Labour leadership poll: Have we learned nothing?

This morning's YouGov poll in the Times on the Labour leadership election is giving me flashbacks to polls at two other recent elections.

The pollsters' post-mortem on what went wrong in their predictions about this year's general election is still under way.

I wrote recently about what happened at the first public evidence session for the British Polling Council's inquiry.

One of the themes that emerged there was whether people who responded to political polling were unrepresentatively enthusiastic about politics.

Now, you would expect people who are eligible to vote in the Labour leadership contest to be unusually interested in politics. But the full details of the poll show that of the 1,056 people polled, 80% said they would definitely be voting in the leadership election while another 15% said they would probably be voting.

I know that the rules for the leadership election have been changed, but it still seems a very high turnout.

Compare it with the 2010 leadership election, when about 72% of constituency Labour Party members voted.

The other poll it reminds me of was the one from YouGov just before the Scottish independence referendum that suggested "yes" was in the lead.

At the time, I pointed out that the results of the poll were in stark contrast to what the bookmakers were saying, with their odds still strongly favouring a "no" result.

There is a lot to be said for looking at what the bookmakers say, because they are not just giving opinions, they are prepared to back them up with cash.

Currently, the odds for the leadership election have Andy Burnham way ahead on evens or 11/10, while Jeremy Corbyn and Yvette Cooper are considerably behind on about 5/2.

And one final word of warning about this poll. My colleague Norman Smith has tweeted: "Team Burnham say if only party members included (not affiliates) then @Andy4Leader beats Jeremy Corbyn 50.5 - 49.5 %".

Remember we're talking about an election that's not even being held until September, and even the 53% to 47% by which YouGov predicts Mr Corbyn would beat Mr Burnham in the final round of voting looks dangerously close to the margin of error.

To be claiming to be ahead by half percentage points just looks silly.