I just heard on MSNBC that there was some debate within the campaign about how seriously she should contest Wisconsin. Obviously the pro-engagement side won out, but only after a bit of a delay. The question is, how costly was the delay?
Perusing the exit polls, I guess I'd say moderately costly, but probably not decisive. Voters who made their decision in the last three days of the campaign--presumably after Hillary had stepped up her efforts--broke for Obama 54-46. By contrast, voters who made their decision in the last week* broke for Obama 67-33, and in the last month, 66-30. So that's certainly an improvement. But, unless that last-three-days figure masks a steep, late-breaking trend toward Hillary (if, for example, she was losing 65-35 three days out, but day two saw a sharp swing the other way), I don't see how this would have put her over the top.
Now, it's true that Hillary appears to have narrowly carried voters who made up their minds today. But I suspect that had more to do with the brouhaha over Obama's alleged plagiarism than with any deeper underlying dynamic.
*My understanding is that these are mutually exlusive categories when pollsters ask the question. Which is to say, even though "last three days" and "last week" technically overlap, voters are only allowed to select one or the other.
--Noam Scheiber