The only person to blame for the timing of the FBI investigation was Hillary, who destroyed government emails, wiped copies of those emails from all computers, stonewalled the investigation having the techs take the 5th, and ignoring the Congressional subpoena for a year.
WikiLeaks released the bulk of their Podesta emails earlier, not later. The FBI and State Sept also dumped emails for public viewing, which again took too long because of the Clinton/State Dept reviewers taking forever.
I doubt the few WikiLeaks dumps near the election had that much impact on voters if the previous mountain of dumps did not.
Trump did not have a lot of last minute conversions, either, no more than Reagan did. The polls then did the same thing. They created a wide margin with the Democrat leading to manipulate voters . Then as election week arrived, the polls magically stated to narrow. That narrowing was to ensure the pollsters didn't get caught being too far off when the election doesn't end as they predicted/tried to influence it.
Yup. Super tired of hearing about how she was a victim of this, timing of Comey's that, timing of whatever. At minimum, she was likely the beneficiary of a smear campaign against Sanders.
Sometimes I wondered if all of those polls showing Clinton winning by a landslide backfired. As in HRC supporters who figured it was in the bag, didn't turn out and vote and inspiring more pro-Trump or anti-HRC to get out and vote and be more vocal and active and rally support against HRC. Who knows. All we know is that Trump won. He's the 45th President of the United States of America. At least give the guy a chance to see what he can offer. I used to think that any one President couldn't do that much wrong/damage to this country. Whoa, I don't think that now.
To me, polls are practically meaningless. However, I do see them as a tool that can be used to sway people, both educated and not-so-educated. Yeah, there are some truly unbiased polls, but the vast majority ask questions that are at minimum biased toward supporting a hypothesis going in, to at times practically leading someone into a given answer.
Given the following options, which would you choose?
A) Do you want a broken down Chevy Silverado?
B) Do you want a brand new Ford F-150?
Well, I can still see some Chevy diehards going for A, but I would think most would go for B.

I've seen some pretty biased polls. Even ones that are seemingly unbiased can end up with results that contain a high level of doubt. One example was a poll on sexual assault on campus. The study was probably an honest attempt at getting data, but what it also tried to account for was how many incidents go unreported. In the end, even the researchers had serious doubts in the data as how could you reliably say whether or not someone who wouldn't report an incident would respond as such on a poll. Essentially, the poll gave some good data, but still not really conclusive on many points.