Forget the campaign horse race: It doesn't matter.

THE MEDIA OFTEN GET criticized for focusing too much on the "horse-race" aspect of politics, as opposed to more substantive stuff: those alleged "issues." In its usual form, this criticism is not compelling. If horse-race stuff affects the outcomes of elections, then it is substantive. And, anyway, it's of more interest to most readers than policy stuff, no matter what they say.

The problem with the media's treatment of horse-race stuff is that we get it all wrong. Most specifically, we grossly overestimate its importance in determining the outcomes of elections.

This is as true on editorial pages as on news pages. When we monitor campaign ads for accuracy and fairness, it is because we think there's a horse race going on, and our job is to keep the competitors within some rules. When we are freed to write columns, as opposed to editorials, we go right to the horse race, talking about this guy's image problem with women or that guy's money problem. It's because we think this stuff matters. This assumption is hardly ever questioned. It should be.

If somebody could prove to you that the outcomes of all elections can be predicted before the campaigns even begin -- based upon conditions that exist before the campaign -- wouldn't that cause you to have doubts about the importance of campaigns? Wouldn't you begin to wonder whether we are wasting time obsessing about TV ads, strategies, money, Dick Morris, and who's up and who's down in the campaign staffs?

Well, nobody can tell you that all election outcomes can be predicted half a year in advance. But a remarkable number can be. If you look at enough factors -- like how the country is doing when an election occurs and what the qualifications of the candidates are -- you can come up with so many good pre-campaign predictions that you will start to lose interest in campaign minutiae. You will realize that much bigger forces are at work in American politics than who has the best clothing consultant.

New predictive model

Professor Allan Lichtman, a political historian at American University in Washington, D.C, offers the best predictive scheme I've seen in a couple of decades of moderate watchfulness. Lichtman started with the suspicion that presidential election outcomes are determined by the circumstances the country finds itself in when an election arrives. Specifically, he suspected that if the people find those circumstances to be about as good as can reasonably be expected, the incumbent party wins.

Testing this theory, he thought up a lot of factors that might, conceivably, be statistically associated with victory. Then he looked at past presidential elections to see whether they, in fact, are. He eventually settled on 13 that turned out to be particularly useful. Lichtman also came up with a similar method to predict the outcome of Senate races.

Take a look at the factors in the adjacent boxes. Then contemplate whether it makes sense to deal with campaigns the way we always have. Lichtmann has a book, by the way, The Thirteen Keys to the Presidency, which he co-authored with Ken DeCell.

NCEW member Martin Gottlieb has been an editorial writer and columnist for the Dayton Daily News for 16 years. Knowing that he can't fully sell the ideas presented here in a short article, he is trying to write a book on the subject.

The presidency: Keys to predicting the winner

PROFESSOR ALLAN LICHTMAN'S system picks the winner when applied retroactively to every election since 1860, which is to say every election in the history of the current two-party system. Lichtman has also applied it in advance to every election since 1984, getting every one right.

Some of the 13 "keys" relate directly to how the county is doing; some don't.

The keys are phrased as true/false questions about whether the incumbent party has a certain advantage. A false answer counts against the incumbent party. Six falses and it loses. (There's nothing magic about the number six; that's just the historical pattern.)

In interpreting the keys, keep in mind that "major" means "MAJOR." For example, in Key 7, only two modem presidents get credit for enacting major changes in national policy: Lyndon Johnson and Ronald Reagan.

The keys:

1. After the midterm elections, the incumbent party holds more U.S. House seats than it did before the previous presidential election.

2. There is no serious contest for the incumbent-party nomination. (In a serious contest, the challenger goes to the convention with more of the first-ballot votes.)

3. The incumbent-party candidate is the sitting president.

4. There is no significant third party or independent campaign. (Third parties hurt the incumbent party. Exception: A splinter from the other party such as Pat Buchanan.)

5. The economy is not in recession during the election campaign.

6. Real per-capita economic growth during the current presidential term equals or exceeds mean growth during the previous two terms.

7. The administration has effected major changes in national policy.

8. There is no sustained social unrest during the term.

9. The administration is untainted by major scandal.

10. It has suffered no major failure in foreign or military affairs.

11. It has achieved a major success in foreign or military affairs.

12. The incumbent-party candidate is charismatic or a national hero.

13. The other candidate is not charismatic or a national hero.

The Senate: Winning factors

ALLAN LICHTMAN'S SCHEME for predicting the outcomes of Senate races has been correct in more than 90% of the individual races in five of the past seven election years. Overall, his success percentage is in the high 80s. If you can find a better system (a.) go with it and (b.) e-mail me.

The keys to the Senate are phrased as true/false questions, wherein a true favors the re-election of the party holding the seat. The incumbent party needs four trues to win.

The keys:

1. The incumbent-party candidate is the sitting senator.

2. The incumbent-party candidate is a major national figure.

3. There is no serious contest for the incumbent-party nomination (i.e., the nominee wins at least two-thirds of the vote cast in the first primary).

4. The incumbent-party won the seat with 60% or more of the vote in the previous election.

5. The incumbent-party candidate outspends the challenger at least 10%.

6. The challenging-party candidate is not a major national figure or a past or present governor or member of Congress.

7. (Mid-term elections): The incumbent party doesn't have the presidency.

7a. (Presidential-year elections): The keys to the presidency predict the victory of the incumbent senator's party.

8. (Mid-term elections): There is no serious contest for the challenging-party nomination (i.e., the nominee gains a majority of the votes cast in the first primary and beats the second-place finisher at least two to one).

8a. (Presidential-year elections): The party holding the seat has a majority in the lower house of the state legislature.