I'm a little frustrated with some of the responses here because they're either a generous analysis that assumes people didn't like her for reasons they didn't care really care about or that speculate in ways that are a little generous to the left.

Hillary has been a boogeywoman for the right since she was First Lady. Mostly it was an issue with Woman in Power, but also Woman Not Knowing Her Place. Republican First Ladies tend to be little more than alcoholic hood ornaments with opinions on reading or drugs, and the last strong First Lady the US had was Eleanor Roosevelt. So for illegitimate and misogynist reasons, the right loathes Hillary Clinton.

She's also one of the most unpopular politicians in the US, which was true in 2016 and is true now. 2016 was a race to the bottom, because both she and Trump were largely despised. The question was whose camp hated who more. And you might think that because Trump won that the answer is that America hates Hillary more, but that's not necessarily telling the whole story.

You probably know what the Electoral College is but if not, it's our stupid fucking antidemocratic mechanism to keep Republicans winning office in the 21st century. Hillary very much won the popular vote, but Trump won the majority vote in the right states and took their electors, all or nothing, so Hillary lost.

Trump was promoted by Hillary's campaign early on via Operation Pied Piper, which was where they promoted unpopular candidates in an attempt to fracture the RNC. Unfortunately, this proved to be a stupid strategy because after Obama largely failed to alleviate the crushing effects of the Great Recession of 2008, people wanted a change. The change candidate, especially running on an anti-corruption platform, proved to be very effective messaging for the Tea Party right.

Now here's the part I see the left skip over a lot, to their peril.

Hillary fucked up 2016 based on an attempt to run a metrics-based campaign. Her manager, Robby Mook, is a technology fetishist and he seemed to earnestly believe you could win the presidency on the back of software. So the Clinton campaign had this program called Ada (for Ada Lovelace, and a massive misappropriation of her fucking name) that took national and state level polls that told Hillary where to go next. As a person who works in tech with data science, you never do this based on a principle called GIGO--garbage in, garbage out. One thing polling sucked at was the Midwest. States that Hillary would need to win to carry her over the Electoral College line but that she ended up losing. Because the algorithm told her things were fine, and she could safely ignore those states, but again, garbage in, garbage out. Bad polls means bad data which means bad decisioning if you don't analyze the data itself.

Purportedly, even Bill Clinton tried to suggest they campaign more in the Midwest (Hillary famously didn't step foot in Wisconsin) but was overruled by the dorks worshipping at the false god of data science.

Hillary ended up losing by tiny margins in states she needed to win, and that fault is directly on the technocrats who relied too heavily on statistics. Which is very fitting, I think, since liberals seem to adulate dorks like the Freakonomics assholes and Malcolm Gladwell and Stephen Pinker, all men who misinterpret data and trust bad faith statistics.

So that's how Hillary blew it. The left has had its own fanciful spin that I think is a too generous interpretation of why she lost, as her platform was very neoliberal and ghastly, but she had a decent enough nationwide turnout. In fact, if you had an algorithm telling you that you were going to win in 2016 but you failed to compensate properly for picking up states via the Electoral College due to some of those states having bad polling, you might have thought you were going to win up until the very bitter end.