4,5,6 is actually bought way more (by hundreds) than would be bought by random.

It's best to use a source of random numbers. Any rules you construct just makes it more likely someone else is using the same constructions.

Exactly. Using a scheme to try to figure out a lottery number that other people won't choose is like using "ZZZZZZZZ" as a password, on the basis that anyone who tries to brute force it will try that one last.

And never play the same numbers because if you play the same numbers all the time and they are the winning numbers the one week you don't play...

No worries, they might win again the following week!

http://www.telegraph.co.uk/news/newstopics/howaboutthat/6202...

Reply on news.ycombinator.com