I'm referring to here[1], which states that Clinton is +15 on average compared to Trump with women. Yes, I did assume that it is only a 2-person playing field because I don't know how those polls were conducted(was it "who are you voting for?", or "would you choose Trump or Clinton?")
So, you conflated share of the voting age population with share of those planning to vote for one of the two major candidates, tried to get a number for that from percentage spread between the two major candidates among those planning on voting for any candidate, and then rounded favorably to your candidate. Resulting in an implausibly high number matching Trump's support in the general electorate even though the same source says that his support among women is substantially lower.
Okay. So what is the correct number? Do you disagree with the core of my argument, or are you just interested in quibbling?
If Trumps support was something extremely small like 1-9% for any group, then I would say there is a pretty good possibility that that group sees something very wrong with the candidate. If it is something more, like say in the twenties or higher, then there is hardly a consensus.
[1] http://fivethirtyeight.com/features/election-update-women-ar...