Facebook Twitter Tumblr Close Skip to main content
A Project of The Annenberg Public Policy Center

The Facts on Crowd Size

White House counselor Kellyanne Conway claimed that “alternative facts” were employed by Press Secretary Sean Spicer when he tried to make the case that “this was the largest audience to ever witness an inauguration, period, both in person and around the globe.” Actually, Spicer got several facts wrong.

He has since admitted that some of the figures he used were incorrect — though he believed them to be accurate at the time.

The crowd-size controversy began on Saturday, Jan. 21, a day after the inauguration, when President Donald Trump claimed the media had misrepresented the number of people attending his inauguration. Trump spoke at CIA headquarters and said that “one of the networks” had shown “an empty field,” while he saw a crowd that “looked like a million-and-a-half people” and “went all the way back to the Washington Monument.”

Trump claimed that “we caught [the media] in a beauty, and I think they’re going to pay a big price.”

Spicer then read a prepared statement later that day, further criticizing the media for “dishonesty.”

Spicer, Jan. 21: Secondly, photographs of the inaugural proceedings were intentionally framed in a way, in one particular tweet, to minimize the enormous support that had gathered on the National Mall. … We know that from the platform where the President was sworn in, to 4th Street, it holds about 250,000 people. From 4th Street to the media tent is about another 220,000. And from the media tent to the Washington Monument, another 250,000 people. All of this space was full when the President took the Oath of Office. We know that 420,000 people used the D.C. Metro public transit yesterday, which actually compares to 317,000 that used it for President Obama’s last inaugural. This was the largest audience to ever witness an inauguration — period — both in person and around the globe.

Photos of the crowd, including one taken at 12:01 p.m. from the top of the Washington Monument, clearly show that the crowd witnessing Trump’s inauguration didn’t extend all the way to the monument. Metro’s figures for both 11 a.m., a half-hour before the inauguration ceremony began, and for the full day show fewer trips taken this year than for past inaugurations.

At a Jan. 23 press conference, Spicer made an alternative argument, saying the number of people watching the inauguration on TV, online and in person had to be the largest ever.

“I have a right to say if you add up the network streaming numbers, Facebook, YouTube, all of the various live streaming that we have information on so far, I don’t think there’s any question it was the largest watched inauguration, ever,” Spicer said.

On that point, Spicer may be correct. While Nielsen TV ratings were higher in past years, some data suggests online viewership was up compared with 2009. But there is no comprehensive measurement available that would prove or disprove this claim.

Facts Versus the ‘Alternative’

Several media outlets wrote about Spicer’s claims about the crowd size this year versus Obama’s inaugurations. And there are several side-by-side photo comparisons of the crowd in 2017 and the visibly larger crowd in 2009 — a historic event as the nation inaugurated its first black president.

On NBC’s “Meet the Press,” host Chuck Todd asked White House counselor Kellyanne Conway, “[W]hy the president asked the White House press secretary to come out in front of the podium for the first time and utter a falsehood?” Conway responded: “Don’t be so overly dramatic about it, Chuck. What — You’re saying it’s a falsehood. And they’re giving Sean Spicer, our press secretary, gave alternative facts to that.”

To be clear, there are not “alternative facts.” There are certainly alternative interpretations of the facts, or alternative data sets, or, in this case, potentially different crowd-size estimates from experts. But a “fact,” in the words of Merriam-Webster is “something that has actual existence” or “a piece of information presented as having objective reality.”

An alternative to that, as Todd pointed out, is a falsehood.

Let’s go through the evidence on Spicer’s and Trump’s crowd-size claims.

Photos. Side-by-side photos of the 2017 inauguration crowds and 2009 inauguration crowds on the National Mall show a noticeable difference, with 2009’s crowds filling up more of the space, from the perspective of the Washington Monument looking toward the Capitol.

See Reuters’ side-by-side photos here, both pictures taken by the news agency. The Jan. 20, 2017, photo was taken at 12:01 p.m., the Reuters photo caption says. Trump took the oath of office at noon.

Reuters editor Jim Bourg posted on Facebook that he was the one who assigned the photographer to take pictures from the top of the Washington Monument. Bourg wrote that he had seen “a lot of inaccurate talk and allegations online about the crowd photos from Friday’s Trump inauguration that are simply not borne out by the FACTS.”

