In the sixth and final article of our special series, US expert Barry Meadow and PPM editor Brian Blackwell discuss various aspects of punting, including betting on trifectas and "value based" approaches.

BB: We ended last month's chat with you mentioning that handicappers, the rank and file punters, will probably never make a discovery that hasn't been noted and evaluated by one's competitors. What about the staking aspect of things? Trifectas for example?

BM: While picking 1-2-3 order of finish isn't easy, the hard part is that you must bet without knowing whether you're getting value. Unlike exactas and quinellas, when you play trifectas you usually have no idea what the numbers you fancy will pay.

You don't know whether you're betting an overlay or an underlay. Is the 2-4-7 combination returning $68 or $680? All you know is you're booking a high takeout on a blind bet.

That makes the trifecta a much less desirable play than the exacta and quinella. Nonetheless, trifectas do offer the one edge that is characteristic of all exotic wagers the throwaway factor. Thank you, numerology and astrology players!
Others spread bets wildly, buying so many tickets that they lose money even when they hit the winning combination. These "donations" mean free money for sharp bettors.

BB: I think the choice of race on which to bet is all-important, though I see punters betting the trifectas on every race with little thought given to whether it offers them any value. Personally, I prefer to attack the races with big fields. They're tough but they do offer the chance of windfall returns.

BM: You have to decide whether the race interests you. If you don't like the race well enough to make your own betting line on it, forget the trifecta. This helps avoid such desperado strategies as boxing a bunch and hoping something of a longshot comes in.

Trifecta costs can escalate rapidly. Wheel (banker) one horse in a nine horse field and a $2 trifecta costs $112 ($56 for $1 units), which is more than it might pay. While it"s true that occasionally a trifecta comes back in five figures, usually the number is considerably less.

If your entire betting stake is, say, $500, don't bother with the trifectas at all. It will require too high a percentage of your bankroll to play properly.

BB: I should say at this point, Barry, that in some parts of Australia punters have flexi betting available, which means they can bet as much as they like on a trifecta combination, and if it wins they receive a percentage of the dividend. For example, they might back a combination that calls for $100 (using $1 units) but they only want to bet $40. So in the event of a win they would receive 40 per cent of the dividend.

BM: Looking over a large number of trifecta results, one notes the influence of the $1 box bet on the size of the payoffs. When the favourites finish 1-2-3 in any order, the return is minuscule. You're better off taking a shot at the exacta.

The two most common strategies, boxing and wheeling (banker selection), are dismal failures. Boxing, taking your top 3, 4 or 5 horses and betting them all in all combinations, generally yields underlaid payoffs even if the favourite finishes 3rd.

BB: It's all about strategy, I suppose, and one man's meat is another man's poison. I would say that the majority of punters can't be bothered trying to assess a formline. It's long been a condition with punters. But I take your point that if a person is serious about betting on trifectas, then he or she should be prepared to spend some quality time carefully assessing and pricing the main chances.

BM: A good strategy for the trifecta, as it is with the exacta, is to key your top horse for first (and sometimes with the second pick) with the legitimate contenders, not merely throw the horse in with every other runner and pray. To catch big payoffs, use a strange one in third slot, since even living filth may stagger in 3rd with the right trip.

If you have a strong opinion, you love No. 5, think that at least one of No. 1, 2 or 3 must finish in the money, and feel Nos. 8 and 9 are so horrible they couldn't get into the picture even with fudging in the photo lab, you can bet fewer combinations in larger amounts:

5-1-2, 5-1-3, 5-2-1, 5-2-3, 5-3-1, 5-32, 5-1-23467, 5-2-13467, 5~3-12467, 523467~1, 5-13467~2, 5-12467-3.

For $36 (using $1 units), you'II win at least once if you're right, and twice if two of your horses finish in the money.

This all comes down to the basic philosophy of the trifecta. That is, are you playing merely to get it, or to make a big score?

If you're looking to crush, bet more money on fewer combinations. If you typically bet three exacta combinations, you shouldn't bet more than 15 trifecta combinations.

BB: I'm a great believer in the banker, Barry. I like to attack the large-field races and a favourite bet is to pick out my win banker and link it with 6 or'7 to run 2nd and 3rd.

That puts me up for $42 if I take 7 with my banker. I'II only bet though when my No. 1 choice is at very good odds, more than 6/1, and the longer the better. I'll also tailor my outlay on each runner if needs be according to the odds I've assessed on them.

I'll bet only a couple of times a week on the trifecta, so there is no great drama involved in working out individual unit assessments. But I make sure I wait for the right races. I have to think that I can make a lot of money should I be right.

BM: Without a key horse or horses, don't play the trifecta. That's my advice. You've got to focus.

BB: Our talk about win betting from last week is probably worth taking a step further, Barry. The 50 per cent overlay aspect is a concept many might find hard to grasp or believe in.

BM: Well, as I've said, assuming your odds line is accurate, the 50 per cent requirement theoretically ensures at least a 22 per cent edge on every bet. (See the win overlay chart below.)

