For a long time now, I have concerned myself with longterm betting strategies. It's been something of a personal battle for me.

Some of you may remember the series I wrote some years ago on 'power' betting, wherein I made the point that flat-stakes betting was really a way of betting against yourself.

As then, I think the point remains valid. Consider this: The first four choices in a race will usually win about 76 per cent of races but you can't make a profit betting them at flat (level) stakes.

You might win 8 out of 9 bets and end up losing, probably not much, but a loss. Add to this your expenses, and you are well down.

My suggestion was proportional betting, a form of Dutching. Since then, I have continued to use that approach with success.

Because it takes into account expenses and losses, you can determine the amount of your bet depending on the price available of your next selection. A win recoups any losses.

Although this approach has served me well, I have always toyed around with various staking formula's. Despite its critics, I think betting has something going for it. I can't see anything wrong with a progression that enables you to break square or make a small profit if you strike a certain-priced winner at any stage of the progression.

Critics will say that a long losing run will prove the killer ... but that applies with any form of staking, be it level stakes or whatever. Losing streaks can knock anything over.

The key is to recognise a slump early and build in a brake on your betting to counter it. This is why the 6-Point and 10-Point Divisor plans are so useful for small-bet punters. They have built-in safety brakes which prevent stakes getting out of hand in a losing run.

In recent years, I have concentrated on applying my form analysis skills to big-race betting. I choose my races carefully. They are usually Group One or Group Two races. I might have only 15 to 20 bets a season. I ensure they are prime selections.

Those of you who follow the Race Forum page will have noted that I backed Toledo in the recent Stradbroke Handicap at Eagle Farm. This was a wonderful 16/1 winner. It was my best 'score' in the past 12 months. Might And Power has provided me with a nice win or two as well.

In the past two years, my strike rate on these big races has been around the 1 in 3 mark. Bearing this in mind, I decided to develop a betting plan that could take advantage of this strike rate.

Knowing that I could secure a winner in every three bets, and that I wanted a 50 per cent 'earn' on very dollar invested, I drew up a plan that would do this for me.

It required that I accept that the price of 7/ 2 was my 'fair' odds about a selection. Not that I don't bet selections at shorter prices. I do. But I always bet them with a bet proportioned to the 7/2 price; that is, 22 units (based on the bookmakers' percentage table for odds of 7/2).

My next step is that when I secure what I consider a 'gross' overlay, as with Toledo, I double this bet. Thus I had 44 units on Toledo in the Stradbroke, a big bet of $440 (my unit is $10) which returned me a fine profit at the 16s.

If we take this form of betting to other levels, we would come up with the following:

No. of Times      
You Win
Expected Profit
On Dollar Bet                   
  25% 50% 75% 100%
1 out of 3 11/4 7/2 9/2 5/1
1 out of 4 4/1 5/1 6/1 7/1
1 out of 5 11/2 13/2 8/1 9/1
1 out of 6 13/2 8/1 10/1 11/1
1 out of 8 9/1 11/1 13/1 15/1
1 out of 10 12/1 14/1 16/1 19/1

The prices listed are the ideal ones you would need, given your strike rate with winners.

For example, you seek a 50 per cent profit and your strike rate is 1 from 5 (20 per cent). If you have, say, 10 bets, you would bet each horse at 13 units (anything from $1 units upwards), again based on odds percentages.

Ten bets would be an outlay of 130 units. Two winners at 13/2 would return you 195 units, a profit of 65 units, 50 per cent on outlay. Should your winners pay more than 13/2, then obviously you would win even more.

It doesn't, of course, always work out sweetly. Your selections may well be at shorter prices. With two winners from 10, the winners will need to be at a minimum of 4/1 each for you to break even.

What I aim for is a flow of good priced winners, with as many big overlays as I can secure. This is a way of approaching your betting that should enable you to have a fighting chance of making a profit or, at the least, holding yourself level on the financial front.

By Alan Jacobs