“Only one news organization had a still photographer atop the Washington monument for the inauguration and I assigned him to be there,” Bourg wrote on Jan. 22. “This photo by Reuters News Pictures staff photographer Lucas Jackson was taken at 12:01:18 p.m. on Friday and not much earlier as many people are trying to claim.”

The New York Times tweeted photos from Getty Images in 2009 and the 58th Presidential Inaugural Committee for 2017, writing that both were taken 45 minutes before the respective presidents’ swearing-in.

PBS’ “NewsHour” also posted a timelapse video of the crowd throughout Inauguration Day, which shows it didn’t quite stretch back to the Washington Monument, as Trump and Spicer said. And as our fact-checking colleagues at the Washington Post pointed out, other journalists had more photographic evidence on Twitter.


Metro ridership. At the press conference, Spicer also gave inaccurate numbers for those taking the Washington, D.C., Metro system for Trump’s inauguration compared with Obama’s.

Spicer, Jan. 21: We know that 420,000 people used the D.C. Metro public transit yesterday, which actually compares to 317,000 that used it for President Obama’s last inaugural.

But that’s not what Metro said, and Spicer acknowledged that during his Jan. 23 press conference.

Spicer, Jan. 23: At the time the information that I was provided by the inaugural committee came from an outside agency that we reported on. And I think knowing what we know now we can tell that WMATA’s numbers are different, but we were trying to provide numbers that we had been provided. That wasn’t like we made them up out of thin air.

WMATA is the Washington Metropolitan Area Transit Authority, also referred to simply as Metro.

Spicer’s 317,000 figure was the number Metro gave for riders as of 11 a.m. on Inauguration Day 2013. The comparable number, for 11 a.m., for Trump’s inauguration was 193,000.

And 2009’s Metro ridership as of 11 a.m. was much higher: 513,000. Even in 2005, for President George W. Bush’s second inauguration, the ridership figure was 197,000, a bit higher than the 2017 number.

(Note, Metro corrected the 2013 date: Inauguration that year was held on Jan. 21.)

The comparable full-day figures from Metro, as reported by the Washington Post, showed 570,557 trips taken from 4 a.m. Friday until midnight. Full-day numbers for the past two inaugurations were also higher: 782,000 for 2013 and 1.1 million for 2009, the busiest day in Metro’s history.

The Women’s March on Washington held on Saturday, Jan. 21 was the second-busiest day in Metro’s history, the Post‘s Dr. Gridlock transportation reporter also wrote. Metro General Manager Paul J. Wiedefeld told the Post that 1,001,613 trips were taken Saturday.

Crowd-size experts. Spicer was right about one thing: It appears, as he said, that no one has scientifically calculated estimates of the crowd size for Trump’s inauguration. Congress has prohibited the National Park Service from making official estimates after it produced an official estimate that 460,000 turned out for the so-called “Million Man March” in 1995, prompting organizers to threaten a lawsuit.

Estimating crowd size is a notoriously difficult task in Washington, D.C., for a variety of others reasons, explained Steve Doig, a professor at Arizona State University who analyzed the crowd size of Obama’s 2009 inauguration. To get a really good crowd estimate, he said, you need a true overhead image. In Washington, D.C., the FAA makes certain areas no-fly zones during the inauguration, and there are few very tall buildings.

Satellite photos would provide the best evidence, he said, but because of the overcast skies on Friday, the low cloud cover did not allow for useful satellite images.

The top of the Washington Monument provides an oblique view, Doig said, though it is not a true overhead shot. Still, he said, looking at side-by-side photos from the 2009 Obama inauguration and Trump’s, that vantage point makes it clear that the crowd was much smaller for Trump’s inauguration.

“You can clearly see that Obama’s [crowd] was substantially higher,” Doig said. “I’d be happy saying three times larger.”

The New York Times reported that Keith Still, a professor at Manchester Metropolitan University in England and a crowd safety consultant, estimated that the crowd for this year’s inauguration was about a third of the size of the crowd in 2009.

Spicer argued that floor coverings over the grass on the National Mall “highlight[ed] where people were not standing, while in years past the grass eliminated this visual.”

Doig called that argument absurd. Yes, he said, grass covering was not used in 2009, but he said the blank spaces would still be readily apparent.

