It seems daunting, doesn't it, when you look at a race programme of eight or nine races? You could be excused for shrinking from the task of attempting to unravel such a big programme of races, possibly embracing more than 100 horses.

Why not, then, consider cutting that programme? Pull the programme into half, and consider the meeting in 'frames' of four races each. Doesn't seem so mountainous a task now. You could have races one to four as your first frame, and five to eight as the second one. If there's a 9th or 10th race, discard them! You could operate on one frame only, or play both.

My study of statistics shows that within each such 'frame' a winner can reasonably be expected to occur at least once if you are betting two horses per race. The idea of the betting method is to play each frame as a self-contained sequence. You aim to make a profit per frame. Since only one winner is needed to earn a profit, and since the records show that a winner should occur at least once in each frame, the rule is to stop betting as soon as a winner is struck.

One method to accomplish all this is by betting to percentage odds. A friend of mine, who has conducted a lot of research into frame betting, says he concentrates on the 2nd and 3rd favourites. Between them, he has found, they win around 34 per cent of all races (in the period of his survey, anyway, which is 18 months). This means, in effect, that a 2nd or 3rd choice will win, on average, at least one in three races.

The study shows that if races are grouped in frames of four, a 2nd or 3rd choice in the betting will win in approximately 80 per cent of such frames. A win average of 75 per cent-three winning frames out of four-is low. Around 80 per cent, or four out of five, would be closer to the norm.

Our objective is to win within the frame of four races and every time we hit a winner we have a clear net profit on the sequence of bets. Because the possible sequence of bets is limited to four races, the possibility of losing runs is held to a minimum.

Why stop betting in each frame as soon as a winner is found? Experience shows that if betting is continuous then the overall win average will be about the same but the total number of bets will be much higher. It's better, by the way, to operate this frame method by attending the racetrack, where you can take fullest advantage of the odds available. Off-course punters will have to listen closely to radio TAB odds, or get themselves online to the Viatel computer service through Telecom, where they can have access to odds throughout the afternoon from the various State TABS.

Off-course punters can play the method by taking the selections of their favourite tipster-his 2nd and 3rd picks, that is. This can work out very well. There is, though, the possibility of longer losing sequences when using a tipster's selections. The 2nd and 3rd picks in the betting show a more consistent pattern of success.

How to back the selections in the frames? Well, we want to ensure that just one winner will see us making a profit. With this feature, there is a graph showing how to back two horses per race to make a certain profit. This will interest anyone who wants to test the frame betting idea. The method has its critics because it is 'due column' betting, and losses are chased with the next bet.

The figures on the graph are, more or less, percentage figures. You can use them to work out your actual bets. If you do not wish to bet to take out \$100, you can merely take a smaller percentage of the figures. When a loss occurs, you carry forward the amount lost on the bet to the next bet in the frame, and increase your 'amount due' figure, and thus your bets.

Let's have a close look in theory, and then in practice, at the frame betting method, using the graph idea for the staking.

I'll assume you are a small punter and that you are betting to win \$10 a frame. ('Amount due' is the sum of money you hope to win on the frame.)

First winner is struck at 3-1. You have had \$18 on it at 3-1 for a total return of \$72. The frame is complete. You have staked a total of \$64 and your return is \$72. This means you have almost won the \$10 you set out to win. Because of the trimming down of percentage figures to 'round off', the \$10 is just missed. But that doesn't matter because the objective has more or less been achieved. You now go on to your next frame of four horses.

Let's now assume we are betting to take out \$100. I'll firstly take Syd Brennan's tips in Best Bets for December 5, 1987.

First winner is struck, Mardisi at 2-1, giving a total return of \$438. The frame now stands at total stake of \$341, and total return of \$438, giving a profit of \$97, virtually the \$100 you were seeking.

Now, I realise that most punters will not be prepared to bet this big, nor chase target figures as illustrated here. So, how else can we back the frame horses and win, without placing too much at risk? What you can do, is back both horses at level stakes if prices permit. Ignore those races where you cannot achieve a profit should one of the chosen pair win.

