大象传媒

Not only an underestimate, but if their estimate was 12,000 why did they say 10,000 to the press at the time?

Also, licencing capacity is surely based on other factors, such as how many people could be evacuated from a space safely in the event of fire, etc. not just on the number who will physically fit into the space? Therefore the Met's assumption that "This, they estimate, is Trafalgar Squares capacity." doesn't hold up. It's not an estimate of the capacity of the square, but of it's safe capacity where "safe" is defined according to a host of H&S regulations. While it could be helpful, it is not, therefore, necessarily a good starting point by itself.

I would have expected there to be statistical methods based on video or photos to estimate crowd sizes these days.

Licensed capacity is indeed a far cry from actually phyisical capacity. For example, I've personal experience of a venue which would have a licensed capacity of 80 people while it regularly has 200-300 and on occasion upto 500 people.

I thought Trafalgar Square was known to hold 60,000 people and that was before the pedestrianization of the 'North Terrace'. That would explain the organisers estimate and has certainly been how I guessed crowd numbers there in the past.

This is a great relief to actually read this in the press. And from the 大象传媒 no less! The police estimates of the numbers at demonstrations are often ludicrous. Though no statistician and having never won at a guess the number of beans in a jar competition, it is often glaringly obvious to me (as someone who often attends such marches and rally's) that the police brazenly downplay the actual numbers of people involved in such protests. The police of course have a poltical motive in doing so. Then again the organisers too, of course, have their own motives and will be sometimes be guilty of overplaying the numbers. Can anyone tell me how the organisers (in this case StWC) actually make their own estimates?

The police have a vested interest in underestimating the number of demonstrators there are for most demonstrations - they are the protectors of the state after all.

This isn't about sloppy methodology or a simple mistake, it is the police upholding the role that they have been given/crafted out for themselves. If it was about sloppy methodology then that would beg huge questions about the intelligence of the police force itself: not just about being a bit rubbish at something, more about having a certifiable learning difficulty.

The disparity was biggest during the Feb 15th 2003 protests: police estimated 40,000 to start with, and even conservative estimates now state that there were about 1 million.

PS. liked the way the cops break the law when it comes to freedom of information requests. "One rule for the....", etc.

It is even more complicated than you think, and they have said, for that matter.

It suffers from the same problem as X-Factor voting. "6 million people voted!"

Okay, let's be honest, most people voted several times. If they only voted 6 times on average that brings it down to a Million.

When the police estimate numbers, they also have to judge that people are often getting counted in several times. Have you ever tried to get protesters to stand still to be counted?

Many years ago I was standing next to a picket line, minding my tape recorder. There were around 20 official pickets and another 60 odd miners who had come for the fight. (They freely admitted it, and most didn't know who the original strikers were.)

There were 30 police men and two vans of SPG parked quietly round the corner. The expected fight broke out eventually. It wasn't very bad. But according to the NUM the next day, Riot police had batton charged the "peaceful" demonstration and caused injury to over 200 people.

The police like to get their numbers reasonably accurate so they can pull in reserves if needed. They don't benefit from getting the count very wide of the mark - but the protesters do!

I have never understood why the police have this incentive to underestimate numbers. To suggest that they are part of the state and the state cannot admit how many people protest is na茂ve. It is surely the organisers of events who have a greater incentive to exaggerate numbers.

About 20 years, I spoke to a man who was counting the number of people going into Hyde Park, for a CND demonstration. I asked him why he was counting tourists as well as protesters and why he was not counting those who were leaving. He explained that he was a civilian employed by the Metropolitan Police鈥檚 public order branch and it was necessary to compile statistics for the Home Office to show how many protesters and how many police were present. It was therefore in the interests of the police to exaggerate numbers because the ratios of police to protesters looked better.

It was only after the demonstration that I read that Dr Julian Lewis, now an MP, had commissioned an independent aerial survey of the demonstration. As I recall, this showed that just less than 100,000 protesters were present 鈥� about half the number claimed by the police and a quarter that claimed by the CND.

The capacity of Trafalgar Square is about 20,000. So when the 大象传媒 and GLA organised the VE Day 60th concert in May 2005, for which there was no incentive to under or overestimate numbers, it was stated that the 鈥渃apacity audience鈥� was just 15,000.

David Tait states that the capacity of Trafalgar Square at the VE Day 60th concert (2005) was just 15,000. As someone involved in organising this event I think it worth mentioning that this was an all seater occasion with disability accessibility, very robust evacuation, fire, and health & safety arrangements in place. This was not a densely packed, all-standing protest event, with people continually entering and leaving the square.

I assume you posed the same question to the Stop The War Campaign organisers? How did they arrive at their figure?

David -
A reasonable question, and I did ask the Stop the War Coalition how they reached their figure. They told me that their stewards do a five-minute count of numbers passing a particular point and they then scale it up in line with the time taken by the whole demonstration. They acknowledged that this is an 'inexact science'.
Where does this leave us? The methodology described by the Police is clearly flawed and leads to an underestimate. The Stop the War Coalition is not covered by FOI and thus is not subject to the same level of scrutiny as the Police. Whether their estimate is soundly based is impossible for me to say.

This is a late response, but the parts only came together yesterday, as I was driving through the roadworks on the southern part of the M1.

There were three components:

1.We are regulary using image recognition and data processing technology that capture vehicle registration numbers in huge quantities and process that data to calculate average speeds of identified vehicles, all in real time.

2. Many (most? all?) Shopping malls have been using - for at least 10 years - image recognition to produce their 'footfall' figures, which they use to justify their rents.
The cameras are quite low-res - the standard is to be able to separate two people walking close together and on-axis to the camer's line-of-sight.

3.These marches take place in city centres - within the purview of large numbers of CCTV cameras.

So "we have the technology". Why is it not being used? Or is it being used, and we are not being told?