Ok, I see. You want me to do the work, which I will not.
National polls are skewed because in some of the big states with large democratic congressional races, republicans don't even think is worth voting. I would argue that if national votes counted towards the presidential race, and seeing that the state polls under-counted republicans, that Trump could have won the national vote. Anyways, the point is that the national numbers do not matter in a presidential election. I don't know why you even bring it up to infer that Trump did not really win.
The NYT predicted that Hillary would win 322 electoral votes; she only got 232. Big mistake.
The NYT predicted that the democrats would take the senate.
The following nonsense is published as a matter of fact. Typical left wing bullsh*t, you don't know what you are talking about and I do.
"The Upshot’s elections model suggests that Hillary Clinton is favored to win the presidency, based on the latest state and national polls. A victory by Mr. Trump remains possible: Mrs. Clinton’s chance of losing is about the same as the probability that an N.F.L. kicker misses a 37-yard field goal."
Now you expect me to believe any numbers they published? They predicted Trump would get 42.8% of the votes with 2.5% margin of error. He got 46.4%.
Hillary was predicted at 45.9% and got 48.5%. They got both wrong.
If you think this is bad, it gets worse. Sam Wang, a scientists from Princeton, predicted the chance of Clinton winning is greater than 99%. He actually claimed that state polling are the most accurate and that is the only polls they use.