With 24 hours until the final presidential debate, the race couldn’t be tighter. Whatever commanding lead Barack Obama enjoyed prior to the first debate has long gone, and his more respectable performance in the second debate failed to do anything other than stem the bleeding.

In September, Obama enjoyed solid leads, or a competitive position, in all swing states save for Indiana — a pseudo-swing state, which had jumped to the Democrats in 2008, for the first time since 1964. After the first debate, the numbers started to shift. One week after the event, it was clear North Carolina could be all but written off by the Democrats — Mitt Romney now enjoys a 4% lead there. Then Florida began to peel away — Romney now enjoys a 2-3% lead there, consistent across several polls. Virginia and Colorado moved in from Obama leads to a knife edge, and then New Hampshire, a solidly Democrat state with occasional cross-over, became a swing state.

Wisconsin, which had veered towards swing state status — when homeboy Paul Ryan was nominated as GOP VP candidate — had gone out again, but is now coming back in. There were even suggestions Michigan and Pennsylvania were coming in as swing states. Until this week, such talk was dismissed by pundits on both sides as overblown — especially since the Republicans are running no ads in Pennsylvania, and few in Michigan. But now Ryan is making a few appearances in Pittsburgh and surrounds, and Romney’s strategists may be contemplating an ad buy there. They would be likely to do so if the Democrats impose a triage model on the campaign, and effectively withdraw from Florida, in order to focus on Ohio, Virginia, Ohio, Colorado and Ohio.

The volatility of the race has produced some truly weird polling behaviour — a process magnified by the failure of pollsters to keep up with the changing nature of communications. Pew Research has swung from an eight-point Obama lead to a four-point Romney lead in two weeks. While most polls see the race as tied, or at most two points either way, Gallup has had Romney up by six and seven points repeatedly. The eventual poll will show whether Gallup is ahead of the pack, or its methodology hopelessly wrecked by the abandonment of landline phones, and the absence of any central registry of numbers. Each polling organisation is trying different methods to allow and compensate for what is effectively the collapse of a centralised public sphere — and also experimenting with cost control in a system where a call flat rate no longer exists.

This election is the one where the polling model that has been in place since 1948 starts to really fray and fall apart. Ironically, modern polling began with an almighty error — forecasting that Harry Truman would lose to Republican Thomas Dewey, an error based on over-reliance on phones, which only about 40% of households possessed. Now, the anarchy of communication is rendering authoritative judgments equally difficult. People have too many phones, not enough. The landline is now decisively age-shifted, but by now it has become more complicated than that. What about those who use Skype on a smart phone? Or who use the “magic jack” flat rate online phone number? Or those who never answer a blocked call? And so on.

The point is that for most of its history, private sector polling has relied on a certain relationship between the state, society and telecoms (door-to-door polling is not done by any major polling organisation) and that has now been dissolved. Its dissolution is part of a wider transformation of social life — into an atomised order, in which the state and social institutions are the distant horizon of fragmented and partial social relations, rather than an imminent presence where people feel involved in the collective practices of national social life. Just in case people feel this is a tad too pretentious, I should add that I am writing this between the 3pm and 6pm showings of the wet T-shirt contest at Dirty Harry’s bar during the Daytona Beach Biketoberfest weekend, so there you go, and FWIW.

“That would be the ideal result, in a way. Obama would regain power, but the country would be forced into thinking about its mad system, and doing something about it.”

In any case, the radical uncertainty of the polls is having various effects. The most important is something that your correspondent pointed out to Crikey readers months ago, something which the mainstream coverage in the US is only starting to pay attention to, and which the dim bulbs in the Australian MSM appear not to have cottoned onto at all yet — the possibility of a split between the popular vote and the electoral college vote on November 7. In my original piece I explored both sides of this — thus including the possibility that Obama might win the popular vote, but lose the college.

That now seems highly unlikely — but increasingly possible is that Obama will win the college via Ohio, and a couple of small states, and lose the country overall. This is reflected in the numbers — Obama is still polling at evens or slightly better in key swing states, but polling badly overall, especially among white males. This is especially so in the south, where his numbers are appalling, a fact which has nothing at all to do with white masculinist racism, nothing at all y’hear me boy?!

Obama’s southern numbers — in some states he is trailing among white males by 22% — point to a truth about the electoral college system. In some situations it can act as a sink for votes, where a massive turn-off can have no effect whatsover. What does it matter if Obama’s vote falls from 30% to 20% in Mississipppi or Utah? He was never going to win those states anyway. And ditto, if the situation was reversed, in California and New York. Millions of votes can go to waste, because of the electoral college system, which is the true hallmark of federalism (compare our pseudo-federalist system, where we add together 150 seat results, regardless of state boundaries). So there is a real possibility that Obama will win the election around 300-258 electoral college votes, while losing the popular votes of millions of, well, crackers, who, as I speak, are still at the wet T-shirt contest at Dirty Harry’s.

That would be the ideal result, in a way. Obama would regain power, but the country would be forced into thinking about its mad system, and doing something about it. The supposition among many has been that a popular vote/electoral college split is so rare as not to worry about — but that discounts the degree to which the political system has shifted from interest to ideology.

Vermont would quite possibly do better under a Romney administration — but it will never vote for Republicans, of this stripe, again. Mississippi desperately needs some basic public services, but it will retain its steadfast self-defeating addiction to poverty, stupidity and the Republican Party no matter what is on offer. The fact that voting is now cultural, rather than economic or interest-derived, plays havoc with an electoral-college federalist system — and no-one, neither the Beltway professionals nor the Nate Silver-esque nerds understand this.

Bring it on, I say. Force the contradictions. And, sheerly as a perk not the thing itself, watch the Republicans seethe under a minority vote president. Well, anyway, we shall see. Now, back to Biketoberfest.

*Guy Rundle is travelling the US with the candidates as the election heats up. His exclusive reports are only available to Crikey subscribers — sign up here