Another idea would be to back two units a win on any horse under 5-1 and place one unit on any horse at 11-2 upwards. Let's see what might happen. With Syd Brennan's tips on December 5 ' you would have lost four units on the first two horses, which were 2-1 and 3-1.

In the next race, there would have been two units on Mardisi and one unit on Twisted Logic at 7-1. The return would have been a total of six units from Mardisi (2-1). All up, then, you have staked seven units for a return of six, and there is a loss of one unit. Had you taken Brian Meldrum of the Melbourne Sun as your tipster, you would have had Giggling (201) and La Zest (20-1) as your bets at \$1 each. La Zest won at 20-1, giving you a clear profit on the frame of \$19.

The problem, naturally, is what happens when you strike a frame that doesn't provide a winner? Your maximum loss is going to be \$16 (four sets of bets at \$4 each). You would have to 'carryover' the loss to the next frame, and double your bets to \$4 on the horses at 5s and under and \$2 on those at 11-2 upwards.

Let's assume you lost \$16 on the first frame. You come to the next and you place \$4 on a 5-2 chance, and \$2 on an 8-1 chance. The 8-1 chance wins. You get \$18 back. You have recouped your \$16 from the first frame but, overall, you have staked \$22 for a return of \$18. You are \$4 behind, but that isn't too bad, and easily recoverable.

Had the 5-2 chance won, you would have got back \$14 and that would have left you \$8 down. Let's look now at a real test case. I took the selections of Shane Templeton of the Melbourne Herald. Here we struck the bad straightoff-in his 2nd and 3rd selections in the first frame (races one to four), Templeton missed out on a winner.

The total stake on his selections (two of which were scratched) was \$8. This was carried over to the second frame (races five to eight). On the 5th race, Templeton's 2nd and 3rd picks were Arch Of Victory at 8-1 and Stable Gossip at 3-1, thus bets of \$2 and \$4 respectively. Arch Of Victory won, for a return of \$18. Altogether, in the two frames, the outlay was \$8 plus \$6 equalling \$14, for a return of \$18, a profit of \$4.

A twist on this is to reverse the bets and place MORE on the longer-priced horse and LESS on the one at the lower odds. If you'd done this in Templeton's case, you would have bet \$10 on the first frame. On the second you would have had \$4 on Arch Of Victory at 8-1 and \$2 on Stable Gossip at 3-1. Arch Of Victory would have returned you a nice \$36. This meant you had backed a total of \$16 on the two frames, for a return of \$36 and a profit of \$20.

Now we'll look at another tipster, in another State, using this particular method. I have chosen, at random, Larry Pratt of the Daily Telegraph in Brisbane.

You would have lost five units on his first frame, but won 14 on his second frame, for a total profit of nine units. He had a winner in each frame, the first winner being Mighty Grey at 2-1 and the next frame providing Cheers For Us at 7-1.

Some of you may feel this method has its faults, and that it does not provide the chance of huge profits. This may be so, but there is the capacity to provide regular winnings week in and week out. Smallish profits picked up on a regular basis can swell considerably over a year.

Study the graph closely and remember, if you are betting to a 'due' column, that to find any particular bet the figures in the graph can be used as a percentage. So, let's say your 'due' column total is \$33 and your two horses to be backed are at 4-1 and 5-1. You would bet 32 per cent of \$33 on your 4-1 chance and 26 per cent of \$33 on your 5-1 chance, a total of \$10.50 on the 4-1 chance and \$8.50 on the 5--1 chance. Get the idea?

Another method of playing both the horses in a race is to assume a price of, say, 4-1 about each, irrespective of what price they really are at in the market. Let's say your frame target is \$10. You have two horses to back to win the \$10 and you assume each is 4-1. Thus, you place \$2.50 a win on each. Should you strike a winner, you stand to get your anticipated target as long as the horse's price is right.

If it doesn't, your next target figure drops by whatever you have won on the race. If both horses lose, you carry over the lost stake and add it to the target figure. So your next target sum would be \$15, and you would bet, say, \$4 on each at a theoretical 4-1. They may be shorter or longer than that in the betting. In essence, your bet on the horses is always one-fourth (25 per cent) of your target figure.

By Statsman

Practical Punting – 1988 March