This post has been updated with Trump's tweets.
And Trump himself on Sunday night reiterated that he preferred the popular-vote method: “I’m not going to change my mind just because I won. But I would rather see it where you went with simple votes.”
The reactions to that comment were similar — that Trump would have lost the election under his own preferred format!
I totally agree with Donald Trump that the popular vote winner should determine the Presidency.— Neera Tanden (@neeratanden) November 14, 2016
Except, it's impossible to know that for sure.
The fact is that just because Trump lost the popular vote last week doesn't mean he would have lost a popular-vote election.
What do I mean by that? Basically, losing the popular vote in an electoral-college election isn't the same as losing the popular vote in a popular-vote election. The former involves a very specific strategy that may cost you when it comes to winning the nationwide popular vote. But you pursue that strategy because the latter doesn't matter. You need to get to 270 electoral votes, not a majority or plurality of all votes.
An electoral-college election involves making explicit appeals to and advertising in around 10 or 12 out of the 50 states. It means Trump didn't campaign or advertise in California or Massachusetts or Washington state and that Clinton didn't campaign in Oklahoma or even Texas (despite polling within single digits there). They knew it would be wasted effort to try to turn a 30-point loss in those states into a 22-point loss, or a 14-point loss into an eight-point loss.
It also means they tailored their messages specifically to voters in key states, which happen to be disproportionately Midwestern and on the East Coast. It meant Trump had little incentive to appeal to Western voters — outside Nevada, at least. And he pursued a very specific strategy that appealed to the key Rust Belt states that wound up delivering him the presidency.
After this post went up Monday, Trump made the same argument himself in a pair of tweets Tuesday.
Trump is taking some liberties here. He has made his preference for the popular vote clear — even on "60 Minutes” this weekend — and it's not clear he would have done that much better in big states if it was a popular-vote election, as he asserts. But it's a valid question to ask.
To be clear, it seems more likely than not that he probably would have lost a race in which the popular vote and not the electoral college decided the president. After all, he trails in the popular vote by 0.8 points, and that number is growing as results from California and other coastal states roll in. Clinton's margin is likely to grow from here — to as much as 2 full points, according to some estimates.
Would his message have been significantly different in a truly national election, in which he had to appeal to moderate Republicans in Orange County along with conservative former Democrats in coal country? Maybe. Maybe not.
But California is a telling example. In fact, it's Case Study No. 1 in why Trump might lose the national popular vote by as much as he is.
Numbers tallied by the Cook Political Report's David Wasserman show Trump is trailing in that state 61.5 percent to 33.2 percent. As things stand, that 28-point Democratic win is the biggest since FDR in 1936 — bigger even than Mitt Romney's 23-point loss in 2012. It's a bloodbath, and even as Trump did better than Romney, he's doing worse in California.
There are more than 4 million ballots left to count, but Clinton is already netting nearly 3 million votes in California alone — a number that is bigger than her overall 728,000-vote lead in the nationwide popular vote.
Possibly contributing to Trump's poor performance in the state is the fact that the only other statewide race -- an open Senate contest -- featured two Democrats running against one another after both advanced under California's unusual "top-two" primary system.
But if Trump simply would have taken 37 percent of the vote in California — the same as Romney took in 2012 — he would be about tied with Clinton in the national popular vote. And that's just an adjustment of about four points in one inconsequential (albeit huge) presidential state.
Those were votes, of course, that neither he nor his campaign needed, and so they didn't even try to get them. Had he and Clinton both sought them and made direct appeals to Californians, perhaps the margin would have been similar. But we'll never know for sure.
What we do know, though, is that less-competitive states tend to have lower turnout. According to the United States Election Project, several swing states ranked in the top 10 in turnout: New Hampshire, Iowa, Wisconsin, Colorado, Virginia and Florida. California ranks 44th in turnout. Only an estimated 53 percent of its eligible voters have had their ballots counted (including the four-plus million that haven't been tallied). Making states like California consequential seems likely to draw a significantly different — or at least significantly expanded — electorate.
Trump's deficit is growing and could soon be historically large for the winner of the electoral college. But it's simply impossible to say for sure that it means he would have lost a popular-vote election.