I am almost exclusively a consumer of mainstream media. I don’t have time (or interest) to read that widely in disparate sources. I don’t have a cable television subscription. I read my local paper, the New York Times, the Wall Street Journal, the Economist, the New Yorker, and sometimes The Atlantic. I appreciate fivethirtyeight.com and read it occasionally. BBC.co.uk, too. I listen to NPR and BBC radio news programs. I watch local news TV broadcasts, but almost never national ones. I don’t particularly value the commentary, editorials or op-eds of any of the foregoing.
I have no idea what @awcntdb is talking about. I highly value the sources I cited because, except for the Economist, the news they report is invariably based on knowledgeable reporters going someplace and reporting confirmable facts, and also indicating the nature of their sources. I don’t mind when reporters have a point of view that they have developed by actually knowing what they are talking about and being where it is happening. I do mind when they fail to report contrary positions and the reasons for them, but that doesn’t usually happen in the mainstream media, at least the print version, and when it does it gets corrected fast.
Anyway, based on my experience, absolutely none of what @awcntdb says passes the smell test. There’s no question the mainstream media was anti-Trump. But it was anti-Trump for the understandable reason that Trump appears to have contempt for everything it stands for, starting with the very notion of accuracy, facts, and expertise, not to mention respect for opposing views and not letting what you wish were true affect what you report as true. No reader of the mainstream media could have failed to miss Trump’s narrative and that of his supporters about how his victory was possible, and why all of the polls were wrong.
Fivethirtyeight contains extensive, sometimes exhausting discussions of poll methodology, if you look past the pretty graphs. It did about as good a job as I can imagine explaining in advance why polls might not be doing an accurate job figuring out who was actually likely to vote, while also explaining that on the whole traditional assumptions were more likely to be accurate.