Romantic data-mining

Feb 04, 2009 20:24

It is well known, at least in certain circles, that the optimal strategy for choosing a spouse is to first date 37% of the people whom you could conceivably marry, and then continue dating people but marry the next one you come to whom you like better than all those you've dated so far. If you date less than 37% of people before doing this then, ( Read more... )

statistics, data mining, maths, social networks

Leave a comment

Comments 1

(The comment has been removed)

silverleaf_ February 4 2009, 15:10:32 UTC
Geez, the "11 possible partners" calculation is eye opening - especially considering the huge age range you used. Combined with the empirical observation that most people do actually end up finding a life partner eventually, it forces us to conclude that either most people have no standards whatsoever or most marriages / long term relationships are ridiculously sub-optimal.

I guess I'll concede that this sort of analysis may not have helped you find a partner, but I don't see how it could have hindered. I mean, either you made an error in your calculations or you really are extremely unlikely to ever meet someone who meets all of your conditions and whom you get along with. Unless your mistakes were simply having such a stringent criteria?

Reply


Leave a comment

Up