Quote:
Originally Posted by ilnjscb
I'm not sure what you mean - 538 is a good source of data. Nate Silver clearly does not like Trump, but he seems to want to be fair to him. I'm not skilled enough at stat to evaluate his algorithms, but he does analyze elections well.
He was taken to task over his support for Trump (he gave him 33% chance instead of 3% like NYTimes) and the "statistician" who did so had to apologize.
Do they make mistakes, yes, of course. But overall they're as fair a source as any.
|
do yourself the favor and check back the day before the elections 2016.
the FBI investigation on hillary went public 11 days before the elections.
donīt you think this had a massive impact?
the polls on november 8. have been on 42,8 for trump and 45,9 for hillary.
this is more or less the CORRECT public result because trump did not win the public votes.
so what is wrong ? nothing !!!! just am FBI investigation against one candidate with an open result on the election day.
for this kind of circumstance the polls have been pretty accurate on 8.
the final result was
Clinton received 65,844,610 votes, or 48.2% of the total vote.
Trump received 62,979,636 votes, or 46.1%
the difference is 1% and i think this is a quite perfect forecast under such circumstances.
am i right or not ?