Remember when the pollsters made a hash of the 1992 election. Could that be happening in the US? twitter.com/MSmithsonPB/st…

— Mike Smithson (@MSmithsonPB) November 1, 2012

The night when John Major defied all the predictions

As can be seen from the screen-shot above the exit poll at the 1992 UK general election was very much out – as were, with one notable exception, every single campaign poll.

Everything pointed in April 1992 to John Major losing and the then 13 year period of Tory government coming to an end.

Only one of the final polls, Gallup, showed a Tory lead and that was of just 0.5%. In the end the blue team won a majority and a GB vote lead of 7.5%. This was an appalling night for the pollsters and led, over the decade and a half that followed, to massive changes in the way that political polling is carried out.

At the moment with the US election, as we’ve been reporting continuously on PB, individual state polls have been pointing to a victory for the incumbent.

This has prompted attacks on the US pollsters on a scale that I’ve never witnessed before.

The charge is much in line with what was being said at the UK polling inquest 20 years ago that the samples are lop-sided to one particular side.

There are, of course, massive differences between the UK and US not least the proliferation of robo-calling and polling firms such as Rasmussen and PPP which make no secret of their allegiances.

My main betting objective is not to lose any money and I’m yet to go all in on Obama.

Mike Smithson

For the latest polling and political betting news from the US and UK

Follow @MSmithsonPB