The Patterson-Canberra Times poll got the Labor vote in Ginninderra and the Liberal vote in Molonglo spot on, and reflected the major parties' levels of support in all three electorates reasonably accurately, except the Liberal vote in Ginninderra, which it got spot on a fortnight ago, at 34 per cent, but underestimated by 8 percentage points the day before the election.

The poll was fundamentally sound when you look at the percentage vote across Canberra. Trouble comes when you break the vote into electorates: the smaller the sample, the greater the margin for error.

A simple example demonstrates the point. Say you throw a coin twice in each electorate. That is six throws. To guess a 3-3 heads-tails split across Canberra correctly, you have two chances in six, or 1:3. To correctly guess a 1-1 heads-tails split in each electorate you have a one-in-two chance in each electorate, which across three electorates is two times two times two, or 1:8.

The reason polls are accurate is that the heads and tails (Liberal and Labor) even each other out the more you throw. If you slice the sample there is less capacity for the results to even themselves out.

The problem came in interpreting the results. Greens support in an opinion poll would never translate at the ballot box. The last seat in each electorate comes down to fairly complicated preference flows.

Greens do not attract many preferences from either excluded candidates or from the overflow of elected candidates. You either love 'em or hate 'em.

The other poll was done by students at the University of Canberra journalism school. It had a smaller sample of 300 as against 1200.

(I have to declare my interest here, because I teach journalism at UC and had a fair amount to do with the poll.)

The students did fairly well. The UC poll got the ACT Labor vote within 1per cent but overestimated the Liberal vote by 3.9 per cent. It, too, overestimated the Greens vote for the same reason as Patterson.

But its interpretation into seats was more accurate than Patterson. Patterson had Labor 8, Liberals 5, Greens 4 out by several seats. The UC poll had Liberals 7, Labor 7 or 6, and Greens 3 or 4: spot on.

Both polls uncovered an underlying trend largely unknown in the 18-month lead up to the election a big swing away from Labor to the Greens. If that had been drip-fed with a poll every few weeks (as with the 2007 federal election) Jon Stanhope would quite possibly have met with the same fate as Kim Beazley.

www.crispinhull.com.au