It was predicted that around 2 million people would come out to watch the Raptors parade in Toronto on Monday, but city officials say the number is likely higher.
Toronto’s chief communications officer Brad Ross told Global News that the earlier estimate of a couple million now represents the “lower end” of overall crowd estimates.
The Greater Toronto Area has a population of just over 5.9 million, meaning if 2 million people did attend the parade, that would be nearly 34 per cent of GTA residents.
READ MORE: Millions of fans flood downtown Toronto for Raptors victory parade
Parade estimates don’t include the head count of people watching from office buildings.
The parade started at 10 a.m. at the Princes’ Gates, located on the Exhibition Place grounds. It made its way slowly through the steady stream of hundreds of thousands of fans lining the streets, ending at Nathan Phillips Square for a celebration rally.
Nathan Phillips Square reached its capacity before the Raptors arrived, and city officials would not let any people onto the grounds after 12:30 p.m. A city official confirmed to Global News that the square can accommodate 65,000 people.
The official said that figure included people in the square itself and the immediate surrounding area.
While photos from the parade route showed massive crowds of people, is it possible to know how many people actually turned out?
How to estimate crowd size
Steve Doig, a professor at Arizona State University and an expert on crowd estimations, told Global News that 2 million people is likely an overestimation.
Doig said that when organizers give out crowd estimates immediately after an event, they are often inaccurate.
To get a “reality-based estimate” of a crowd, Doig says you need two things: the size of the area the crowd occupies, and the estimated density of the crowd.
READ MORE: Overcrowding becomes issue as millions of fans turn up for Raptors championship parade
“If the crowd is gathering in a particular location, you can use Google Earth to estimate the area of the crowd in square meters,” he said.
“Then you need an estimate of the density of the crowd, and there’s a couple of ways of doing that.”
Doig said for a “loose crowd,” where everybody is about an arm’s distance from each other, the density is about one person per square meter.
“If a crowd is gathering to hear people speak… they may be getting in tighter, in which case you could go down to perhaps half a meter per person,” he said.
WATCH BELOW: Raptors victory parade: Snowbirds fly over Nathan Phillips Square
Based on those figures — the square metres of space and the amount of people possibly in that space — you can calculate approximately how many people are in the crowd.
For a parade, estimating crowds is a bit trickier, but not impossible.
Doig says you take the width of the street that the parade is on, from storefront to storefront, and subtract the path the float travels through. That gives you an idea of the area the crowd is possibly standing in, he said.
“Then, you need to make some kind of a reasonable estimate. Are the people really packed in? Are they deep all the way from the storefronts to the police corridors?”
After you estimate the parade route turnout, you also need to estimate how many people are at the parade’s end stop. With the Raptors parade, this was Nathan Phillips Square.
“You estimate the crowd there, too,” he said, and add that amount to the parade route numbers.
According to Charles Seife, a professor at New York University who has written about crowd counting, aerial photographs are one of the best ways to estimate turnout in a public or open space.
WATCH BELOW: Raptors victory parade: Fans crash through barriers at media area at Nathan Phillips Square
Seife told Global News that experts break up aerial images into squares, and do their best to estimate crowd density “on a fairly granular level.”
Even still, crowds are “dynamic,” meaning people move around, leave and come in. This means accuracy can be hard to determine.
There are also tools and websites used to estimate crowd turnouts, like MapChecking.com. This site takes the surface area of a space, the estimated number of people per square metre, and calculates crowd size based on that.
WATCH BELOW: Raptors victory parade: Jeremy Lin talks about bringing a title to T.O.
Lastly, before an event, Doig it’s a good idea to know the maximum number of people that can fit into a space.
“We wind up quoting these wild estimates of crowd size, and really, it would be impossible to gather that many people in that kind of space,” he said.
While estimates of 2 million could be right, it could also be inflated because people can’t really “eyeball estimate” crowds very well, Doig said.
“A lot of times these estimates are made because a crowd size is kind of a metric for enthusiasm,” he added.
Seife echoes this, and says that event organizers have a tendency to overestimate.
“Often there’s kind of a prestige thing… There’s a vested interest in making things seem large rather than small,” he said.
“The official estimates are almost certainly going to be overdone. If you get an objective crowd scientist doing it well rather than what the city says, I suspect you’ll find a reasonable difference.”
WATCH BELOW: Raptors victory parade: Kawhi Leonard accepts plant from the crowd
Doig says eyeball estimations often lead to “crowd dueling,” which was seen when U.S. President Donald Trump accused the media of lying about his inauguration crowd size.
“People who are in favour of the crowd want to make the number as big as possible, and those who were opposing it want to deflate the estimate as much as possible,” he said.