Let me deal, firstly, with one of the simplest possible examples of using both tipsters and the pre-post betting market. This is a relatively simple technique and it will probably allow you to move on to more complex techniques in regards to 'converging factors', the subject about which I have been writing these last four months.

We all have to walk before we can run and very much so in the rather tricky area I am dealing with at the moment. The example that I am about to give will act as an introduction on how to set about applying converging factors to your betting benefit.

In this example, I shall make some elementary assumptions. Let's say we are dealing with the daily double at a major meeting (though it could be any meeting). I assume that:

(A) The winners of the 2 races will be in the first 3 horses in the pre-post betting market.

(B) That 3 chosen experts, say Scout, Harry and Form, will each give a winner in one of the two races.

The assumptions may be ambitious, or not, but I merely want to show you how we could deal with the two races, using the above assumptions.

At the moment we are dealing, then, with A and B as probabilities.

So let's take a theoretical race.

First 3 in the pre-post betting:

FIRST LEG: Lilianje, Valdivia, Norcrest.

SECOND LEG: Gallant Turk, Trim Curry, Happy Wanderer.

If we term them 1, 2, 3 in order of betting, we have nine possibilities for the two races:

1 2 3 1 2 3 1 2 3

1 1 1 2 2 2 3 3 3

Next, the tipsters' selections for the two races:

SCOUTHARRYFORM
LilianjeValdiviaNorcrest
G.TurkTrim CurryG.Turk

So far, so good. We have all the data. What is the procedure? Well, we have taken assumption A that the winners will come from the first 3 horses in the betting, and we have considered only those horses.

We have got the selections into terms of our figures instead of names of horses and we have considered only those horses.

Remember, the possibles, in those terms, are:

1 2 3 1 2 3 1 2 3
1 1 1 2 2 2 3 3 3

Now, if Scout is to give one winner, and he has 1-1, then, from the nine possible doubles, we shall not require the following combinations:

2-2, 3-2, 2-3, 3-3.

The reason is clear: They do not agree with Scout at all. Now, if Harry is to have a winner, with 2-2, then we shall also get rid of 1-3, 11, 3-1, because they don't agree with Harry.

And if Form is to give a winner, with 3-1, then out goes 1-2 because it does not agree with Form.

We have only one double left: 2-1. In terms of horses, Valdivia and Gallant Turk. Valdivia was Harry's top selection in the first leg while Gallant Turk was the top pick in the second leg of both Scout and Form.

Remember that this approach is all about making assumptions -and then reducing your overall bets to accommodate the assumptions coming true.

Using the tipster tricks a bit further, let's select a 'main' expert and assume that he will give us at least one double in the races selected. The races chosen will be those in which our expert, and another expert, will have their selections in the first 3 in the prepost betting market.

Let's assume a theoretical meeting at Sandown, with 4 races covered: The selections of our main tipster (Scout) and the other expert (Oracle) are as follows, numbered according to betting position:

SCOUT 2-2-3-1 ORACLE: 2-3-2-2

They both, for the 4 races, give selections that are in the first 3 in the betting. So we can consider taking all combinations.

But we are basically concerned with the picks of the Scout. If he takes all possible doubles on his selections, there can be six (races 1-2, 1-3, 1-4, 2-3, 2-4, 3-4).

But I always assume that no one expert will have it all his own way. So I will take only the races in which he disagrees with the other expert, so far as doubles are concerned.

There will, then, be only these doubles: Races 2 and 3, 2 and 4 and 3 and 4. They both agree on race one.

This is the essence of the approach. We only use the main tipster's selections in races where he disagrees with his fellow expert/s. Remember, too, we are backing only the selections of the main tipster, no-one else's.

• This is the completion of this series of four articles on converging factors. The series was compiled using material published some 40 to 50 years ago in Britain, under the title Converging Factor Racing Formula, by Promath.