The answer to “what happened?” in Saturday’s state election is “not much.” Only three seats (of 93) have definitely changed hands, and two of those were just Liberals winning back from independents what should be safe Liberal seats. The opposition has only made one gain from Labor, Tweed, with a slight chance of a second in Port Stephens.

Another two Labor seats are still at risk to independents; at this stage my guess is Labor will hold Lake Macquarie and lose Newcastle (the reverse of what the ABC website is saying.) There is also some doubt about Coalition v independent contests in Goulburn and Dubbo, but it looks as if neither will change hands.

The Liberals have maintained their lead over the Nationals, even though (unless Port Stephens tips over) they have failed to take a seat from Labor. The Nationals did well to hold Murray-Darling and take Tweed, but they got a fright from an independent in Barwon. The Greens improved to 8.8% of the vote, still behind the Nationals, and actually outvoted Labor in one seat (Peter Debnam’s Vaucluse: the first time, I believe, this has happened in a general election.)

The overall swing will take a while to work out, as final results come in. On Saturday night, Antony Green was saying 2.7% to the Coalition; my quick and dirty calculation from the ABC website gives me 3.8%. Today’s Sydney Morning Herald says 3.2%.

In any case, it’s a good result for the pendulum, which said just one seat would fall with a swing of 4.0%.

It was also — much as it pains me to praise anything coming from the Daily Telegraph — a good result for the Galaxy poll, which, despite being the newest of the leading pollsters, has now been the most accurate in each of the last three state elections.

As usual, the “why” is much harder than the “what”. Labor, especially federal Labor, is claiming that industrial relations was a big factor. Since Labor state governments were doing well before Work Choices, and still are, that claim is a bit implausible (it was supported by — surprise! — a union-commissioned exit poll, but there may still be some truth in it.)

John Howard, yet again, maintained that the election was about state issues. In a general sense he’s probably right; indeed, other things being equal, the survival of unpopular Labor state governments should be a plus for him.

Nonetheless, most people don’t distinguish strongly between state and federal politics, so the failure of the Coalition to make up much ground is a bad omen for the federal government. It may be that some of Kevin Rudd’s phenomenal popularity has rubbed off on Morris Iemma.