The world media, and especially U.S. media, are following with intense interest and concern the November presidential elections in the United States. Almost all the stories discuss which of the two principal candidates, Donald Trump and Hillary Clinton, is likely to win and by what margin. The media also are filled with explanations of the polling results, which of course vary over time.

However, almost none of the coverage of the election poses the question, who does the respondent expect to win regardless of the respondent's own preferences? We do not know how many persons feel certain about their prediction. Whatever the number today, it is likely to grow as we approach the final moment of choice. My guess, and it is really only that, is that perhaps at most one-third of the electorate will feel they know what the results will be. Please keep in mind that feeling sure about the victor is quite distinct from feeling sure about one's own preferences.

The most obvious consequence of advance certainty affects those voters who are sure that their preferred candidate is certain to win. It is one that the candidates themselves always fear. Voters who feel sure that their preferred candidate will win may think it unnecessary to make the effort of actually voting. This is why candidates engage in elaborate efforts to get their pledged voters to actually vote.

We might call this the "laziness" factor. Sociologists call it self-defeating prophecies. Will such voter "laziness" affect Trump voters or Clinton voters more? It is hard to say because the "laziness" factor varies according to the intensity of voter preference. This factor is being publicly discussed in terms of the negative preference of the voters. Do Trump voters or Clinton voters contain a higher percentage of persons whose fear that the other candidate might be elected determines their vote? And why?

There are many more consequences of voter predictions than the "laziness" consequence. Take the case of voters who value a so-called balance between the legislature and executive branches of the U.S. government. If such voters feel sure about the victor, they might shift voting from their preferred candidate to the other candidate in order to obtain a "balanced" political result. Here we are entering a zone of much opacity. It is anyone's guess how much the "balance" factor will change the final vote of voters.

Another factor is what I would call the desire to send a "message" to the victor and to the party that performs most strongly. A Clinton voter may be more ready to support a third party candidate if it won't throw the election to Trump. Ergo, if such voters were sure that Clinton would win, they would feel it was "safe" to use their vote to send a message.

Voters who supported Bernie Sanders may then vote for Jill Stein of the Green Party, Gary Johnson of the Libertarian Party, or simply abstain rather than voting for Clinton. Similarly, Trump voters may vote for Johnson, abstain, or actually vote for Clinton. Or if sure that Trump will win, they may concentrate their energy and money on congressional candidates.

The point is that when individual voters think they can predict with certainty an inherently uncertain outcome, such self-assurance can change the real consequences in complicated ways. The combination of what I call the laziness, balance, and message factors should make us all less confident about how we pursue our voter preferences and how we try to persuade others to make the same analysis that we are making.

That brings us to the last factor, which I shall call the factor of the "importance of voting" at all. There are many eligible voters who are skeptical that voting makes any real difference in what happens after the election. This group may be subdivided into those who feel it is of no importance at all and those who waver on this question. The waverers may be persuaded not to vote for their only mildly preferred candidate if they feel they know the outcome but not if they feel uncertain about the outcome.

What of the candidates themselves? Do they feel they are sure who will win? It seems not. They both act as though they were nervous about the results of a close election. Trump is alleging that he has the votes and therefore ought to be declared the winner. He says that if he is not declared the winner, it must be because the system is "rigged" and therefore he will have been "cheated" of his victory.

This argument may be simply a way to make sure his supporters come out to vote. Or it may be a face-saving explanation of anticipated defeat. In addition, he has called upon his supporters to watch the polls for cheating, which may be a way of intimidating "minority" voters from coming to the polls. The most likely reason is that he is preparing the ground to challenge the legitimacy of a Clinton victory, and thereby to continue the campaign after the elections, in anticipation of future elections.

Clinton is also arguing the uncertainty of her victory. In her case, the explanation is simpler. She is genuinely uncertain. She therefore needs to stimulate her supporters not to be "lazy." More important, she is probably seeking to discourage "message" voting. And she is making a public case against "balance" voting. Finally, she wants to keep the spotlight off her own weaknesses by turning eyes toward Trump and his weaknesses. Trump's egomania, which makes him seek always to keep himself in the spotlight, helps Clinton in this regard.

Once again, this is not an analysis of who will win or why one should prefer a certain behavior or why other people support a given behavior. It is simply an effort to factor into the results the ways in which the certainty some voters have or will have about the outcome will affect the actual outcome.

- Immanuel Wallerstein, Senior Research Scholar at Yale University, is the author of The Decline of American Power: The U.S. in a Chaotic World (New Press).

Copyright ©2016 Immanuel Wallerstein -- used by permission of Agence Global.