How exaggerating protest numbers could backfire on students


That’s how many people, according to a firm hired by Radio-Canada, were at Place du Canada at 2:35pm on Wednesday during the monthly protest against tuition hikes and Bill 78.

As I predicted, the number prompted outrage among protesters and their supporters. Reactions from “no way” to “fraude“. Some presented evidence to back up their cases, but in all cases they involved subjective comparisons. Many judged based on a single aerial photo (whose source I couldn’t find easily). Others based their numbers on known capacities of large stadiums. But many just pulled numbers out of nowhere. CLASSE’s official estimate was 100,000. Numbers as high as 250,000 were being thrown about.

My estimate was 24,000, and I provided my methodology. I stood at the corner of Berri St. and René-Lévesque Blvd., estimated that people were crossing a line on the pavement at about 10 per second on average, and multiplied that by how long it took the whole protest to pass by – about 40 minutes. (I had to leave early to go to work, so I estimated that based on the tail end being at St. Laurent and de Maisonneuve 25 minutes after the first people crossed at Berri and René-Lévesque. I figured it would take the tail end about 15 minutes to reach where I was originally counting from.)

Comparing apples and pineapples

Measured in time, the protest was about half as large as the one I saw on May 22, whose size I also tried to estimate. That estimate was 50,000, but with the understanding that I was only counting the people who passed my location.

The closest thing to an official estimate we’ve had of one of these protests before now is QMI Agency’s call of about 150,000 for that May 22 protest. They did not reveal their methodology, but it was the only time a news agency offered its own number.

So the media have been relying on protesters’ own estimates of the protest size. And obviously, there’s clear motivation for them to inflate those numbers.

It’s not just the student protests. Remember that big march against the Iraq war in 2003, that drew 200,000 or 300,000 people? Radio-Canada did an official crowd estimate there too, and the actual number was much lower (see the video at the bottom).

In first-day stories, the media have been careful about their estimates. They refer vaguely to “thousands” or “tens of thousands”. When they use the protsters’ numbers, they’re clearly attributed. But as time passes, the care starts to slip, and without any competing numbers to refute them, those estimates of 100,000 or 200,000 people slowly become fact.

So now, as media outlets start to realize they can’t abdicate their responsibilities and really need to do their own crowd estimation, more accurate numbers show a dramatic drop. People start to make comparisons in their heads: If the May 22 protest drew 200,000 people and the Aug. 22 one drew only 12,000, the student movement is clearly dying out.

Let’s compare

The biggest problem with amateur estimates of crowd sizes is that people don’t know what a crowd of 10,000 people or 100,000 people looks like. So they try to compare with the Bell Centre (21,000), the Olympic Stadium (60,000), or the Place des festivals (which they say is 100,000 based on estimates of Jazz fest mega show sizes, but those estimates include crowds watching secondary screens on de Maisonneuve Blvd., at Place des Arts, on Ste. Catherine St. or even on Clark St.)

But these can be deceiving. The Bell Centre and Olympic Stadium look a lot smaller than they are, because of all the empty space in the middle. And crowds there are packed in very tightly, unlike a moving protest march.

Rather than subjective comparisons of size, let’s compare the crowds using another method: transit.

  • 20,000: A packed show or hockey game at the Bell Centre causes a strain on the metro system as it clears out. The STM routinely adds extra trains at the Lucien L’Allier station to handle the thousands of people who hop on at the same time. There’s a noticeably large crowd at transfer stations like Berri-UQAM, consistent with what you’d find during a busy rush hour.
  • 50,000: The biggest events at the Olympic Stadium have been in the 50,000 range. When they end, the metro system is severely strained. Extra trains are parked near the Pie-IX station, and for a good hour they fill up and depart westbound toward Berri-UQAM. Even then, extra trains are added to the other lines as well, and security officers herd crowds towards the ends of the platforms to get as many people as possible onto the trains.
  • 80,000: Remember that U2 show at the Hippodrome? Officials pleaded with people to take public transit because parking would have been a nightmare. Though the shows let out before midnight, the metro was kept running past 2am because that’s how long it took to get everyone on the trains. It was so bad that not only were people directed to walk to two metro stations, but a fleet of dozens of buses was brought into service to shuttle people to the Jean-Talon station via a special reserved lane marked with pylons for the whole route.

Which of these do you think is the best comparison to Wednesday’s protest?

When 10,000 isn’t enough

What’s most disconcerting about all this is that the bar has been set unreasonably high for large protests in this city. Tens of thousands of people taking to the streets isn’t enough anymore. It has to be hundreds of thousands before anyone starts noticing.

That’s unfortunate. Whether you agree with the student movement or not, they amassed enough people that it took them more than half an hour to walk by in a march as wide as five traffic lanes. No matter their actual number, a descriptive word that’s synonymous with “enormous” is called for here.

