So how many marched on saturday?

As usual there is a huge difference between the estimates of the police and the organisers. According to the BBC:

One of the biggest marches of the day was in London.
Organisers say 100,000 people took part, while police put the figure at 45,000.

I don’t know the rationale for either figure but I am naturally suspicious of the police figure. They have historically been on the wrong side of every major social conflict since the first bow street runner er… hit a poor beggar on the head. As far as I can remember the one occasion the police estimate matched the organiser estimate was the Countryside Alliance march.

However can we really trust the organiser estimate? They have an obvious interest in over estimating just like the police have an interest in under-estimating. Now I don’t doubt their sincerity and veracity – but many, many others do. My friends always claim it somewhere inbetween the two figures (which is silly when you think about it). Why don’t Stop the War employ a few crowd specialists to devise a method (or just bloody count people as they leave Hyde Park)? Then at least we could challenge the police figure with more credibility.

This issue has been raised before most notably by John Vidal in the Guardian. He interviews a company who provides a crowd counting service:

So is there an accurate way of counting the numbers at demonstrations? Yes, says Dr Jake de Salis, of the planning consultancy Intelligent Space which specialises in crowd numbers. He was employed by the GLA to put a real figure on how many people went to the Notting Hill carnival where the figure of 2m people was always quoted by both police and organisers.

Using hundreds of people with tally counters and stopwatches, analysing CCTV cameras, sampling head counts, collecting bus and tube figures and evidence from police aerial pictures as well as a tested formula based on how many accidents there are in mass crowds, he found that the figure was more like 1.2m.

What are we waiting for?