If the weather report is accurate less than about 75% of the time, then the weather report doesn’t increase your certainty over the weather at all. If you don’t have sufficient confidence in the Met Office, they can say it will be sunny today, but you’ll take your umbrella with you anyway.

Watching the weather report, therefore, gave you no benefit whatsoever.

In fact, there’s a new website called ‘Umbrella Today?‘ which has turned the whole idea of reporting the weather upside down. Instead of predicting the temperature, the pollen count, the pressure and the precipitation, it gives you one single piece of information: do you need to take your umbrella with you today, or not.

The folks behind Umbrella Today? have looked at the traditional weather report and asked ‘what is the information that you are looking for, when you watch the weather’. Quite simply, you want to know if it’s going to rain, so you can decide whether to take your umbrella with you.

And that’s what they tell you.

A lot of attention in the political geek community is given to opinion polls, where a self-selected bunch of strangely fickle political enthusiasts answer a series of questions concerning which party they will probably vote for.

So much attention, in fact, that there are a number of popular blogs whose only Unique Selling Point is their poll-watching.

A common theme, in poll reporting, is that the ‘winning’ side of any particular poll is allowed the smug satisfaction of believing that the polls show their views and ideas are popular, and by extension, the best ones. The ‘losers’ say that the polls don’t really mean anything anyway.

There’s a problem with that, in the words of Blackadder: its bollocks. I’m going to write this in big text so that my point is very clear.

Polls are utterly meaningless.

They provide no useful information whatsoever. There is no strategic advantage to watching political polling whatsoever. There is no policy-making advantage. There is no statistical advantage. There is no tactical advantage.

After reading a poll, you have just as much information as you have after watching a weather report. Less, in fact, because polls – like the weather – can be misleading and are influenced by factors which may not be immediately apparent.

Let’s be clear on this: I want Gordon Brown to lose the next election. I want that very much. I dislike the Labour Government and wish them to fail. To a lesser extent, I’d like the Conservatives to win, but Labour losing is of utmost importance to me.

And so, you’d think I’d be happy that the Conservatives have recently been consistently ahead in the polls recently. Wouldn’t you. Evidently those polling results show that “their views and ideas are popular, and by extension, the best ones”.

But they don’t. And I just wanted to make sure I’ve said that, on record, while the polls are showing what I want them to show.