As has now been pointed out many times, perhaps most forcefully by Antony Green, conducting a redistribution while an election campaign is under way is sheer madness. But the Australian Electoral Commission took the view that the legislation leaves it no choice in the matter, so it has been doing just that, releasing draft new boundaries for Victoria last Friday.

This really will have to be fixed. While Green possibly overstates the amount of confusion that will be created among ordinary voters, it’s a completely unnecessary diversion for the AEC and political professionals of all sorts. There’s no reason why any redistributions required during the life of a parliament shouldn’t all be conducted at the same time, well away from an election.

But, since we’ve got the draft boundaries, what do they mean? Victoria’s seat entitlement is unchanged on 37, but the changing pattern of enrolments made it logical for the commissioners to abolish a rural seat and create an urban one in its place.

They’ve done so, I’m pleased to say, in very much the way I suggested: Murray disappears, Mallee and Indi both take slices of its riverine territory, and McEwen shifts northwards to take Shepparton, losing its weird rural/urban mix. That leaves room for the creation of a new seat, Burke, in the north-west of Melbourne (it’s basically the northern half of the old Calwell plus the Shire of Macedon Ranges).

Despite the fact that they abolish a Liberal seat and create a Labor one, these proposals are actually bad for Labor. That gain is more than outweighed by the fact that Labor-held McEwen, Coragamite and (just barely) Deakin would become notional Coalition seats, for a net loss of two.

Several seats would move the other way, but not enough to change hands; perhaps most noticeably Aston, which has been given a strangely unmotivated extension to its south (mostly Endeavour Hills), would become a very marginal Liberal seat. The changes would also shore up Labor’s majority in Chisholm and probably in Melbourne Ports.

Redistributions are very much a matter of swings and roundabouts. Labor has had a good run lately, notionally gaining seats in the recent changes in Western Australia, Queensland and New South Wales. Now that run of luck seems to have come to an end.

It’s not unheard of for Victoria’s boundaries to favor the Coalition. In the 1996 election, for example, Labor won a narrow majority of the two-party-preferred vote in Victoria, but only 16 of the 37 seats. A decade later, however, that advantage had reversed itself; in 2004 it won a majority of the seats with only 49% of the vote.

In the 2007 election, Labor’s vote across the state was 54.3%, but its vote in the median seat, Bendigo, was 56.1% (on an ideally fair set of boundaries, those two figures would be the same). As both votes and population shift, the numbers change; redistributions throw another unknown into the mix, but in the long run things tend to even out.

And in the end it is the voters who make or break governments, not the boundary commissioners. There are plenty of cases of redistribution change turning out to be less important than everyone thought; sitting member influence and local patterns of campaigning play an important role. Candidates in places such as Deakin would be best focused now on winning votes rather than worrying about what their seats will look like next time around.

Anyone who wants to object to the proposed boundaries has until Friday August 27 to do so.