If we gather together all 199 polls taken since January 2008, sort them on the basis of the ALP two party preferred vote estimate, and then measure the percentage of times that every two party preferred result has occurred, something pretty astonishing happens.(click to expand)

allpolldist1To read the chart, just pick an ALP two party preferred result on the bottom axis and it tells you the percentage of polls that have registered that exact TPP figure. We can break that down even further into Phone Polls (consisting of Newspoll, Nielsen, Morgan Phone Poll and Galaxy) and non-phone polls (Essential Report and Morgan Face to Face)

polldisttypes

As you can see, the phone polls skew to the lower end while the non-phone polls skew to the higher end – although some of that is to do with the non-phone polls being in the field more times during the (dare I say it) “Honeymoon” period between January and Easter of 2008.

We can take these numbers and turn them into a cumulative distribution curve, that let’s us see the proportion of times that the Government has polled greater than or equal to any given two party preferred figure – and we’ll do this for all 199 polls, and then again for just the phone pollsters:

allpollcumulative phonepollcumulative

Remember when a party getting 55% in a poll created headlines of impending doom for their opposites? To show just how blase we’ve all become lately, 97% of all polls taken during the Rudd government have shown the ALP to be on a two party preferred off 55% or greater. Even if we just use the phone polls, 91.2% of all phone polls have shown the ALP to be on a TPP of 55% or greater.

Poll after poll, these extraordinary results roll in and we all just go “Oh yes, there’s another one“.

Landslides have become normal.