Slightly OT: Crowd Guesstimation

The events unfolding Egypt are a long way from southern gospel music, obviously, but watching the massive crowds assemble in Cairo and then reading some of the attempts online to guesstimate the size of these gatherings brought two things to mind: 1)I am spectacularly crappy at crowd guesstimation whenever I have to give it a go at concerts, and 2)I wonder if there is some kind of reporter’s lore or passed-down tips among journalists or other observers of political gatherings and demonstrations to figure out how to guesstimate the size of these crowds when no easy counting method presents itself.

Obviously when we’re talking about an event in an auditorium, one can (theoretically) rough out a grid in one’s mind based on seats and rows and make a back-of-the-ticket-stub calculation about crowd size by comparing an approximation of capacity against how full the room seems (though I’m blogging proof of the limitations of even this method).

But the Washington Mall, or Times Square, or Tahrir Square? Based on a conversation with a National Parks Service official, USA Today came up with this formula: 

Estimating the size of a crowd requires aerial photographs, dimensions of the gathering space, and an estimate of the crowd’s density.

Uhm, ok. But … Clearly most reporters (to say nothing of a blogger or freelance photographer in a far flung land) can lay hands on only some of this data in most cases, unless they are much better resourced than most media outlets these days. So what says the wisdom of the crowd around here? Any promoters or performers or die-hard fans have any reliable ways of guesstimating crowd size in open-air or ticketless or other unstructured settings? Let the crowdsourcing begin …. Lords knows I could use the help.

Update: In addition to the impressively thoughtful comments on this question, fellow blogger Daniel Mount chimed in via email:

I was involved in politics in my teen years (and through early twenties; I’ve been out for a few years). During the time I was a congressional campaign and later staff aide, my job sometimes included a physical head count, even if that meant standing in the back of the room for half of the event. I can tell you firsthand: Local newspapers, at least, average no better than you do. I have seen thirty-two vigorous protesters recounted the next day as “hundreds,” and 200+ shrink into a few dozen.

(Photo: Southern gospel nights at Silver Dollar City, via

Email this Post


  1. Larry Ward wrote:

    30. Estimating Crowd Attendance
    Many festivals have it easy if they are paid or ticketed events. They simply
    count the tickets sold or add up the turnstile counter total. For the majority
    of the free events the task is much tougher. You can guess, as many do, but
    your sponsors require and the media and vendors will want to know. The
    more folks you have through the event, the easier it is to attract sponsors
    and other vendors.
    1. The grid method is a tried and proven way to count the crowd, but
    works best for events where the crowd is sitting and watching a
    concert and not moving around. You take an aerial photograph and
    then mark off the picture in 100×100 foot squares. If sitting on the
    ground or in lawn chairs, there are usually a fi xed number of people
    in a 100×100 area. Simply count the number of folks at your event
    in one square area, add up your squares, and multiply to estimate
    your total. This method requires an airplane or helicopter, but that is
    a different problem, isn’t it?
    2. The Wal-Mart method is more accurate as it is a well-known fact
    that everyone that visits your town will go to Wal-Mart to buy sunscreen,
    chairs, coolers, etc. Enlist the help of someone to count the
    number of people going through the front door of the local Wal-Mart
    during your event. Then at some other time count the number of
    folks not during the festival. The larger number during the festival
    less the number on a regular day is exactly the number at your event.
    Now you may point out that 4% will go to K-mart instead, that is
    true, however exhaustive studies show that 4% is offset by the folks
    that visit Wal-Mart more than once during the event.
    3. By far the most accurate way is to weigh the contents of the porta
    potties. This gets a little complicated due to its scientifi c nature but
    try to follow along. The average festival attendee, if healthy, will
    urinate ½ gallon of liquid per day. The weight of one gallon of water
    is seven pounds per gallon. Have your toilet clean out crew weigh
    the empty weight of their truck. Once they clean out the toilets each
    night, weigh the trucks full. The average pump truck holds 600 gallons.
    If the truck is full and weighs, for example, 4200 pounds full
    150 | Fair, Festival & Event Promotion Handbook
    of, well, you know, then each truckload full represents 1200 people.
    Now it gets a little trickier. You have to compensate if you sell lots
    of beer. Of course folks will pee more if they are drinkers, so you
    have to about double the amount of volume pumped or half the attendance
    results. Also you need to have a temperature allowance.
    If it is a really hot day, the normal person will sweat out a lot of
    moisture through their skin and not go to the toilets near as much.
    You should add about 30% to your attendance fi gures due to this
    variable. After applying the variables of beer verses no beer and the
    temperature compensation, there is the well known style of music
    factor. We have yet to fi gure out the reason, but it appears that Blues
    festivals in particular attract attendees that have trouble fi nding the
    porta potties and simply go behind a building or fence. If you have
    trouble following the extensive math, we will loan you our patented
    SEWER SUCKER CALCULATOR as long as NASA doesn’t buy
    it from us first.
    I trust you have by now figured out this chapter is written in jest. We thought
    that much of this book is too serious and you needed to grin somewhere
    along the way. It is also a way to test you to see if you are really reading it,
    if you did read this, email us at and let me know.

