On Tuesday night, Donald Trump's campaign sent out an email to reporters headlined "A TALE OF TWO RALLIES: SEPTEMBER 27, 2016."

The first rally was one Trump held Tuesday in Melbourne, Fla., that drew 15,000 people with, according to the campaign, thousands waiting outside. It looked like this.

The second rally was a Hillary Clinton event, also Tuesday, in Raleigh, N.C., that had a reported crowd of 1,400. The release also pointed to this image of the Clinton crowd.

The message — in case you somehow missed it — is simple: Lots of people want to see Trump. Far fewer want to see Clinton. Crowds = passion. No matter what the polls say then, Trump is going to win.

This look-at-the-size-of-our-crowds argument has been the standard pushback by Trump and his surrogates against any negative news for a while now. "We go to Oklahoma, we have 25,000 people, Trump said last month in Florida. "We had 21,000 people in Dallas, we had 35,000 people in Mobile, Alabama. We get these massive crowds. Look, if [Clinton] had 500 people I would be surprised."

Trump campaign manager Kellyanne Conway has been singing the same song. “I mean, Bernie Sanders had an event yesterday in Ohio for Hillary Clinton,” she said earlier this month. “There were 150 people there. That’s like a second wedding where I come from.”

There are two reasons to be somewhat skeptical of this argument.

The first is that it has long been Trump's stated strategy to do large-scale rallies. He enjoys them and believes they were at the heart of his success in the Republican primary race. "My best investment is my rallies," Trump told the Associated Press in May. "The people go home, they tell their friends they loved it. It's been good." In that same interview, Trump was broadly dismissive of the sort of data collection operations that have become the hallmark of modern campaigns. "I've always felt it was overrated," he said. "Obama got the votes much more so than his data processing machine. And I think the same is true with me."

Trump is trying to get massive crowds. So, no one should be all that surprised when he gets them. Clinton, on the other hand, has done very few massive rallies either in her primary victory over Sanders or as a general-election candidate. That, too, is a strategy. Whether it's because Clinton is more comfortable with 1,400 than 10,000 or because her campaign has simply decided that smaller crowds are more helpful for their organizing and get out the vote goals is hard to know.

The point is that the campaigns have long been pursuing different strategies when it comes to rallying their supporters. You can believe Trump's big crowds theory is superior but you shouldn't assume Clinton is operating under that same theory of the case.

Second, there's very limited evidence to suggest that large crowds translate into large vote totals — particularly in a general election.

Start with this: 125 million votes were cast for president in 2012. In Florida alone more than 8.4 million people voted. Assume, for the sake of argument, that Trump had 27,000 people in Florida on Tuesday to see him. If every one of those people not only voted but voted for him that would amount to .6 percent of the 4.1 million votes that Mitt Romney got in the state in 2012.

And, speaking of Romney, remember how he and his campaign team were utterly convinced that he was going to win in 2012? That conclusion was, at least in part, based on the fact that at every rally he held in the closing weeks of the campaign, thousands of people turned out to show support for him. "Mitt drawing larger crowds" read a headline from Politico in October 2012; the report read, in part:

Since his strong presidential debate performance last Wednesday night, Romney has seen a bump in the number of people attending his rallies, which the campaign calls a sign of new enthusiasm in the final month of the campaign.

In the past week alone, Romney’s campaign says at least three of its rallies have, per the campaign’s crowd counts, exceeded 10,000 people: an Oct. 4 event with country singer Trace Adkins in Fishersville, Va., which was Romney’s largest event ever at 14,000 people; a rally last Sunday in Port St. Lucie, Fla., that drew 12,000; and one in Cuyahoga Falls, Ohio, that fire marshals estimated also drew 12,000.

As you may recall, Romney lost that race. He lost Virginia, Florida and Ohio — all states where Politico reported massive crowds were turning out to see him.

The problem with conflating crowd size or energy with momentum or a "hidden vote that the polls aren't catching" is the sheer size of the electorate in a presidential general election. Trump's crowd sizes, arguably, were more telling in the primary because 25,000 people in a Republican primary is a significant chunk in many states. It's not even a drop in the bucket in most swing states in a general election; in the smallest of the swingiest states — New Hampshire — more than 700,000 people voted for president in 2012.

Even Trump, who rarely admits a strategic error, wondered aloud about whether his focus on big crowds might be a mistake.

“I hear we’re leading Florida by a bit,” Trump said in that same speech where he hit Clinton's crowds. “I don’t know why we’re not leading by a lot. Maybe crowds don’t make the difference.”

Maybe not.