Maybe Victorian election coverage can now turn its focus away from the inner city. Even if (as I claimed yesterday) the Greens can still win seats, the Liberals’ decision to preference against them surely puts paid to any idea of the Greens putting a Coalition government in power.

It should now be clear to everyone that the result will be determined by what happens in Labor’s marginal seats.

There are several good guides to the marginals available, from such experts as Antony Green, William Bowe and Ben Raue. Rather than try to duplicate the detail they provide, my aim in this series is to survey the different areas of Victoria (particularly for the benefit of those not familiar with the geography) to indicate where the key seats are and how the different areas might behave.

I’ll be looking mostly at the 15 Labor-held seats with margins between 2% and 7%. Assuming it wins the three ultra-marginals, the Coalition needs to win nine of those 15 to give it the 44 seats needed for government (since in that case, as explained a fortnight ago, independent Craig Ingram would have no real choice but to support them).

Victoria’s three large provincial cities — Geelong, Ballarat and Bendigo — are spread over nine seats; four of them also contain substantial rural areas or other towns, and another (Lara) extends into the western suburbs of Melbourne. Labor currently holds all nine, but only Lara (on17.7%) could really be called safe; the others all have margins below 11%, and five of them are in the critical group below 7%.

In the state’s second city, Geelong, South Barwon (2.3%) is widely expected to fall, but Bellarine (7.9%) and Geelong (8.3%) are more difficult targets. Ballarat’s three seats are all in the critical group (although Ripon (4.3%) lies mostly beyond the city, extending north to Maryborough and west to Stawell), as is Bendigo East (5.4%), the seat of senior minister Jacinta Allan.

These cities all swung strongly to Labor in the 1999 election that put it into government. Although Labor’s overall swing that year fell short of what the pendulum said was necessary, it got the votes in just the right places. Having picked up Bendigo West in 1996, it won another five in 1999 — finally falling across the line in Geelong by a handful of votes to give it the minimum number of seats to take government.

Labor’s vote increased again in 2002, when it added Bellarine and South Barwon. Not surprisingly, it lost ground in 2006, but not enough to forfeit any seats.

The feeling in 1999 was that the Coalition had neglected regional Victoria, and Labor won support with its promises to restore services in the country and build fast rail links to the provincial cities. Now the boot is on the other foot (as many have discovered, country people are hard to please) and it’s the Coalition that is appealing to disenchantment with an apparently Melbourne-centric government.

These are not rural seats; most of their voters are very much city dwellers. Talk of anti-Labor and anti-Green anger in “the bush” is not necessarily relevant here. But their concerns are nonetheless different from those in metropolitan areas; even Geelong, which is no further from Melbourne than some south-eastern suburbs, is clearly part of regional Victoria.

Labor took advantage of this difference to take government. If the swing is on in the provincial cities, it could lose it the same way. Ted Baillieu needs to win seats on a broad front, but four to six gains among these seats would put him well on the way to victory.

Correspondingly, a failure to make headway there would almost certainly doom his chances.