Presidential election polls, October 1976

Forty years ago, this coming Tuesday, very many Americans, who normally enjoyed taking a little bet on this or that, refuse to gamble. They complained that all the fun had gone out of presidential elections because a scientific device had appeared that could decide ahead of time who was going to be the winner. It was something called the public opinion poll and it was initiated by a magazine known as the Literary Digest.

The Literary Digest announced that it had discovered the secret of the old Greek Oracles. But in modern society, the oracles were not left to wise old men on mountain tops, they'd been run through something very subtle and scientific called a sampling system. The Literary Digest announced a few days before the election – Franklin Roosevelt was running for the second time – that he, who was hailed as the national liberator in 1933, the father of the New Deal, the white hope of a Western world already beginning to cower before Hitler, he was going to lose to the Republican nominee, the governor of the state of Kansas, one Alfred Landon, a wiry, mild, engaging little man. 

To most people, Landon's victory would have seemed a preposterous dream a year before and even during the campaign it looked like a very doubtful prospect. Roosevelt was the master of a strategy and a tactic. His strategy was to appear at the helm of the ship of state, bear his big face and his huge shoulders to the howling winds of depression and swear that election, or no election, he'd dare not leave the bridge. He succeeded superbly in appearing not only presidential, as we now say, but in appearing like the only president the country dare depend on. His tactic was his well-known mastery of the new medium, known then in Britain as the wireless and later as sound radio – a very odd phrase since it's practically impossible to imagine radio without sound. 

No statesman, before or since, could manage these two stratagems with equal mastery, to appear like a god and to sound like a favourite uncle. Mr Landon, on the contrary, looked like a dapper Kansas farmer and he had a voice with the decibel volume of a mouse. But the Literary Digest was apparently on to something which in those days seemed to us to promise a scientific revolution on a par, today, with the computer. 

Well, the election was held. You know, of course, it's not decided directly on the popular vote, that's to say the totals each state is assigned, according to its population – so many electoral votes which go in each state to the man with a majority of the popular vote. And a majority of the electoral vote is what counts. In 1936, there were 531 electoral votes to fight for. When the votes were counted, Roosevelt had 523 electoral votes and Landon, the Oracles' scientific favourite, had eight. Well, that finished for a time all public trust in polls. It incidentally finished off the Literary Digest. It folded. 

Well, we know that since then, though politicians are loathe to believe it, there has been a revolutionary advance in statistical method. By composing a true, statistical sample, and it is possible to do that with a sample of, say, 3,000 people representing a voting population of 80 million, the best of the pollsters – Harris, Roper, Gallup – cannot possibly drop a brick so large as the one that squashed Alfred Landon and killed the Literary Digest, but they've had enough experience to know that the polls and the computers that can now tot up their results can not yet digest all the factors that go into an election result. 

And let's just list quickly some of the things that affect the way people vote: the balance of rich and poor and the comfortable in the middle; the proportions of these people who have registered and are qualified to vote; the actual number which did get out and vote; the range of weather across the continent on polling day – it could be 90 degrees in Phoenix, Arizona and ten below zero in Upstate New York; the conflicting prejudices of the so-called ethnic groups inside any given community – how, in Chicago, say, the Polish Americans will offset or strengthen the Irish Americans, how the Czechs come into it; the health or sickness of the industry or crop by which any locality makes its living; the local unemployment rate; the momentary fame that was brought to a town which one or other of the candidates visited; the effect on the voter of seeing one or both of the candidates' wives up close; whether you think Ford is looking more or less presidential every day and whether you suspect that Jimmy Carter dyes his hair. 

A score of other considerations, not least this year. The wary second thought about what sort of president the second man would make – and this may weigh more heavily than it's ever done, for we are constantly and properly reminded that of the last six presidents, three of them, Truman, Johnson, Ford, were never expected to be anything higher than vice president. Ford, only a little over two years ago, was never expected to go higher than he'd gone as the Republican leader in the House. 

