How to Manufacture a Popular-Vote Victory

The Clinton campaign, which is losing the pledged delegate race, is now talking up a different metric: the cumulative popular vote.

“I’m very proud that as of today, I have received more votes by the people who have voted than anyone else,” Hillary Clinton said on Wednesday, a day after she won the Pennsylvania primary by more than 200,000 votes.

Her characterization is true only in a highly technical way: If you count the votes she received in Michigan (where hers was the only name on the ballot) and Florida (where an outlaw primary was held in January), and if you ignore a series of caucus states where hundreds of thousands of Democrats participated but no official popular vote tally was kept, then yes, she has received more votes than Barack Obama.

By any other reasonable standard, she has not, and most likely will not. This is highly significant because, even though the popular vote is an officially meaningless metric, the entire premise of Clinton’s longshot campaign now rests on winning it.

Since she cannot win the pledged delegate race, it is the only plausible way in which she can still lay a moral claim to the nomination when the primary season ends. With a popular vote victory, she can ratchet up the pressure on superdelegates to side with her, arguing that they’d be voiding the “will of the people” if they didn’t.

As her spin on Wednesday indicated, there are several variables at work that allow for differing calculations: Do you count Florida and Michigan? Or just Florida (where at least Obama’s name was on the ballot)? Or do you give Obama the “uncommitted” vote from Michigan? And do you add in vote estimates from Iowa, Nevada, Maine and Washington state, where popular vote tallies weren’t kept during caucuses?

Let’s go through it one step at a time, starting with the current baseline popular-vote total (in other words, not including anything from Florida, Michigan or the four caucus states that didn’t tally popular votes):

Obama: 14,447,566

Clinton: 13,965,162

Now, let’s project the remaining primary contests:


Turnout estimate: 905,000

Clinton: 479,650 (53%)

Obama: 425,350 (47%)

North Carolina

Turnout estimate: 1,750,000

Clinton: 770,000 (44%)

Obama: 980,000 (56%)

West Virginia

Turnout estimate: 445,000

Clinton: 271,450 (61%)

Obama:173,550 (39%)


Turnout estimate: 540,000

Clinton: 329,400 (61%)

Obama: 210,600 (39%)


Turnout estimate: 750,000

Clinton: 330,000 (44%)

Obama: 420,000 (56%)

Puerto Rico

Turnout estimate: 1,000,000

Clinton: 600,000 (60%)

Obama: 400,000 (40%)

South Dakota

Turnout estimate: 130,000

Clinton: 54,600 (42%)

Obama: 75,400 (58%)


Turnout estimate:160,000

Clinton:64,000 (40%)

Obama: 96,000 (60%)

Add these all together, and here is a reasonable projection of the popular vote total for all of the remaining states and Puerto Rico:

Obama: 2,780,900

Clinton: 2,899,100

That’s a gain of 118,200 votes, for Clinton, owing entirely to her projected landslide win in a high-turnout primary in the Commonwealth of Puerto Rico. Without Florida and Michigan, that would make the end baseline popular-vote total Obama 17,228,466, Clinton 16,864,292. That’s a margin of 364,174 votes for Obama.

Now, here’s where the math gets funny. The Clinton campaign, should it reach this point, will insist that Michigan and Florida be factored into this popular vote total. She won Florida by 294,772 votes and received 328,309 votes in Michigan. Mix both of these in, and she would indeed edge into the popular vote lead by 258,907 votes.

But the arguments against doing this are well-known and hard to refute: No one campaigned in either state, turnout was much lower than in other states (and, indeed, much lower than in each state’s G.O.P. contests – something that didn’t happen anywhere else), and – most glaringly – Obama’s name wasn’t even on the ballot in Michigan.

If you’re going to count Michigan and Florida, Obama’s side will argue, you’d better also consider the four caucus states (Iowa, Nevada, Maine and Washington) where hundreds of thousands of Democrats participated but where no official popular vote tally was maintained. Obama won three of those four states, losing narrowly in Nevada. Add reasonable and generally accepted estimates of each candidate’s performance in these caucus states together and Obama gains 334,084 votes to Clinton’s 223,862 – a difference of 110,222 votes, severely cutting into the lead that Clinton enjoys when Florida and her share of Michigan are factored in.

But will it even come to that?

Whereas in Florida both candidates’ names were on the ballot and the result was probably a fair representation of what the margin would have been under normal circumstances, Michigan is different. It’s unlikely that superdelegates – or the media, or anyone who isn’t already a hard-core Clinton supporter – would seriously accept the notion that Clinton should be credited with 330,000 bonus votes (those that she won in Michigan’s outlaw primary) and Obama with zero. In fact, polls show, there’s a very reasonable chance that, had a re-vote taken place, Obama would have won the Michigan primary outright.

So if Florida were counted, but not Michigan, and not the uncounted caucus vote, Obama would still be ahead by 69,402 votes. Add to this his uncounted caucus vote — it’s hard to see how you include Florida but exclude the caucus estimates — and he leads by 179,624 votes.

As for Michigan, the most compelling argument is simply to discount the state’s primary results completely. But, for the sake of argument, if you’re intent on including them, Obama ought to be credited — at the very least — with the votes that “uncommitted” received in January, many of which were intended for him anyway. It’s true that some weren’t, but it’s also true that, has his name been on the ballot, Obama would have received even more votes than uncommitted did in the primary. So awarding him the uncommitted total is actually a conservative estimate of his support in the state. As was mentioned above, he very likely would have won the state outright had there been a legitimate primary. But credit those 238,168 uncommitted votes to Obama, and his overall popular vote lead — this includes Florida and the caucus states — stands at 89,483.

The point is that under the most basic and probably the fairest criteria — simply counting every state and U.S. possession where there was a legitimate primary or a caucus where popular votes were tallied — Obama will finish the primary season hundreds of thousands of votes ahead of Clinton. Considering that an estimated 36 million primary votes may ultimately be cast, Obama’s advantage, obviously, won’t be huge. But it will still be significant: You’d have to forecast some extraordinarily good results (putting it mildly) in the remaining primaries for Clinton to overtake Obama.

And even counting the extra caucus states, Florida and Michigan (with uncommitted going to Obama), Obama will still very likely lead Clinton at the end of the process, though by a more narrow margin. The only way she will be able to claim a popular vote win would be through very selective, self-serving criteria: counting Florida and only her votes in Michigan (as she did in declaring herself the popular vote leader this week).

No doubt, if they reach June, the Clinton campaign will declare a popular-vote win anyway. But will anyone believe it?

How to Manufacture a Popular-Vote Victory