I have a feeling that approach is no better when compared with going on dates with a random sample of 88 women (within some range of compatible age/attractiveness), but I don't have the data to prove it.