So the pollsters... the pollsters hedge their bets and there's nothing abject or sinister in that. It shows nothing but good sense and a healthy regard for what is imponderable in life, in the life of the voters. The pollsters say that they can be wrong by as much as four per cent. In fact, and in the past 20 or more years, they have rarely been wrong in this country by more than one or one and a half per cent at most. Still, to be wrong by one and a half per cent about the winner's total can mean an overall three per cent error and we keep hearing from commentators in states with the big totals of electoral votes – Ohio is a good example and has decided several presidential elections – we keep hearing that in this section of the country the election is too close to call. We do hear that Ford has written off Massachusetts, the only state to go for the Democrat in the thunderous 1972 landslide of Richard Nixon. Remember him? And even the Democratic leaders in California are uniformly glum and fear that that big state – it has 45 electoral votes, the most – is lost to them. 

But we could go on all night and day matching one politician's hunches against another's. I've gone into the polling business not for its intrinsic interest, if any, but because it has taught us that the newspaper editors favourite hobby – what are the decisive issues? – is not a responsible method of looking at an election. It's an escape hatch into an orderly world of ideas that isn't, and perhaps never was, on land or sea. 

Well, I'm sure there are listeners who are beginning to pick their teeth and smack their lips with impatience. When is the man going to get down to the really important things that will decide the prosperity and the fate of American in the next four years? Which man is most likely to halt the spread of nuclear arms and, on balance, how do the voters feel about this? How do Americans feel about the new leadership of China and which man do they trust to handle it skilfully? Do Americans trust Ford or Carter more to respond effectively to a new Arab oil boycott? 

This is the sort of thing that editorial writers love to muse over. These are the questions that well-informed men want answered by the majority of the American voters. Well you know something? The spread of on-the-spot reporting that television has made possible and, much more, the refinement of the statistical sample has made one thing brutally plain. There is no such thing as a well-informed man or woman outside the tiny community of their own friends and the people they buy things from and nobody knows the answer to those questions about nuclear arms, China, the Arabs or any other political problem that will face the next president. 

We can make a fairly certain guess that if Ford goes back to the White House, we shall have four more years of the same but, even then, once he becomes his own man, actually elected for the first time to the presidency, he, too, could develop surprising strains in his mental chemistry. Carter, or rather what Carter might become, is a puzzle – a sausage, as they used to say in Lancashire – beyond the probing of any poll or statistical sample. 

In the final survey, the polls are agreed on one thing: the election is closer than anybody'd dreamed even three months ago when we were making jokes about calling off the election and inaugurating Carter right after the formality of the Republican convention. I have as many hunches and ideas and suspicions as anybody about why this is so, but I don't remember an election in which the strength of the known candidate was so uncertain and the effect of the unknown candidate on most voters was so dicey. 

We do know that there's a third man who, more than any other American, could decide the outcome in Ford's favour. And he is old Eugene McCarthy who battled and lost for the Democratic nomination in 1968. He's on the ballot in 28 states as an independent and the polls give him between five and six per cent of the vote. In big states where the outcome is close, a good showing for McCarthy would draw votes from Carter and throw the majority to Ford. And then there's the high percentage, eight or ten, of the undecided. This, I think, could be Carter's nemesis. For to be undecided at this late date must mean, in most places, undecided about Carter. But it might just mean undecided whether to take a flyer on the new man and ditch the old. The polls say rather nervously that Carter has a slight edge, but Ford is edging closer every day. 

So, to sum up! There are 126 candidates for president representing such exotic parties as farmer labour, socialist, conservative, the know-nothings, the single tax party, women's lib, temperance, Cuban, freedom party, senior citizens – you name it. On Tuesday night, I confidently predict that of those 126 candidates, 125 of them will lose. The next president of the United States, without a shadow of a doubt, will be the man christened James Earl Carter, known as 'Jimmy', or the man christened Leslie King Junior, known by grace of his stepfather's adoption as Gerald Ford.

This transcript was typed from a recording of the original BBC broadcast (© BBC) and not copied from an original script. Because of the risk of mishearing, the BBC cannot vouch for its complete accuracy.

Letter from America audio recordings of broadcasts ©BBC

Letter from America scripts © Cooke Americas, RLLP. All rights reserved.