They claimed to have used some kind of "average" ticket price as the basis for their calculation, but Simon was the first to spot that the Arsenal price used was actually that of the most expensive ticket.

If they'd used a more significant figure (such as the cheapest ticket, surely a better reflection of the cost to the average punter) then Arsenal would have come out looking very good. As it was, Arsenal came out worst, and the papers were full of "Arsenal is most expensive day out" headlines, which will have raised an eyebrow on the faces of all Gooners who travel to away games!

Well, credit where it's due. Nationwide have withdrawn the claim (the whole report in fact), and apologised to the Arsenal. See Total Football for grovelling quotes.

Using the wrong numbers is only one of the problems with the survey. They did the wrong things with them too. The "average" ticket price used was something they called the "median", which they defined as the mid-point of the range of prices (ie half the sum of the cheapest and dearest tickets). Anyone who's done any statistics will know that this is a) a stupid statistic to use b) not what is generally meant by "the median", which is a common and well-defined term in statistics.

Speaking as a mathematician, I'm hoping whoever Nationwide paid to conduct this survey gets their comeuppance. It's rubbish like this that gives us a bad name. The truth is, you can't "prove anything you like with statistics". But you can pretend to, and then laugh all the way to the bank....