This morning’s opinion polls contained good news for Kevin Rudd. But as usual – and despite the assurance of The Age‘s editorial that “The polling says it straight” – there was some equivocation in the message.

The most striking result is ACNielsen’s claim, in The Age and SMH, that Labor would win 56% of the two-party-preferred vote, a swing of 4% since the last poll and nearly 9% since the last election. That result was said to be independent of whether Rudd or Beazley was leader: the switch to Rudd increased Labor’s primary vote, but at the expense of minor parties whose preferences were coming its way anyway.

It’s hard to see anything in the last fortnight’s events that would justify such a big swing, so the Nielsen result probably should be regarded as an outlier. Nonetheless, as Bryan Palmer’s graph at Ozpolitics shows, the trend over this year is clearly in Labor’s favour.

Beazley’s supporters were claiming that as an argument against change, but the polls have also consistently shown his personal standing as on the wane, and Newspoll today gave Rudd a 43% to 27% advantage as preferred leader. (It blows out to 48-27 when Beazley is coupled with the invisible Jenny Macklin.)

Such polls, however, are notoriously hard to interpret. What matters is the view of swinging voters; it’s no good for a leadership candidate to appeal to people who will never vote for them, and only slightly better to appeal to those who are rusted-on voters anyway. As Peter Brent at Mumble mischievously reminds us, Newspoll back in 1994 picked Bronwyn Bishop as preferred Liberal leader.

Even if we knew that swinging voters approve of Labor’s new leader, it would be nice to know where they’re located. Anecdotally, Rudd is supposed to be good for Labor in Queensland, where it has a swag of marginal seats in its sights. New South Wales could be a harder task, especially since the state election will probably dominate the news there for the next few months. But neither Nielsen nor Newspoll gives us a geographical breakdown.

If Labor really wins 56%, of course, the distribution of it won’t much matter – it will be a landslide of epic proportions. But no-one expects things to be that clear.