July 20, 2008 6:17 pm

One simple way to predict a victor

  • Share
  • Print
  • Clip
  • Gift Article
  • Comments
Cole illustration

One cannot help but be struck by the current disconnection in US presidential politics between, on one hand, the excitement and enthusiasm that attend Barack Obama’s candidacy and, on the other, the tightness of the race according to recent polls. The first suggests a sweeping victory for the Democrat in November, the second a close election and the distinct possibility of a win for John McCain.

Most would agree that Senator Obama has so far waged a polished and efficient campaign. He stumbles occasionally. His endless iterations on troop withdrawals from Iraq, for instance, have given Senator McCain a valuable opening. But he recovers quickly from his missteps and despite them projects an air of competence and assurance that belies his lack of experience.

Mr McCain, in contrast, has all the experience one could wish but little, these days, of the composure and gravitas that it is supposed to confer. When he trips up, the error seems to stick and he looks ridiculous. Mr Obama looks presidential. Mr McCain, sad to say, does not.

Mr McCain has to contend with the US media’s increasingly shameless bias in favour of Mr Obama. And his position is inescapably awkward – not much liked within his own party, forced to concentrate on keeping the Republican vote intact. Also, Mr Obama has the edge in campaign tactics. Handled right, as it most probably will be, this week’s foreign tour could prove a masterstroke in that regard.

Yet look at the polls. A recent Gallup reading says that Mr Obama’s slender lead has narrowed; last week Rasmussen’s tracking poll called the race a tie. State-by-state polling, filtered through the electoral college arithmetic, gives Mr McCain a real shot at victory. All this despite the fact that the incumbent Republican president is deeply unpopular and the economy continues to tank.

How does one make sense of this? The simple answer may get me ejected from the guild of political commentators, who have a lot of space to fill between now and November – but I report it nonetheless. It is that these early head-to-head polls and the vast enterprise of political analysis, nit-picking and minute speculation they support, are, to a first order of approximation, worthless. In short, you resolve the paradox by ignoring them.

Note that if you do, science is on your side. Alan Abramowitz, a politics scholar at Emory University, has shown that summer head-to-head polls convey almost no information about the forthcoming election. (Subsequent head-to-head polls are not much better.) Instead, he has a simple “electoral barometer” that weighs together the approval rating of the incumbent president, the economy’s economic growth rate and whether the president’s party has controlled the White House for two terms (the “time for a change” factor). This laughably simple metric has correctly forecast the winner of the popular vote in 14 out of 15 postwar presidential elections.

The only exception is 1968, when the barometer (calibrated to range between +100 and –100) gave Hubert Humphrey a wafer-thin advantage of +2; he lost, with a popular vote deficit of less than 1 percentage point. The barometer not only picks winners but pretty accurately points to winning margins, too. In 1980, Jimmy Carter had the biggest postwar negative reading (–66); Ronald Reagan beat him by nearly 10 percentage points.

President George W. Bush’s net approval rating (favourable minus unfavourable) is currently –40; the economy grew at a 1 per cent annual rate in the first quarter; and Republicans have had two terms in the White House. Plugging the numbers into Mr Abramowitz’s formula gives the Republican candidate a score of –60, about as bad as it gets: second only to Mr Carter’s in the annals of doomed postwar candidacies. The barometer says Mr Obama is going to waltz to victory.

Why has this barometer been so much more accurate than the wisdom of Gallup? That is hard to say – but as a factual matter, its superiority is indisputable. Even if you do not buy it, it ought to inform your reading of the polls. A wide winning margin, which is what the barometer predicts for Mr Obama, renders moot all the detailed electoral map analysis of swing states, solid states, toss-up states, states leaning one way or the other. All this wonderful stuff might matter if the margin in the national popular vote is thin. If it is wide, the toss-up states move together and that is that.

The unsettling thing about this way of predicting the outcome, of course, is that it does not matter whether the Democratic candidate is Mr Obama or Hillary Clinton – or Joe Biden or Dennis Kucinich, for that matter. The Republicans’ choice of Mr McCain was equally beside the point. On the merits, one candidate may be much better than another – a separate and endlessly interesting question. When it comes to predicting the result, the barometer says that as long as the incumbent is not running, it makes no difference.

Are there special factors that could throw the calculation off? No doubt, and this year one above all cries out. Mr Obama would be the first black president, a possibility the barometer has not yet had to contemplate. Who knows what difference his colour will make, whether it will help him on balance or hurt him. History suggests neither; that the choice of candidates, their strengths and weaknesses and the way they fight their campaigns, matters less to the outcome than one might suppose and infinitely less than the political commentariat is honour-bound to maintain. History suggests Mr McCain is toast.

Send your comments to clive.crook@gmail.com

Read and post comments at Clive Crook’s Washington Blog

Copyright The Financial Times Limited 2016. You may share using our article tools.
Please don't cut articles from FT.com and redistribute by email or post to the web.

  • Share
  • Print
  • Clip
  • Gift Article
  • Comments
SHARE THIS QUOTE