We’ve all heard of the Birthday Problem: The counterintuitive statistical fact that in a random grouping of 23 people, there’s a greater-than-50% chance that two of them have the same birthday.

Out of curiosity and a desire to distract my brain for a few minutes, I decided to test it out. I’m sure this is the kind of thing that a smarter person would know how to automate, but I just went to random.org (a truly great website that I hope never gets a redesign) and had it generate 10 sets of 25 numbers between 1 and 365. (I used 25 because that’s what I’d heard it was and I didn’t think to look it up before I started.) I pasted each set into a column in Excel and sorted each column into descending order, to make finding pairs easier.

Granted, using 25 numbers instead of 23 raised the odds a bit, but I didn’t expect it to be this strong: 8 of the 10 sets had repeated numbers. Five sets (half!) had more than one repeated number, and one set had *three* repeats.

That datum that five sets had multiple repeats may be what surprised me most. Eight sets overall had repeats, and five of those eight had multiples, so a set that had one repeated integer was *more likely than not* to have another repeated integer.

Obviously this is a very small sample size, but I’m comfortable with that because I’m a dilettante at this stuff and not obligated to cleave to scientific rigor.

If anyone would like to see my raw data for some reason, here is the spreadsheet.

Back to learning lines. I start rehearsals on Monday for *Blithe Spirit*.

I bet you know how to calculate the odds? Just calculate the chances of NOT matching a birthday... For 2 people it would be 364/365 - cause one day has been used up... And then 1 - (364/365) are the odds of a match. For three people it would be 1 - (364/365 * 363/365) - 363 cause two days have been used up when you're trying to find a new day for the third person... and then 1- (364/365 * 363/365 * 362/365 * 361/365....). It gets to 50% pretty quickly

-Mike