If you look into this closer, you realize the poll was garbage. If you ask a question in 2018 with no context in the form; "Would support or oppose 'a policy of Medicare for All'"? most respondents are going to think if you're asking if everyone should be allowed to have access to Medicare when you're 65.

When a poll asks about "a national health plan, sometimes called Medicare-for-all, in which all Americans would get their insurance from a single government plan," only 23% of Republicans were in favor of it. [1]

When you explain that this plan would not only replace private insurance we have today, but in fact make such insurance illegal, Republican support drops by ~50%.

[1] - https://www.politifact.com/factchecks/2019/jul/02/jess-mcint...

And if you phrase it as "are you in favour of a healthcare system where you are forced to pay for other people's healthcare, without any opt-out, the government decides if you get treatment, and price-fixing the cost of drugs resulting in reduced pharmaceutical profits thus threatening the economy" you would see support drop even lower. Now, you could argue that phrasing is technically accurate but it's clearly biased to get respondents to answer in a particular way.

The way you phrase questions in polls matters -- phrasing it as "making private healthcare illegal" is placing a bias in the question (if it was phrased as "giving everyone the same excellent standard of healthcare as provided by Medicare" it would also be a bias but in the opposite direction).

> most respondents are going to think if you're asking if everyone should be allowed to have access to Medicare when you're 65.

In 2018, a very large number of Americans understood what "Medicare For All" specifically referred to. Sanders had been campaigning on it for more than 3 years at that point (it was one of the major topics of debate in the 2015 Democratic primary), legislation had been drafted and so on. I don't know why someone would think "Medicare for All" would mean anything other than "Medicare for all [Americans]." The vast majority of Americans over 65 are already entitled to Medicare in the US (unless you only recently became a citizen or haven't paid Medicare taxes for the past 10 years, and don't have a disability).

So what? Why does that statistic mean anything in a highly fragmented contest? Bernie got 26% of the vote in Iowa, about the same in New Hampshire and just shy of 50% in Nevada. So in all three states more people voted against Bernie than for him.

Why does that statistic not mean something in a highly fragmented contest? The 1972 and 1976 Democratic primaries both started with 16 candidates; the 1996 Republican primaries had 12 and the 2000 Republican primaries had 13. Being the only candidate in 11 cycles to win the popular vote in the first three primaries/caucuses isn't an automatic ticket to the White House, no, but it's not nothing.

Assuming Sanders is even a close second in South Carolina, he's going to have tremendous momentum going into Super Tuesday, and historically it's unwise to write that off.

Canadian here, more of our elections nationwide are won with less than 50% of the vote than with more (last federal election had 124 candidates with >50%, of 338 seats).

Social democrats are staunch believers in liberal democratic polity and a capitalist-oriented economy, both of which are antithetical in _any_ socialist system, even democratic socialism.

American socialists will lie to you otherwise, or perhaps they just don't know.

Moreover, Bernie has half a century of history praising communism and kissing up to Marxist-Leninists whereas this softer, friendlier, more populist Bernie is less then a decade old and won't address his endorsements. I don't believe it for a second.