More 9/12 Crowd Data: Yeah, It Was Big
On Monday, PJM published my piece titled "March on Washington: How Big Was the Crowd." It contained several estimates of the actual number of people who attended the mach on the Capitol on September 12.
The legacy media seems to have converged on an estimate of 60,000 to 70,000 people. This appears to be sourced to the D.C. Fire Department, although as several people have pointed out, no one seems to recall any other estimates coming from the D.C. Fire Department. On investigation, it turns out to be Pete Piringer, public affairs officer for the D.C. Fire and Emergency Department, as quoted at Politifact:
[Piringer] said the local government no longer provides official crowd estimates because they can become politicized. But the day of the rally, Piringer unofficially told one reporter that he thought between 60,000 and 75,000 people had shown up.
“It was in no way an official estimate,” he said.
We asked Piringer whether there were enough protesters to fill the National Mall, as depicted in the photograph.
“It was an impressive crowd,” he said. But after marching down Pennsylvania Avenue to the Capitol, the crowd “only filled the Capitol grounds, maybe up to Third Street,” he said.
Emphasis mine. Remember those points: "in no way an official estimate" and "filled the Capitol grounds, maybe up to Third Street."
- The Daily Mail first reported 2 million, then scaled it back to 1.2 million.
- Barbara Espinosa reported that actual traffic counters from the march organizers reported about 450,000 crossing Pennsylvania at 11th after about an hour of the march, and nearly 1.5 million total.
- There was also the time-lapse photoset from the traffic camera over the Freedom Plaza, which is where 14th St NW crosses Pennsylvania Avenue at E street.
- Our own estimate of more than 850,000.
- The Noble and Ancient Order of the Gormogons, with an estimate of about 1 million.
- Another estimate by Henry Vanderbilt, reported at Transterrestrial Musings. Using two different methods, Vanderbilt arrived at one estimate of up to 320,000 in the march down Pennsylvania alone, and 350,000 to 500,000 on the Mall and Capitol grounds.
- The estimate of 350,000 from Political Gumbo -- and let me just mention: this is the way it's done, kiddies. Ken Vaughn looked at my computation, made his own, documented his assumptions, and came out with a different number for the number marching on Pennsylvania Ave. Frankly, from his argument, I probably trust this estimate over my original one; remember that it's only the marchers, though, so it's certainly a lower bound number as well.
There were several sources that, over time, were clearly spurious, such as the famous "aerial photograph" that turns out to be from a Promise Keepers rally months ago. There were also some spurious debunkings which depended on a claim that the traffic cam photos were from another event, based on the flag at half-mast. I found that a little questionable, because I went to the webcam at Freedom Plaza myself during the march and saw images comparable. If these were being faked, the fakes were being fed into the webcam stream upstream of me. It turns out, in any case, that flags were at half-staff that weekend because of the Patriot Day holiday, and in any case, since then there have been many well-sourced photographs of the crowd from different vantage points; it appears that we should continue to take the traffic cam videos as reliable.
Since I wrote that piece, though, we have two new sources of information. First, the ridership statistics from D.C. Metro became available after being delayed, apparently because of a fatal accident on the Metro tracks. The Heritage Foundation, using these figures, computed that Metrorail ridership was about 235,000 greater than the previous weekend. As they say, that in itself is more than three times the (unreliable and badly sourced) number reported in the legacy media.
Second, there is now a high-resolution photo from FreedomWorks, which you can see in the poster here.
After the issues with the Promise Keepers rally photo, I called FreedomWorks to verify this one. I spoke to Joseph Onorati, who confirmed to me that the photo is from the actual 9/12 rally. It was taken by Matthew A. Beck, a volunteer photographer, from the Capitol itself; he was taken to a high point in the Capitol by Rep. Tom Price (R-GA), one of the speakers at the rally.
Let's look at the photograph (and it might be worthwhile to look at this Google Map view of the area). It's a panoramic photo, from about 1st Street on the south side at the left end of the photo, past Maryland Ave, the Mall, and Pennsylvania Ave around to 1st Street again to the north at the right hand end of the photo. In it, we can see that the Mall is indeed full back to about 3rd Street, but that at the time of the photo, Pennsylvania Avenue was also full back to at least 3rd Street, and there are significant numbers of people on 1st and on Maryland. (Remember that the area west of 3rd Street was reserved for another event, but notice that the side paths are very full for a good ways back from there.)
I spoke to Clarice Feldman, a Washington attorney and PJM contributor who was there; she had arrived by another route and ended up in a very crowded area at the base of the Capitol, out of view in this photo. She confirmed that the crowd was very dense; examining this photo suggests the same thing. (Look at the people in the photo. In a dense crowd from a photo from above like this you see only the very upper part of the body; in a less dense crowd you see a larger part of the body.)
Now, let's look back at the National Park Service method as published in USA Today. If, in fact, the Mall were only as full as a seated crowd of invited guests, as shown in the article, and if the crowd only went to the diagonal streets Pennsylvania and Maryland, then the Park Service method would give an estimate of about 250,000. But a crowd of this sort is rather more dense than a crowd of seated invited guests because of the need for aisles and access in a seating plan; you can see that in the overhead photos of the inauguration.
At this point, with this photograph and the Metrorail ridership, it's clear that this must be a new lower bound: the "60,000" number turns out not to be any sort of official estimate. Also, when quoted fully, it turns out that the D.C. Fire Department source agrees with the photo (remember the line I emphasized above?).
One more factor that we haven't used in any estimate yet: charter buses. There has been variously reported a figure of 4,500 bus parking permits. Now, there were other events on, and I can't confirm that number reliably. But a busload of people is normally about 50. If only half as many buses as reported were actually there with people who attended the rally, that would account for another 112,000 people.
What can we take away from this exercise? Here are the main points:
- The estimate widely used in the legacy media is not from an authoritative source, and it isn't even consistent with itself: "full back to 3rd Street" is around 250,000 by Park Sevice methods, not a quarter of that.
- Many estimates, using different assumptions and different methods, arrived at numbers well into the hundreds of thousands.
- This is clearly consistent with the panoramic photo that we can source reliably.
- With everything above, and with several more estimates, I don't think there is a plausible argument for any total attendance figure much less than 500,000 to 600,000. That is, nearly ten times the reported attendance.
How much does this matter? It's hard to say. It might be that a precise number isn't needed, and at any rate it may not be possible to get a good one. What we can say is that the number being reported is wildly wrong. Wildly too small.In any case, the number that will really matter is the count at the polls in 2010 and 2012. Given this, I'd say that any politicians who hope to retain their jobs better look very carefully at that panoramic photo -- and remember.