Moderate Democrats are worried about one thing, reports The New York Times: defeating Trump. They're fretting that Joe Biden can't seem to run a proper campaign, and Elizabeth Warren might be weak in the general, and Pete Buttigieg might not get minority support. "Would Hillary Clinton get in, the contributors wondered, and how about Michael R. Bloomberg, the former New York mayor?" writes Jonathan Martin. Or maybe even Michelle Obama?

It's not hard to see what moderates are concerned about, but it's got nothing to do with beating Trump per se. It is all about advancing their own political agenda, and maintaining their comfortable grip on the Democratic Party's levers of power, as it always has been.

For nearly five decades, it has been a truism among Democratic grandees that The Left Cannot Win, based mostly on one data point. When George McGovern got stomped by Richard Nixon in 1972, it was taken as proof for an entire generation of politicians that a leftist political candidate is electoral poison, and so must be prevented from trying.

But this is garbage history. For one thing, as New York's Ed Kilgore writes, the 1972 election was not lost because McGovern was some fire-breathing Leninist radical. On the contrary, while he was on the party's left flank, he had been a loyal Democrat for his entire career, building up a party apparatus virtually from scratch in South Dakota and serving in the Kennedy administration. Nixon won mainly because he was an incumbent and the economy was extremely strong, because the civil rights movement had dynamited the New Deal coalition, and because his undercover campaign of dirty tricks worked quite well at dividing the Democrats. (People sometimes forget that the Watergate scandal was touched off by a botched burglary of the Democratic National Committee headquarters to replace some malfunctioning wiretaps.) It is virtually beyond question that any Democrat who went up against Nixon would have lost.

For another, the composition of the electorate has changed dramatically over time. An 18-year-old first-time voter in 1972 would be 65 years old today. Back in 1970 about 83 percent of the population was non-Hispanic white; today it is about 60 percent. It is senseless to pretend as though one highly-contingent election is anything more than a hazy demonstration of political bedrock realities, especially almost 50 years after the fact.

But more importantly, McGovern was not the last Democratic presidential nominee to lose an election. They also lost in 1980, 1984, 1988, 2000, 2004, and 2016. In almost all of these elections, the candidate was on the party's center or right: Jimmy Carter started the neoliberal deregulation trend with airlines and trucking as president from 1977-1980, Walter Mondale ran on austerity and balancing the budget in 1984 (and incidentally lost almost as badly as McGovern did), Al Gore was a centrist New Democrat who ran alongside the pathological warmonger Joe Lieberman in 2000, John Kerry defeated the more progressive Howard Dean in 2004, and Hillary Clinton of course bested Bernie Sanders in 2016. (To be fair, both Gore and Clinton won the popular vote, and Gore would likely have become president if not for naked political interference from the Supreme Court.)

Only Michael Dukakis, who defeated centrists Al Gore and Gary Hart and went on to lose to George H. W. Bush in 1988, was anywhere close to the party's left flank. Even then he still defeated Jesse Jackson in the primary, who was briefly tied with Dukakis on the strength of his "Rainbow Coalition" messaging about building up a cross-racial working class movement — which as Ryan Grim writes in his book We've Got People, inspired wild panic among the Democratic party elite, who professed absolute certainty that a left-wing black man could not possibly win.

The political function of "electability" is well demonstrated by the response to the consecutive losses in 1980, 1984, and 1988. Rather than discuss the possibility that perhaps the party should have gambled on Jackson — maybe his Rainbow Coalition could have paid dividends, and after all he couldn't have done worse than Dukakis — the party put up yet another moderate in the form of Bill Clinton, who finally won. And just as the McGovern loss was taken as proof positive that the left cannot win, the Clinton victories were taken as proof positive that only moderates can win — conveniently memory-holing the previous three consecutive losses.

Rahm Emanuel was pushing this line after 2006 as the head of the Democratic Congressional Campaign Committee, and claimed triumphant victory when his handpicked pro-Iraq War moderates won multiple close elections in the wave election that year. But as Grim writes, he conveniently elided that when lefty Democrats won primaries and Emanuel cut them off from money and support, several of them went on to win, while several of his moderates lost — and almost all of the rest were wiped out in 2010.

Of course, candidate quality does matter, as does campaigning. But there is always a great deal of slack in the political system — room to push one's particular politics and move popular opinion. At bottom political elites always have their own perspectives, and do not neutrally sample public opinion for their policy views, if such a thing were even possible. On the contrary, as Matt Stoller writes, "Parties don't poll for good ideas, run races on them, and then govern. They have ideas, poll to find out how to sell those ideas, and run races and recruit candidates based on the polling."

The conservative movement's perspective on electability is a lot more sensible. Their avatar Barry Goldwater got stomped in 1964 by 22.6 percentage points. But instead of concluding that America was an unalterably center-left nation, and accommodating themselves to civil rights and the welfare state, they kept pushing and pushing, and eventually got their guy in the White House in 1980.

Which brings us back to the 2020 primary. During the presidency of a reality TV host, I have little confidence in anybody's ability to predict anything. It's a time of unusual political chaos, with an unprecedented incumbent and only a few dozen prior data points in any case. But I am very confident that none of the Democratic power brokers currently looking for a moderate dark horse to save them from a Warren or Sanders presidency are honestly concerned about electability. They are looking down the barrel of candidates who would try to sharply raise taxes on the rich, and disrupt the comfortable revolving door between the party and the private sector. If one's C-suite chair is sufficiently padded, the prospect of President Sanders is just as bad as a second Trump term, if not worse.

Editor's note: This article initially misstated the outcome of the 1988 presidential election. It has since been corrected. We regret the error.

Want more essential commentary and analysis like this delivered straight to your inbox? Sign up for The Week's "Today's best articles" newsletter here.