It does male a difference in polling, though. In both the 2008 and 2012 elections, most of the "undecided" during the last campaign weeks used to be low-to-middle income, some-or-no- college women. If they don't - as is commonly assumed - break 50:50, or for the challenger, but vote majority democrat instead, all these nice "horserace" and "possible last-minute swing" narratives become futile.
P.S: The female vote also appears to be less 'swingy' than the male vote. Compared to 2008, Obama lost 1% support among women, compared to 4% among men. I haven't checked on older elctions whether such female "swing resistance" is a particular 'Obama-phenomenon', or a longer-term pattern. However, it is common marketing knowledge that women tend to display stronger brand loyality than men. As such, I think it may be much more difficult for Republicans to regain lost female votes, than for Democrats to win over more men.
These two paragraphs seem to contradict each other. Women are overrepresented among undecided voters, and yet the female vote is less swingy than the male vote? How are both of those things true?
In any case, I think the crux of Nichlemn's point is that the gender gap itself by definition shouldn't favor one party over the other. A candidate is wise to focus on convincing voters who are persuadable, whether those voters are disproportionately male or female. The actual breakdown of votes between the two sexes is irrelevant. It really doesn't matter whether you won a narrow election victory with gender parity, or with one sex voting disproportionately for you and the other against you. Both of those outcomes count the same.
