The most important principle of race-track betting is obvious and axiomatic. Yet it is one that most horse players violate regularly.

A bettor should make his largest wagers on the races in which he holds his strongest convictions, when he sees the greatest potential profits. In less promising situations he should bet less -- or nothing at all.

Nongamblers would assume that any rational man would behave this way. But in the emotive atmosphere of the track, the size of horse players' wagers is influenced more by their mental state at the time of a race than by a rational calculation of risks and rewards.

A bettor who has just doped out the last two winners is more likely to make a strong bet on the next race -- regardless of his insights into it -- because he is temporarily feeling omniscient. A bettor who has just lost two tough photo finishes in a row and has seen his bankroll dwindle may feel gunshy at the very time he ought to be aggressive.

Similarly, a horse player may decide on a type of bet -- win, place, show, exacta triple -- because he feels conservative or greedy at a particular moment, instead of making an intelligent risk-reward judgment.

We are all human and are all susceptible to such emotional weakness. Yet anyone who hopes to win at the track -- or even to survive at the track -- must formulate some kind of a rational, consistent betting strategy.

The first principle of any such strategy is that a handicapper must possess a cogent opinion before he bets at all. When he does isolate a horse who looks like a winner, he must determine whether win, place, show or exotic betting will offer the greatest value.

Part of the answer is easy: place and show betting are losing propositions.

The mathematics of the game are such that place and show bettors get less of a return on their money than win bettors. Pick horses by any method and you will find that, in the long run, betting them to win will always provide a greater return than place or show.

The place and show pools offer an illusion of safety, not value.

The choice that a bettor must confront in almost every race is between playing a horse to win and playing him in the exotics. Here, too, he must judge whether one type of wagering offers a clearly greater value than another. j

A horseplayer who has a strong opinion in the ninth race at Bowie or Belmont should know that the state and track take an extortionary 25 percent from all the money bet in the triple. Betting solid horses to win ought to be much more profitable in the long run than playing them in the triple.

There are some situations in which the exotics may be especially lucrative. One arises when a "hot horse" gets strong last-minute betting action that drives down his odds. The late money will almost always affect his win price more drastically than payoffs in the exacta or triple, making the exotic pool an attractive value.

Maryland bettors are (or at least ought to be) familiar with trainer Richard Curtin. At Bowie this summer, he unleashed a first-time starter named Rampo, who was bet down to 8 to 5. A 7-to-1 shot ran second and ordinarily the exacta would have returned $30 or so. This time it paid $57.40 -- an enormous bargain.

In most situations, though, a horse-player cannot generalize about whether the win pool or the exacta offers greater value. He should analyze the board showing potential exacta payoffs, though few bettors do.

Few could deal with the following question:

In an eight-horse field, one horse is going to the post at odds of 7 to 2. The exacta payoffs coupling him with the other members of the field are $40, $50, $56, $80, $100, $100 and $225. Is he a better value in the win pool or the exacta?

It is not too difficult to compute a horse's equivalent win odds in the exacta pool. I do it by jotting on my program the amount of money I would need to bet on each combination to get a return of $1,000. In the case of the exacta combination paying $40, I would need to buy 25 $2 tickets -- an investment of $50 -- to receive $1,000. The compilation for the whole field looks like this: (TABLE) Exacta Price(COLUMN)Investment needed to return $1,000 $40(COLUMN)$50 $50(COLUMN)$40 $56(COLUMN)$36 $80(COLUMN)$25 $100(COLUMN)$20 $100(COLUMN)$20 $225(COLUMN)$9 (COLUMN)Total$200(END TABLE)

Thus it would take a $200 investment to get a $1,000 return. The horse is a 4-to-1 shot in the exacta pool. (A $200 bet at 4 to 1 produces an $800 profit, plus the original wager, to equal $1,000.) Anyone betting this horse to win at 7 to 2 would be costing himself money, when he could make exacta wagers to get a half-point higher return.

When there is no clear-cut difference in the value of the win and exotic pools, however, a bettor must make a choice that is more philosophical than mathematical. Should he bet conservatively, or should he take greater risks to earn greater profits in the exotics?

The answer ought to depend on his temperament and his bankroll. If a bettor chooses to play the exotics regularly, he must realize that he will cash fewer bets; there will be many times when his top choice wins and he still loses his entire investment.

Some bettors don't have the resources to play the horses this way. When they miss a payoff that they might have had, they engage in self-flagellation and moan, "I shoulda bet him to win." Such bettors will be happier putting their money in the win pool.

But a few bettors can stoically accept the unpleasant consequences of playing the exotics and console themselves with the knowledge that they will be compensated by some big payoff in the future. They are equipped to bet the horses in an aggressive fashion, and shoot for big profits, which happens to be the optimal way to play the game.