By Lawrence Alfred Powell
This year's United States presidential election has been widely portrayed in international press reports as a neck-and-neck horse race, with polls alternately showing Barack Obama or Mitt Romney in the lead.
But compelling as that analogy might be, American elections are not horse races, and the two candidates are not 'running neck and neck' either. With three months to go until the election, Obama is ahead in the numbers that matter - though much could still happen between now and November to alter the stakes.
NO HORSE RACE
At first glance, this election does appear very close. As of this writing, the most recent CBS/New York Times national poll has Romney slightly ahead with 46 per cent, to Obama's 45 per cent. The most recent Gallup poll has them tied at 46 each, and Rasmussen shows Obama ahead by 46 to 45 per cent. A similar seesaw pattern has prevailed in various polls for the past few months.
But the outcome of American elections is not determined that way. It's not a horse race, with the winner of a majority of popular votes winning the presidency. The United States does not have direct popular election of the president. Instead, a system of indirect election via an 'electoral college' was established long ago in the 12th and 23rd Amendments of the Constitution, as a compromise between election of the president by a vote in Congress and election of the president by a popular vote of citizens.
This Electoral College consists of 538 electors, equal to the total voting membership of both Houses of Congress (100 senators and 435 representatives), plus the three electors allocated to Washington, DC). Each state's allotment of electors combines the number of members in its congressional delegation - two for the senators (all states have two), and one for each member of the House of Representatives (which varies according to size). This represents a compromise struck between the principles of state representation according to population size (which gives advantage to large states like New York and California) and equal representation of all the states.
A majority of 270 of these electoral votes, not a majority of the popular vote, is what is required to elect the president. Thus it is possible, especially in close elections, for a candidate to win the Electoral College yet lose in the popular vote, or vice versa.
OBAMA AHEAD, SO FAR
If we average the various poll numbers from over the past month, we can get a more precise sense of how this election is actually shaping up. When this is done, Obama averages 47.8 per cent, across eight recent national polls, to Romney's 43.9 per cent, which puts Obama slightly ahead at this point, though not comfortably.
But remember, it's the Electoral College outcome that really matters in the end. In that tally, Obama has at least 253 likely electoral votes to at least 191 for Romney, with 94 votes still up for grabs in swing states that, at this point, are too close to call.
In the above Electoral College map, one can see that if the polling trends of the past few months were to continue through November, Romney would easily pick up the electoral votes of southern states like Louisiana (where Romney leads by 16 per cent), Arkansas (R +20), Oklahoma (R +35), Texas (R +7), Mississippi (R +12), Georgia ( R +9), South Carolina (R +9), West Virginia (R +15), Tennessee (R +6), and Kentucky (R +8).
He is also leading strongly in the Midwest plains states of Nebraska (R +14), Kansas (R +17), Missouri (R +6), Indiana (R +13), South Dakota (R +6) and North Dakota (R +14), and in western mountain states like Montana (R +8), Utah (R +48), and Arizona (+9). Notice that, as in all past American elections since 1980, these are the classic Republican strongholds of 'the south' and 'the west'.
Obama will most likely take the northeastern West Coast, and upper Midwestern states. In the northeast, he is well ahead in New York (O +26), New Jersey (O +14), Massachusetts (Romney's home state, where he was governor! O +19), Connecticut (O +10), Rhode Island (O +17), Vermont (O +25), Maine (O +15), Pennsylvania (O +7), and Maryland (O +20).
He is also running strong in the West Coast states of California (O +17), Oregon (O +9), and Washington (O +12), and in the upper Midwest states of Minnesota (O +11), Illinois (O +14), Wisconsin (O +5), and Michigan (O +6). And in Hawaii, Obama is ahead by 27 per cent.
That leaves seven states up for grabs - Florida, Ohio, New Hampshire, North Carolina, Virgina, Iowa, and Colorado. These are the swing states most likely to determine the outcome of the election. On the map, I've classified these as too close to predict at this point - either because of the slim margin between candidates, or a historical tendency to vacillate back and forth unpredictably between the two parties in previous elections.
A reasonable educated guess as to the final shape of the November election is that if present poll trends were to continue unabated, Obama would win re-election by at least a margin of 275 to 263 electoral votes - regardless of whether he wins a majority in the popular vote. This assumes Romney can, ultimately, win the big prize of Florida (29 electoral votes), as well as Virginia, North Carolina, Iowa, and Colorado, and that Obama squeaks by in Ohio and New Hampshire.
This is actually a somewhat conservative projection for the Obama camp, who might conceivably win more than that if he also picks up Iowa and Colorado, in both of which he is currently slightly ahead. In that case, the Electoral College result might be as skewed as 290 to 248.
But for Romney, the hill to be climbed in order to reach the White House looks much steeper. Between now and November, he would have to somehow manage to prevail in both Florida and Ohio, as well as some combination of Iowa, Virginia, and Colorado, to put him over the 270 mark. This combination could give him a margin of 281 electoral votes to Obama's 257. The problem is that, at present, Obama is slightly ahead in all of those states. So the odds so far are in Obama's favour.
POSSIBLE CONFOUNDING FACTORS
But as Jamaica's election last December illustrates so well, a taken-for-granted electorate can be fickle and candidates' stakes can morph rapidly. Emergent scandals, media gaffes, televised debates, vice-presidential picks (e.g. Sarah Palin), the recent Supreme Court decision favouring corporate campaign contributions, the Republican party's voter-turnout suppression (e.g. 'voter ID' laws), voting machine fraud (e.g. 2004), a sluggish economy, unexpected international events - any of these could still alter the game and swing the election.
The outcome in Ohio, for example, could still slip into the Romney column if he chooses Senator Rob Portman as his running mate. This is important because in recent elections, Ohio has been critical, with a longstanding reputation as a bellwether state.
Or, he might do the opposite, make a foolish VP choice - like when George McGovern chose Senator Tom Eagleton of Missouri as his running mate in 1972, without first going through a careful vetting process. It turned out Eagleton had a history of electroshock therapy for severe depression. McGovern went on to lose to Richard Nixon, by 520 to 17 electoral votes.
Lawrence Alfred Powell is honorary research fellow at the Centre of Methods and Policy Application in the Social Sciences at the University of Auckland, New Zealand, and the former polling director for the Centre for Leadership and Governance at UWI, Mona. Email feedback to firstname.lastname@example.org and email@example.com.