“I don’t know how [Spicer] kept a straight face when he said that,” Doig said.

If you look at the photo of the Obama inauguration, he said, unoccupied areas would have shown up as green or brown (rather than white), but he said they would have been easily visible.

“It’s not as if everyone was wearing camouflage,” Doig said.

As for Trump’s claim that when he looked out, he could see people “all the way back to the Washington Monument,” that’s entirely possible, Doig said, given Trump’s low vantage point.

There were reports that demonstrators may have been blocking people from getting onto the Mall, Doig said. And, he said, there were a couple of “mini entry points” where it appears demonstrators were trying to block entry.

“Perhaps as many as hundreds may have been delayed getting in,” Doig said, “but not hundreds of thousands.”

Back in 2009, Doig did an analysis of the crowd that attended the Obama inauguration using a satellite image taken on a clear day. The image was then broken into grids to make counts. Doig’s estimate of 800,000 was far lower than the 1.8 million figure reported by the Washington Post, based on counts by Washington, D.C., officials. The higher estimate included people along the parade route and estimates of people watching from windows, Doig said, while his did not.

Such an estimate will not be possible this time due to the lack of useful satellite photos, Doig said. But for the purposes of fact-checking, the exact number is less relevant than whether or not it was the largest crowd ever for an inauguration. And Doig said the photo from atop the Washington Monument –while less than an ideal overhead photo — provides enough evidence to make clear that Obama’s first inauguration was far more heavily attended.

TV and online viewership. But at the Jan. 23 press conference, Spicer emphasized the combined audience — in person, watching on TV and watching online — was the highest in history. Asked if he was now saying that it was the largest in-person crowd for an inauguration, Spicer said, “I am not. I am saying that it was the total largest audience witnessed in person and around the globe.”

According to Nielsen, TV viewership for the Trump inauguration — an estimated 31 million people — was about 19 percent below the number that tuned in to watch Obama’s first inauguration in 2009. And it is far below the 41.8 million who tuned in for Ronald Reagan’s first inauguration in 1981.

But that only tells part of the story, as Spicer highlighted at his Jan. 23 press briefing. If you add in those who watched it online via livestream, Spicer said, “Sure, it was the most watched inaugural.”

That’s a difficult number to quantify, let alone to compare with the 2009 inauguration. But some data suggest Spicer is correct that online viewership was up dramatically from 2009.

Spicer argued that if you add the 31 million TV viewers estimated by Nielsen with 16.9 million who viewed it on CNN, that alone would be the highest in history. The 16.9 million, though, isn’t a hard number. If someone were to log on and off repeatedly, that person would be counted multiple times.

A CNN spokeswoman told us that at the peak of CNN’s livestream coverage at 12:15 p.m., there were 2.3 million devices (desktop, mobile, connected TVs) streaming CNN’s feed of the inauguration.

Digital comparisons to the inauguration in 2009 are difficult, though. Online viewing of Obama’s first inauguration set records at the time. A CNN spokeswoman told us that in 2009, there were 23.7 million live video starts on CNN during the inauguration. But she cautioned that that figure is likely inflated because technology at the time often required users to restart frequently. “There were roughly 8.3 million visits to the streaming content in 2009, which was much lower than 2017,” the spokeswoman said.

And that’s just CNN. As Spicer noted, there were lots of other platforms from which viewers watched the inauguration via livestream, such as on YouTube and Facebook.

Akamai Technologies, a content delivery network for most of the major networks and newspapers, reported that the Trump inauguration was the largest single live news event that the company has delivered. At its peak, the company recorded 4.6 million people watching the inauguration simultaneously (up from a peak of 3.8 million back in 2009), said Chris Nicholson, a spokesman for Akamai. Many more people may have tuned in to the inauguration online at other times, he said. And, he noted that Akamai’s customers represent “a big chunk but not all” of the online providers of the live video of the inauguration.

Does all that add up to the most viewed inauguration in history? Maybe, but data necessary to make apples-to-apples comparisons are limited. The firm comScore, for example, says it is developing technology to measure audience across platforms — including TV and online — but that has not yet been rolled out.

Regardless, Trump and Spicer argued that the media misrepresented the size of the crowd at the inauguration, and we don’t find any evidence of that. To the contrary, it was Trump and Spicer who provided false information to feed a false narrative about crowd size.