The values on that map are a little frustrating, have we got anything which offers more precision for those enormous 1-91 and 108-205 ranges? I look at that and think "Across the country the ratio could only vary between about 9 men for every 10 women and 9 women for every 10 men and the diagram would still be totally correct - is that even really significant?".

("Is that even really significant?" is a whole other question that I'm sure someone can easily answer. I don't have any intuition about how skewed a gender ratio has to be for it to be noticeable to a member of that population, certainly if someone told me that there were only 9 women for every 10 men in a city I wouldn't immediately think my chances of hooking up would be hurt too much...would they?)

Is the contention the word "huge" because the disparity is only ~103:100 in a lot of places? Because that's actually pretty big when it comes to dating, because of the "available pool" problem: Most members of both genders are paired up already, and each pair removes one member from each pool. Those with a different sexual orientation from you or those not actively seeking a relationship come out of the pool too. So you go from 103:100 to something like 13:10, which is pretty huge.