  2. Irishlad wrote:

    f(x)=ax3+bx+c+cx +g.
    c= crowd

    Quite simple really.

  3. judi wrote:

    When I used to report on gatherings in spaces where I didn’t know the capacity, I would count the number of seats in a row or section and multiply that by the number of rows/sections to get an idea of the capacity. Then I’d estimate based on 1/2, 1/4 space filled, etc. I’ve also read about the National Park service grid system and it is very complex….thus few organizers trust it, especially if the estimate doesn’t jibe with their publicity goals. If it were some outdoor gathering where people largely stayed put, like a rock concert, you could mentally measure a square foot-sized space, and estimate from there. Example: a square yard (3 feet by 3 feet) might hold 9 people if they were very skinny (or children) and took up only one square foot of space each. But might be easier to work from a known quantity…like an elevator. Measure one and see what the square foot/yard size is in relation to the maximum capacity the little sign says it holds. Then assume that at most gatherings people aren’t packed that tightly, but in the Egyptian demonstrations they certainly appeared to be. So how many 5×6 (25 people/sardines max capacity) elevators are in that riot-filled square? Easier still might be to mentally image a larger room. A bedroom, or a deck, or something of that ilk, whose measurements you know, or take. How many people would it hold? (Think about your last holiday party.) Now how many of those rooms in the outdoor space? I have to admit I’ve tried this only on a small scale. For a garden variety outdoor alumni picnic I use the clump system. I will count off a group of 10 or 25 and then try to estimate how many of those clumps are present. If there are about 10 clumps of 25, then there are about 250 attending. Try practicing the next time you are at the beach and a whole flock of gulls or sandpipers descends, or you are on a boardwalk looking down at a school of fish. With practice, you can apply a system you work out to crowds of people. But when it comes to riots, or claims of a million marchers, I am as clueless as you.I’m not sure an accurate crowd estimate is possible under those circumstances, even with aerial photography and a grid system, because you are trying to pin down a moving target. (There I go with the gun imagery again.)

  4. Nate Stainbrook wrote:

    I have never been good at guesstimating SG crowd sizes; so thanks for the tips 1 & 3…

  5. Jake wrote:

    Re. the picture at the beginning of this article, I would say that if they are Democrats, there are about a million people there. If they are Republicans, probably a couple hundred. LOL

  6. CVH wrote:

    These are all tried and true methods of crowd calculation but there’s one factor that’s been overlooked: the SGFAF, or Southern Gospel Fan Ass Factor. Given that the size of many of the fans rivals that of the Kingsmen back in their “ton of fun” days, you have to divide whatever total head count you come up with by 2.132 to allow for the increased space used by the SGF’s As.

  7. Rick in southGeorgia wrote:

    Re:#5- Very funny, except I think you have that backwards. :)

  8. irishlad wrote:

    CVH, I saw a couple of guys in smocks once at the NQC who could have quite easily turned that estimation into 3.123.

  9. quartet-man wrote:

    Now, I am one of the last people to talk about size, but this reminds me of something that happened in Government class in High School. There was a smart-mouthed kid in class who would often make comments to the teacher. She often made them too and most of the time they both took things fine.

    Anyhow, the teacher made some sort of comment about the average US household having 3.2 members (or something like that). Anyhow the kid in class asked “.2, what’s that?” The teacher answered “that is you.” The kid replied “well, you are the other three.” :D

  10. KermitTheHermit wrote:

    I have always hated the word “guesstimate.”

  11. DRL wrote:

    I hate the phrase “wisdom of the crowd.” When did the crowd suddenly become wise? The mob yells, “crucify Him!”

    When I was a kid, the phrase was “the masses are asses.” If the crowd is doing something, you’ll usually find wise people doing exactly the opposite.

  12. Irishlad wrote:

    #11 Tell that to the crowd in Egypt.

Post a Comment

Your email is never published nor shared. Required fields are marked * Please note: Comment moderation is enabled and may delay your comment. There is no need to resubmit your comment.