BB: Yes, we can see from that chart that if you price a horse at evens, then you should be asking for at least 6/4 about it (3/2 in the chart). If your price is 5/2, then you should be looking at securing 7/2 or better. If your assessment is 4/1 on your top choice, then you need to get at least 6/1. These figures seem quite reasonable to me. As you have stated, Barry, the 50 per cent requirement is based on the odds line being accurate, or as accurate as it can be.

BM: The Win Bet Chart below sets out clearly how much you need to invest following my approach. As you'll see from the chart, if you have a selection assessed as a 3/2 chance (6/4) and you can secure 3/1, then the bet called for is $29. If you could get this horse at 4 / 1, then the bet would be $33.

If you make a horse 6 / 5 and the crowd has him at 9/5, you would bet $35. If they send him away at 2/1, raise your bet to $40. If an overlay is not listed on the chart (e.g. your 3/5 shot goes off at 8/5, play the highest number, in this case $80.
The figures on the Win Bet Chart assume that the race is "average" (rated, say, B on an ABC scale of interest in a race). If you rate a race as a C, bet less; if you rate the race an A, bet more.

The numbers were derived by assuming that instead of your odds line being accurate, it is only accurate to the point where you will earn a 6 per cent edge on minimum overlays, 8 per cent on overlays of one extra column, 10 per cent on overlays of two extra columns, and 12 per cent on overlays of three extra columns, and then using half the Kelly recommendation, rather than the full Kelly, for security. Thus if it's in error, it errs on the side of safety.

Note that the largest bets are those with the greatest chance of winning. That is, the horses listed at even money and below. Although your bets also increase as the edge becomes larger, the key is the frequency of wins, not the size of the advantage.

The small bets in the chart (e.g. $6 to win on a horse you rate 6/12 who goes off at 9/1) won't bring you a million dollars in a week. However, their conservatism makes it unlikely that you will go bust. And, assuming your handicapping improves with experience, your capital will grow larger which will enable you to increase your bets.

BB: I notice in your excellent book Money Secrets At The Racetrack that you get into the area of two-horse overlays. How does the average punter work this approach?

BM: At times, two contenders will be listed at least 50 per cent above your odds line. Recently, I checked the results of 12 meetings which yielded 420 races in which I had two or more horses above the 50 per cent overlay mark (of which nearly all had only two horses above the line). Playing both, or all, overlays would have resulted in net profits for nine of these 12 meetings.

My conclusion: If two horses are overlaid, bet them both. When you bet more than one horse in a race, you are guaranteed at least one losing bet, so you must subtract that losing wager before considering whether the bet is still an overlay.

If one horse is 2 / 1 and another is 5/2, even if both are overlays, you'll wind up with less than even-money for your action.

That's why I insist on at least 3/1 for any horse that's part of a two-horse overlay.

Here's how I see things with two horse overlays:

3/2 or below 3/1 8/5 or 9/5, 7/2
2/1,4/1 5/2,9/2
3/1 or 7/2 6/1 4/1,7/1
9/2,8/1 5/1, 9/1
6/1 10/1

If only one horse meets these more difficult overlay requirements, then bet that horse only. For example: Let's say a 6/5 shot on your line is listed at 2 / 1, and a 4 / 1 shot is listed at 8/1. Skip the 2/1 horse and bet the 8/1 chance only.

BB: I know in these situations a lot of punters would want to Dutch the pair to collect the same profit no matter which of them wins. What are your views on this?

BM: Dutching helps prevent long losing streaks, since generally when you bet only one overlay you'll be passing the one at low odds, which is more likely to win. However, you are also laying out a lot more money to collect small profits with Dutching, and often both overlays will lose.

BB: We're coming to the end of this series. I think it's been a most useful discussion with enough ideas in it to help punters trim up their thinking about betting, to make them think twice about some of their bad habits and to instil in them some self-discipline to successfully tackle their racetrack betting. How would you sum up, Barry?

BM: Well  to win you must understand the game well enough to make an accurate odds line for your contenders. Then you must follow a strategy. The key to winning is to find the overlays and bet them correctly.

Underbet and your bankroll doesn't grow as fat as it should. Overbet and you can easily go bankrupt, even if you bet with an advantage. The bet charts we've illustrated have been scaled on the conservative side but as you win you can increase your stakes.

If you've never made an odds line before, it may take you some time initially. As with any skill, the more you practise, the swifter you'll get.

BB: Barry, thanks for taking part in this discussion. It's been great to get the benefit of your years of experience. Unlike a lot of experts, you are there at the rockface, you're betting every day, in big numbers, so you are not talking only from theory.

I'd recommend any Aussie or Kiwi punter to buy your bestselling book Money Secrets At The Racetrack. lt's full of ideas. And it makes a lot of sense.

•    Barry Meadow's website is

Click here to read Part 1.
Click here to read Part 2.
Click here to read Part 3.
Click here to read Part 4.
Click here to read Part 5.

By Barry Meadow and Brian Blackwell