Well it’s been some time now since the smoking ban kicked in and London is coping with the withdrawal symptoms. Tobacco sales seem to be going down but statistics as a science seem to be going the way of the dodo. Misrepresented, misremembered, about to be extinct as a science and only used in an imaginative way.

I am sensitive about statistics. This is not to say that I like statistics, quite the contrary. I find mathematics unbearable and this combined with the cold ‘logic’ of statistics leave me shuddering every time someone mentioned the word ‘regression’ which I can only explain – almost by reflex – exclusively under exam conditions.

But because I dislike statistics – and they are pretty necessary in the real world – I made sure that I know all about their limitations. Mark my words: Statistics can never prove anything. They can only suggest a connection. What is that connection is up to the person that is reading the statistics. As it happens we are hasty creatures, hence the conclusions we draw from numbers are hasty too – more often than not.

Back to the smoking ban in public places it seems nowadays that everywhere I look in London there are more people outside the pub than inside it. Being a smoker I understand those that need to nip in and out. Being an ex-resident of a Mediterranean country I have to admire their courage in dealing with the horrible weather we have been having all through the summer and early autumn.

Yet some reports are suggesting lately that tobacco sales had a spectacular drop in England and Wales (Scotland started the ban earlier) during July and August. Why? Ah, of course – say the all-knowing journalists – it is due to the ban. Some newspapers published a 7% drop on average those two months year on year while others went with the most impressive figure of 11% drop during July year on year. You would have thought that the whole tobacco market is crumbling and companies related to the trade are running amok. Right?

The truth is that nobody is panicking (yet) and there is a very good reason for it. I know because I received the same piece of communication that journalists up and down the country received from AC Nielsen. The company found that July showed an 11% drop in cigarette sales and August showed a 2.8% drop in sales. But they don’t just stop there – they also interpret the results, as any company of that statute that works with statistics for a living would do.

Predictably AC Nielsen puts the drop down to the smoking ban and also on the awful weather we had in early summer. This naturally explains the big difference in number between July and August since during August the weather picked up and people could have more of an opportunity to stand outside to smoke. Yet AC Nielsen being responsible people also stressed something else that journalists seemed to forget as well – that last year we had the Football World Cup and people not only tended to smoke more but they tended to drink more. It’s not only tobacco sales that are down you see, it’s alcohol as well.

Yet – what sounds better as a headline? “Tobacco sales down not only from smoking ban” or “Tobacco Sales tumble to the abyss following smoking ban”?

Ahhhhh – the joys of statistically induced hysteria.