The first system I ever developed was for the trots (35 years ago) and I suspect it was suited to that code due to the smaller fields and because it relied on current information as its main force. I designed it, tested it over three years' results, and it looked solid enough.

It worked and probably still does I have used it with horse-racing but I believe it belongs to the harness-racing era of my life. Why I did it is another story for another time.

It went like this: I looked at the tipsters' most favoured column for the three main chances, and the horse at the longest price was bet eachway. This gave me an average win price of around 7/1 and enabled me to prove to myself that I could actually come up with a system that worked.

(Please note: In 23 words I have just given you a system that took me six months to produce!)

For horse racing, at that time, I used to bet on the daily doubles in the following way. I would look at the tipsters' panel and, ignoring the most favoured column, I'd take every horse mentioned by each tipster in both races as a multiple double. If the winner was mentioned somewhere by one of the tipsters, then I got the double. It worked.

For greyhounds, I used the trots plan but only had win bets. The annual results were not good enough to produce a profit.

Read PPM magazine and you will find lots of these types of systems. Any of them are worth a look. If you like the feel of what you are reading, and don't believe it will be difficult for you to determine the selections, then start researching it for a potential plan for yourself.

Once you start your own research you'll come up with other ideas that can be applied to the original plan. You'll end up improving the original out of existence and either it will hold up or it will not.

In all cases be aware of win ratio and percentage return as these factors are your best guides to forecasting future results and betting bank requirements. Most good systems produce winners 20 to 25 per cent of the time and will show a percentage return of at least 10 per cent.

Assuming your ambition is to earn a sufficient income from betting on the races, and that you are prepared to develop your own plan and build up a betting bank, you should understand what these things actually mean.

Let's assume you are betting on 10 races per week and your betting bank is $5,000. This means your unit bet is $50 (1 per cent of $5,000) and the weekly outlay is $500. The annual outlay is $25,000 ($500 x 50 weeks), which translates to an annual income of $5,000 if you are getting a percentage return of 20 per cent.

The problem is that, although you have doubled your bank, the $5,000 is not really enough, is it? So you have to continue to plough it all back and compound the bank by increasing your bet by $1 as each $100 is made.

If your ambition is to earn $50,000 per annum, your betting bank has to be $50,000. It will get there but you need to be patient because some years won't work out as well as other years. Your unit bet in this case is $500 and your weekly outlay is $5,000. The annual turnover is $250,000.

The time it takes to reach this stage will depend on your selection skills and the compounding effect of the betting bank.

That's the easy part as it's all rather mechanical. But when you are betting in $500, or even $300, units, with a potential outlay of $5,000 for the 10 bets, your attitude is going to change.

You may be influenced by the fact that the last five bets have lost and lose your nerve, not wanting to throw away the next $500. To deviate from your plan could be a disastrous move.

If you have lost confidence in your system and want to stop betting, either to do more research or to throw it away and find another one, well, okay. But assuming you've already built up most of this betting bank from your plan (and remember, things do go wrong for everyone when it comes to betting), my advice would be to hang in there or stop betting and give the whole thing a rest.

You can always come back and pick up where you left off. There will always be another race meeting.

I now come to point one, which was about developing your own system. This must be done correctly and you must be completely satisfied about two things: The expected win ratio (at least 20 per cent) and the expected percentage return. The development of a system is easy enough and you can bet on it with only $1 units, assuming you have at least $100 as working capital.

Here's one of my ideas to get you thinking.


  1. Bet on Saturdays at Sydney, Melbourne and Brisbane metropolitan tracks only.
  2. Do not bet if the track is slow or heavy.
  3. Look at TAB No. 1.
  4. Its last run must have been at a city track in the same State as today's race.
  5. It must have run a place in its last race (1st, 2nd or 3rd).
  6. Its last run must not be more than 28 days ago.

This approach will reduce the potential bets from 24 to a reasonable number and give you every chance of succeeding.

You can add some refinements to it as your research continues. I must point out that the research and development of your system should be enjoyable and be considered a labour of love. If you're successful - and there is no reason at all why you will not be if you persevere - then your rewards are immense.

If you're half successful, then you won't be too disadvantaged. If you're unable to develop a satisfactory system, then you can simply give it all away and be wiser and richer by not betting.

There is, then, much to gain from succeeding and little to lose from not succeeding.

In one of his articles, The Optimist said that for about five weeks in each year we really make waves and for the rest of the year we are struggling to stay above water. I wrote to him, stating that the article was pretty sobering stuff, and I believe its message is worth repeating.

Researching a system means testing it out on paper to see if it works and then to see if it can be improved, particularly if it's not working.

I've seen systems that work well and then go down big time. Most systems have their cycles, a point which must be taken into consideration. It may be that a particular system never makes money in the months of June, July and August, for example. Now, it's going to take you a long time to determine this, as you may have to see two or three years of results before making such a decision. How's your patience level?

The professionals know a lot about when not to bet. This is a refinement process rather than an elimination rule. They know what tracks do not turn a profit and bet accordingly (maybe trimming the size of their usual bets). They know what track conditions to avoid and are aware of many other traps that await the average punter.

They've learned the hard way and have taken a lot of time to analyse and refine their betting procedures.

You must do the same and be prepared to allow for the time it takes to see where you are going wrong or where things can be improved or tightened up.

The fact is that a $5,000 betting bank can compound to an $80,000 betting bank in five or six years. That's not a long time, really, and it allows you time to find the weak spots in your system.

They are there, believe me, but time is the only way you will determine what they are. So every bet made for a system is clearly documented before and additional information is noted after the race meeting. In time, this information becomes valuable. Patience is always a virtue in the racing business.

We often read about 'when not to bet' scenarios, but they will only apply to one particular method or system. You can't blindly adopt their principles to your system without confirming the outcome over an extended period of time. Sorry ... no shortcuts here.

Refinements are different in that your system is either making money, breaking even or losing on an annual basis. A refinement is not used to turn a losing system into a winning one; that would require an additional, or an altered, elimination rule. The refinements simply tighten things up a bit and, of course, improve the annual profit and, perhaps, the win ratio.

You'll learn that each system will lend itself to all kinds of changes, which can become confusing. For example, let's say you're following a system, hopefully one you have developed yourself, and you see that restricting your bets to TAB numbers 1 to 4 does absolute wonders for your win ratio and percentage return.

I've noticed this myself. Whilst this may be true, what you may be doing is simply reducing the number of bets the system throws up and the true factor to be considered is the weight a horse is carrying compared to its previous last few starts.

The analysis process is quite challenging and does require you to think things through. Fortunately there are many people out there who will help in this area now. Including me.

Computer spreadsheets are an excellent way of entering your betting information and applying 'what if' scenarios to see what can happen.

Some other aspects (which could be costing you money) worth considering are the following:

  1. Midweek meetings.
  2. Provincial meetings.
  3. Public holiday meetings.
  4. Sydney, Melbourne or Brisbane - one State may be costing you money.
  5. Racetracks.
  6. Race distances.
  7. TAB numbers.
  8. Favourites.
  9. Apprentice jockeys.

And the list goes on.

• Keith Roth continues his series in the August PPM, and writes about specialising and increasing your win-strike rate.

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

By Keith Roth