Mark “Reckons” Thompson’s statistical analysis appears to show a correlation between seat safety and the likelihood that expenses will have been abused. Everyone’s quite rightly impressed by this, including Polly Toynbee and now Guido. First comment post-Guido link? “Ow my head hurts now.”Priceless.

I wish to sound a note of caution, however. We’re not quite at the point where we can say that this is now a fact to be deployed by people like Guido and Polly to support their arguments.

The sample size is taken from the hundred or so MPs that the Telegraph have covered so far. Expenses abuses don’t immediately leap off the page and announce themselves, so the the Telegraph has tackled the data in a very specific order: Biggest Names First.

They’ve apparently had a team of over 20 journalists working on the story full time (in full ‘research mode’). They’re not going to tackle the data in alphabetical order. They’re going to start with Cabinet members, then Shadow Cabinet members, big beasts, big names.

The thing about big names and cabinet members is that they’re more likely to be in safe seats than not. Could this be the real source of the bias towards ’safe seats’ observed in the statistical analysis of the currently limited data sample?

The Telegraph have announced their intention to investigate every single MP and that’s important – it means that just because we haven’t heard about an MP doesn’t mean that they’re clean. There’s still something like 530 more MPs to go, the vast majority of which no-one’s ever heard of.

What will be interesting is to redo the statistical analysis once we know about all the MPs and the ‘Telegraph looking for big names first’ factor will have been neutralised. I still think what Mark’s done is amazing – it’s just not fact yet. To his credit he’s not yet claimed it is – but others are beginning to, and that worries me.