The state seat of Melbourne is absolutely irrelevant in terms of who forms government, yet it is centre stage in one of our most fascinating modern political dramas. A byelection on July 21 will deliver the next episode.

Melbourne has been Labor’s for 103 of the last 104 years. That’s about as safe as safe gets — or it was until the Greens stepped into the arena in 2002 and started happily munching away at the ALP primary vote.

Between 1999 and 2010 the ALP vote has dropped from nearly 60% to just over 35%. The Greens first contested the seat in 2002 and achieved nearly 32% in 2010.

A recent Roy Morgan poll of current voter intentions found the Greens would win the byelection without breaking a sweat. The poll gave them 54% two party preferred support on a first preference vote of 48.5%. The poll gave Labor a primary vote of 37.5% with the independent vote split between Stephen Mayne on 7% and the rest of the field holding a similar figure.

As per regular practice in unwinnable byelections, the other major party (in this case the Liberals) are not contesting. They had just under 28% of the primary vote at the last state election.

A Greens primary vote of 48.5% would give them the seat regardless of other factors. That feels like an over-inflated figure given their 32% result at the last state election and Adam Bandt’s 36% over the larger federal electorate of Melbourne.

Presuming their actual vote is in the mid to high 30s, the Greens will need a significant swag of preferences from elsewhere. Some or all of these could come from Mayne, other small independents and/or further erosion of Labor support.

However, the biggest reserve of votes to be liberated is that of Liberal supporters. With no candidate to vote for, the question is who will they support? Some will not vote or vote informally, but the majority will make a choice between the options on offer.

Some will choose Mayne or other independents, but they still need to assign their preferences. What this means, in the end, is a choice between the two serious contenders. These conservative voters will be the ones will ultimately decide the outcome in Melbourne.

Traditional analysis might present this scenario as an unappealing choice between the lesser of two evils. Hardly worth voting when your own party isn’t even willing to enter the race.

Much like open primaries in the US, this also presents an opportunity to meddle in the political fortunes of your opponents. If Liberal voters will decide the outcome, then these voters have a powerful weapon at their disposal.

If Liberal voters back Labor over the Greens, they maintain the status quo. This means the Baillieu government’s slim parliamentary majority remains and business continues as usual. Disruption and unpredictability are minimised.

The positive here is the denial of the Greens in the lower house. There are plenty of Liberals who believe the rise of the Greens is a far greater threat than even the ALP and would endorse a policy of shutting them out — as was done formally via party preferences at the last state election.

On the other hand, if Liberal voters back the Greens, they will deal a historic blow to their mortal rivals. With Bandt already ensconced at federal level, a Green victory would evict Labor from its own heartland.

That would bring on another round of navel gazing about the direction of the party and possibly even a crisis of confidence in Daniel Andrews’ leadership. It would reduce the ALP team by one and prevent them introducing fresh talent into their ranks.

In the longer-term it intensifies the existential dilemma and political wedge facing the ALP as it fights on both sides of the political spectrum. The public fuss about the Greens from Labor in the last week shows how easily a party under threat can be distracted from the main game of fighting the Coalition for the hearts and votes of the middle ground.

It seems a reasonable assumption that Liberal voters in Melbourne are better informed, educated and politically savvy than their suburban counterparts. This is a chance for them to have some real influence with their vote, something of a novelty for inner-city conservatives. To what end they use that influence, is another matter altogether.