But so long as we continue to measure protests like we do penises, this obsessive war over numbers will only distract from any real issues we might be trying to debate.

UPDATE: OpenFile, in a story about how difficult crowd estimates are, comes up with 80,000 based on a march 3 km long and 18.5 metres wide with 0.7 square meters per person. That seems a bit too dense to me. But at least it’s a scientific effort.

12 thoughts on “How exaggerating protest numbers could backfire on students

  1. Vaudreuil Bob

    24,000 more or less, sounds about right to me. They passed in front of our offices on Rene Levesque. We were watching them from above.

  2. Walter Zuker

    Excellent discussion on quantitative information being “fed” to the media.

    Sadly this is the rule and not the exception.

  3. Kevin

    I have one question for people who think that these figures for protests are too low: have you ever been to a sporting event?
    Go to an Impact game beforehand and afterward. Watch the crowd snake along (and that is what is does, crawl to the parking lots an metro station). Get a sense of density, speed, flow.

    Then find out how many people bought tickets and went.

    I am of the opinion that anyone saying there was more than 20, 000 people at this week’s protest thinks that Saputo stadium has a capacity of 300,000.

  4. Alex H

    “The closest thing to an official estimate we’ve had of one of these protests before now is QMI Agency’s call of about 150,000 for that May 22 protest. They did not reveal their methodology, but it was the only time a news agency offered its own number.”

    QMI / TVA Nouvelles has been adamant about counting every “nightly” protest and reporting it like it was a big thing, even on the nights when it seemed to be 5 homeless guys banging their shopping carts together. They appear to be veyr big supporters of the students, or more likely big supporters of anything that makes the Charest government look bad. So their estimates of 150,000 is likely very, very, very, very exaggerated.

    It’s one of those times where you can see the political bias of the media coming out.

    1. Fagstein Post author

      They appear to be veyr big supporters of the students

      You’re accusing Quebecor of showing a systemic bias in FAVOUR of the student movement?

      1. Alex H

        I am saying that, while other media generally let the story go (or buried it) Quebecor (especially TVANouvelles site) was more than willing to count the days. Even on the days where there were very few protesters, they counted it as the “XXXieme” nightly protest. I don’t think it is actually a support of the students, as much as a way to put pressure on the Charest government.

        Now, what would be interesting is to find out how many students were protesting, and how many others were protesting. I have a feeling that many of these so called student marches were really just the usual collection of professional protesters, union members paid to attend, and others who just felt it fun to bang pots and be noisy.

        1. Fagstein Post author

          I have a feeling that many of these so called student marches were really just the usual collection of professional protesters, union members paid to attend, and others who just felt it fun to bang pots and be noisy.

          I really doubt there are 10,000 of these people in the Montreal area. I’ve seen no evidence that anyone (except the cops and journalists) were paid to attend these protests, and union activity has been quite low. Unions do show up to protests, but their presence is obvious because they are well-organized and carry union banners and flags. None of this has been present at the vast majority of these protests.

          1. Alex H

            Actually, I can confirm 100% that my ex-neighbor, a younger dock worker at the port of Montreal, was encouraged by his union to appear at these nightly protests whenever possible. The particularly large May event was packed full of union members and other non-student types looking to rally against the government and hopefully get a more union friendly government in, like the PQ, who would literally bend over backwards to make the unions happy.

            I am not clear if the unions were paying “strike pay” type benefits for people appearing, but it was clearly organized. Oh, and they were apparently told NOT to appear as union members with their local or union badges showing, but to appear in student garb, possibly with a back pack or other similar student look items. It’s amusing, thankfully I no longer have to deal with that crap.

          2. Alex H

            Let me point to this:


            Why would they bother to give coverage to 40 people, if it wasn’t against Charest? This story was TOP of the tvanouvelles site for a while, and then in the second row “larger” image feature story position. It’s 40 people. Not even enough to fill a single city bus. You get more people out to watch a little league baseball game. Why is this a major story, unless you have an agenda to push?

  5. wkh

    I don’t think transit is a good way to compare. The people attending don’t all get on or off a metro or bus at the same place like the do with those other examples, or at the same time. A lot of people walk to the start point, and most go to a bar rather than a bus or metro when done walking. Or back to their residence as the manif passes nearby.

    I think 12K sounds insanely low. I don’t think the numbers honestly matter beyond “enough to block traffic and cause drama.” I also think 80K sounds insanely low for the May demo. I don’t believe 400K but 80? Naw.

    1. Fagstein Post author

      I also think 80K sounds insanely low for the May demo. I don’t believe 400K but 80? Naw.

      I’ve laid out my methodology for my estimates for both protests. For your math to work some part of my math must be wrong. Which part?


Leave a Reply

Your email address will not be published. Required fields are marked *