Niefelheim disasters--bug?

[PhilD]

Meglobob said:
The odds of flipping a head 10 times in a row are only 1/1024, when you consider how many random rolls you must make in Dom3, you are bound to get a 'losing run' every now and then.

Yeah, but flipping heads 100 times in a row are this to the 10th power - roughly, 1/10^30. This is really astronomically low. If every human being in existence (roughly 6.10^9 of us) tried this every second for a year (roughly 3.10^7 seconds), the probability of someone. somewhere, getting 100 heads in a row would be of order 1/5.10^13, not that much different from someone winning the "Loto" TWICE IN A ROW ("Loto" is a chance game where you pick 6 numbers out of a possible 49, and then exactly 6 numbers are rolled, and you "win" if you guessed all 6 - something you have about 1 chance in 40000000 of doing).

Or, to put it another way, if all 6 billion human beings had starded flipping coins around the Big Bang, a hundred flips per second, it would still be very unlikely that one of them had flipped 100 heads in a row by now.

Now, I'm the first to tell my students (I do teach probability theory, among other things) that if you try long enough, something extremely unlikely like flipping 100 heads in a row will happen. But it helps to have an idea of what "long enough" means. In this case, it's really long.