I didn’t want to completely hijack a thread, so I moved it to a new one: here’s where we are now…
**
Pardon my skepticism, but this seems too good to be true. Does it really apply in the world (is the birthday pattern really even?)? I thought that there was a greater percentage of people born in Jan. - Mar., which could really blow up the odds and get expensive and embarrassing fairly quickly.
Actually, this is what’s sometimes referred to as the “Birthday Inequality”. If the birthday’s are truly randomly distributed, then you get the probabilities I mentioned above. If they’re not randomly distributed (more in Jan.-Mar., for example), it can only increase the chances of at least two sharing a birthday, so you’d have an even better chance of winning a bet, in that case.
To test your theory, I went through my PalmPilot’s database, which has 72 ppl. in it, all with different birthdays. Either I’m one of the 1%, or something is amiss.
Please note I’m just playing devil’s advocate here; something just smells funny, and before I lose hundreds of dollars I like to make sure :).
With 72 people, it’s 99.95% likely that at least two will have the same birthday. Did you choose the 72 people randomly, or did you intentionally pick 72 that all have different birthdays? If you picked them randomly and they all have different birthdays, all I can say is that’s a huge fluke; there’s only a 1 in 2000 chance of that happening.
I can assure you that this isn’t just some quirk of the math, though, it really does “work” in the real world. Here’s some more about it if you’re still skeptical, and from there are additional links to even more pages, as well: