The GOP won big on Tuesday, and its winners include two new female senators -- Shelley Moore Capito in West Virginia and Joni Ernst in Iowa -- as well as the first black female Republican in Congress, Utah's Mia Love.
But the GOP's gains came in large part due to men, and in fact, there was a larger "gender gap" than we've seen in 20 years.
According to exit poll data, the 20-point gap between the voting preferences of men, who favored the GOP by 16 points, and women, who favored Democrats by four points, is bigger than any election since the 1994 GOP wave, according to exit poll data.
Here's how the gender gap looks over the last 20 years:
It didn't matter much in this election, given the GOP's increasing strength with men. And it's not like Republicans did especially poorly among women; their showing was one of the better ones they've had in recent years -- albeit worse than 2010. It's just that GOP momentum didn't register as much with women.
And once the pendulum swings back a little bit, the numbers Tuesday suggest women could remain a difficult demographic for Republicans.
I've argued before that the gender gap is somewhat oversold, in large part because it has existed for a long time now and is generally pretty consistent -- even before the Democrats started accusing Republicans of a "war on women." In addition, as the chart above shows, Republicans are doing better and better among men (who, after all, are nearly half of the electorate). Clearly, the GOP can win big even with a 20-point gender gap.
But the idea that the party that is already so reliant upon white voters is also increasingly reliant on men -- and has far fewer women members of Congress than Democrats -- isn't a helpful image as Republicans try to expand their brand. And party leaders are doing their best to fight that very perception.
Being the party of white men isn't what GOP leaders want. And despite winning on Tuesday, the gender gap suggests the party hasn't really chipped away at the second half of that perception.
Updated at 10:10 a.m.