ASU2003

I didn't do so well at discrete math, but I think your odds are a little off there. Let's say you had a sample size of 10,000 women. 20% are the top 20% in intelligence (you could use something else that is more definite like religion, hair color, weight, eye color, ...), so you have 2,000 remaining. Now lets say you wanted the next item, which is the top 20% in humor. You have to go back to your full 10,000 women to get the funniest top 20%. If you wanted the funniest 20% out of the women who passed the first condition, you may be left with the all of the un-funniest girls to choose from (imagine that smart=unfunny, stupid = funny). If you applied the first condition to take out the people who aren't smart enough, you may get women ranking numbers 8000 - 8400 in humor. Or you might find the opposite, where (smart = funny, unsmart=unfunny) and the top 20% in both intelligence and humor would leave you with 400, when in fact you should have 2000 left that he would be happy with..

But like I said, I might be wrong, statistics and discrete math weren't my favorite classes. And maybe I just didn't understand what you meant with your 'correlations and the anti-correlations cancel out' statment (sorry I re-read your post one more time once I wrote everything out).

There's also the fact, that he might just happen to be visiting a comedy club on women's night on a college campus with lots of international students that liked weird food in San Diego or Miami, he would be in a 'target-rich' environment. Where the odds of meeting his 'acceptable' girl are a 0% to 20% chance(or whatever the rarest condition % is) in the general public. His odds in real life went way up because every girl he sees here could meet his standards.

While I'll agree that you shouldn't filter out people lke this, because you may be throwing out a few girls you would be happy with while chasing after the 'perfect' girl that may not exist.

Then again, the only othere option to get the true number is to use the census data to get a accurate number of 18 million 20-29 year old females. http://en.wikipedia.org/wiki/Image:Uspop.svg
But you would have to resurvey those remaining to see how many remain after applying more filters, (unless that data is collected by the census bureau and they have the information and data from everyone) Using my filters, lets say 18 million are in the age bracket, after resurveying those you get 17 million are non-smokers, after surveying those there are 14 million are left if you consider exercise or activivity levels, and maybe there are actually 60,000 left if you say they need to live 50 miles away or less from me.

I don't know, I've written too much. But it is something I have spent a lot of time thinking about. The only plus side is that so many people end up in relationships, that it can't be too hard. I can go to the grocery store and see a few girls each time that I would like being in a relationship with, and if they asked me out, I would say yes.