Plenty of election stories in this morning’s papers, but the majority are the sort of warmed-over generic previews that could have been written any time in the past few weeks (or, with some name changes, months). If you’re looking for hard news, the most interesting might be the piece by Phillip Coorey in The Age, foreshadowing a preference deal between Labor and the Greens.

Sure enough, a press release from the Greens this morning confirmed the story: Labor “will direct its Senate preferences to the Australian Greens ahead of all other political parties”, and “Local branches of the Greens have chosen to direct preferences to the Labor party ahead of the Coalition in a number of lower house seats”.

Even a large part of this could have been written months or years ago. Every election, Labor and the Greens circle each other warily before signing up to a deal on preferences that turns out to amount to very little. But it’s worth thinking about just how important those preferences might be this year, and what each party hopes to gain.

The Greens’ attitude is powerfully shaped by the experience of 2004, when Labor announced a preference deal and promptly ratted on it a day or two later, dealing instead with the Assemblies of God Party, Family First and ultimately delivering a Victorian Senate seat to Steve Fielding instead of the Greens’ David Risstrom (and almost doing the same thing in Tasmania).

This year, as I explained last week, the Greens would have to be very unlucky not to end up with the balance of power in the Senate. But they are haunted by the fear that Labor will again pay heed to its own fundamentalist wing and do something to deprive them of one of the necessary seats. That may be an irrational fear — I think Labor was burnt badly by the Fielding experiment — but as long as they hold it, Labor Senate preferences will be worth going out on a limb for.

In the lower house, Labor preferences are irrelevant — there is no seat where they are likely to be eliminated before the Greens. But Greens preferences, as everyone reminds us, will be critical to Labor’s fortunes in a host of marginal seats.

But beware the equivocation here, from “preferences”, meaning where votes actually go when they’re counted, to “preferences”, meaning what a party says on its how-to-vote cards. They’re not the same thing. The first matters a lot, but it doesn’t follow that the second matters much, if at all.

It’s scary how many people still don’t understand this, but preferences in the lower house, unlike the Senate, aren’t controlled by the parties. While some parties have mostly obedient voters who just follow the how-to-vote cards, others, such as the Greens or previously One Nation, attract a more ornery crowd who make up their own minds. And of course the smaller the party, the less likely its voters are to get given a how-to-vote card in the first place.

Greens voters mostly come from the left, and Greens preferences flow strongly to Labor regardless of what the party says on its how-to-vote cards. Directing preferences to Labor rather than leaving them open might make a difference of about 5%, maybe less. In a typical outer-suburban marginal, where the Greens vote isn’t very much to begin with, that’s only going to be about 0.2% or 0.3% of the total.

In a close election, of course, that could still swing enough seats to be decisive, and even if this election doesn’t look like being close, party strategists have to operate on the assumption that every seat matters. More to the point, Labor was never likely to do anything with its Senate preferences other than give them to the Greens, so the commitment costs it nothing (and, as 2004 demonstrated, it will rat on it anyway if it turns out to be inconvenient).

That extra 0.3% or so might not make a difference, but getting it as a free bonus is still a pretty good deal.