There are plenty of approaches to betting on horses. Most of them are well worn. But my basic idea will travel one that is not so worn. In fact, it begets entirely new methods.

Some will say that nothing matters except form. Statistics are hurled at us in profusion. Then there are certain races to avoid, horses for courses, and horses for distances. Many of these things have some basis in fact.

I am all for reasonably intelligent and logical factors but we have to handle them in a reasonably intelligent and logical manner as well. My view is that it's essential to consider factors in groups.

The 'law of averages' has been decried time and time again, and it's pointed out that something cannot happen because something else has! If we are going to make statements of that kind they must be within certain conditions.

Something may have to happen, provided something else HAS, if certain selected conditions prevail. In certain conditions, the IF must mature!

Perhaps history not repeating itself can be the result of an individual factor, of which we are not aware. But a group's history is always repeating itself. Prophecies concerning groups can be very accurate.

Given similar conditions, we may assume that the behaviour of a group will tend to repeat itself. It may concern a larger or smaller part of the group than it did last time, but the trend is almost always very apparent.

And there is always a sound reason for the trend, if we look closely enough.

Okay, why do favourites win about 30 per cent of races? And second favs account for 25 per cent? Favourites in wfa races may even have a higher percentage. These things are very consistent. They come about every season, pretty near the same. The figures are constant and that gives us a clue. If the figures are so constant, and if groups are so fond of repeating, cannot we stick to figures and groups?

There is, I believe, salvation in a group. A particular 2yo favourite will fail but will a group of 2yo favourites all fail? I do not think so.

The best newspaper expert of the lot can tip a loser and indeed often does but will a group of experts, whose selections vary all fail for a complete meeting? I don't think so.

What we must do, then, is to find a method of turning those group trends to our advantage in a given race, or races. We must ally the best of one group to the best of another. We may assume that while some may fail, all will not do so.

Suppose we deal first of all with numbers. We assume that it is possible for a sequence of three factors to have two possibilities each, so that the full permutation would be 2 x 2 x 2 = 8.

If the two possibilities are represented by ciphers 1 and 2, then the written out permutation would be as follows:

1 2 1 2 1 2 1 2

1 1 2 2 1 1 2 2

1 1 1 1 2 2 2 2

But if we can assume, on reasonable grounds, that the correct or desired combination will have BOTH a 1 and a 2 in it, then we can safely eliminate these:

111 and 222 as both are comprised of the same numbers. We are then left with only six combinations (211, 121, 221, 112, 212,122).

If we also are given any reasonable grounds to assume that there will be a pair of the ciphers 1 in the correct or desired combination, we are able to dispense with three more: 221, 212 and 122. That leaves us with 211, 121 and 112.

If we could again assume that cipher 2 will not appear in the second or third factors, then we will have only one column left and that will be: 211.

This is the one that fulfils all the conditions or assumptions made. Note that our first assumption did not venture to prophesy any final combination; it merely said that there should be a 1 and a 2 in the final reckoning.

It did not rashly guess in what order. There will, of course, be reasons for the assumptions. And the soundness of those reasons will depend on whether our final column is the correct one or not.

If they are sound, then we must be correct. We have, in fact, synchronised varying factors in a bid to bring the full possible permutation down to one single correct column.

What we are after is a definite indication for a given race, or races, from a number of factors that are generally apt to provide winners. If we stick to the group technique we can benefit by the best of them, whilst evading the worst.

In the simple example on numbers, three conditions were specified. If they were all obtained, we would find the correct column. If one or more of our conditions let us down we would fail.

In actual betting, this has not GOT to be so. One or more of our factors can fail and we can still get a winner. We can allow for errors and failures. Successful betting, then, is:

- Dealing in groups
- Synchronising various sound factors
- Allowance for failure, and more allowance for failure.

We can take trebles, for example. If we pick 2 horses in each leg, that means a total combination of 2 x 2 x 2 equalling 8 combinations.

Let's say we bet the first 2 favourites in each leg but we only expect the TOP favourite to win two of the legs and a 2nd fav to win only 2 legs. That means we eliminate some of the combinations.

In all, the linkups would be (1 equalling the fav, 2 the second fav):

1st LEG | 2nd LEG | 3rd LEG |
---|---|---|

1 | 1 | 1 |

1 | 1 | 2 |

1 | 2 | 1 |

1 | 2 | 2 |

2 | 1 | 1 |

2 | 1 | 2 |

2 | 2 | 1 |

2 | 2 | 2 |

Instead of betting 8 combinations, we bet only 6. That means we can ADD to the amount we place on the bettable combinations to boost our profits.

Let's take another example and assume that we expect a second favourite to figure only once in any treble. That means we would bet only the following combinations:

1-1-2

1-2-1

2-1-1

Now we are betting only three combinations. If we can get 2 winning favs up and a 2nd fav wins the other leg, whichever one that happens to be, then we have landed the treble for just a 3 unit outlay.

This is the beauty of making the stats work for you. In next month's article I intend to take this 'converging factors' idea a stage further.

I have stressed that if we make statements to the effect that something must happen because something else has, they need to be statements made in certain definite conditions.

We can be as 'certain' as we like, provided the appropriate conditions obtain.

Example: If there are a lot of socks in a drawer and they are of two different colours, black and white, and you are looking for a matched pair in total darkness, you need take out only three socks to know you have a pair of the same colour. You MIGHT take three of the same colour out in three attempts but you MUST take two of the same colour.

Given the conditions, then:

- A mixture of black and white socks
- The taking out of three socks at random, and a pair of similar socks must ensue!

The fact that you take three out will guarantee a matched pair. If you took only two out, you might very well succeed in getting a pair to match but there is NO guarantee.

Suppose you had socks in three different drawers and in each drawer there was 75 per cent white socks and 25 per cent black socks. If you took two socks from each drawer - six socks in all - you could get three of each, you might even get more black than white, but it is very unlikely.

You are more likely to get at least two pairs of white socks, if not more, from the six socks taken at random. You might take a black pair from one drawer but it is most unlikely that you would do so from two drawers and highly unlikely that you would take a black pair from all three drawers. That is obvious common sense.

The chances of getting at least one pair of white socks from the three drawers are excellent. There are MORE white socks available. The majority of the white socks is an undeniable and existent factor. And we cannot ignore factors with such credentials.

If there were three horses running in the first race and three horses running in the second, then the two races, or the double, must be won by one of these combinations:

1-1, 1-2, 1-3, 2-1, 2-2, 2-3, 3-1, 3-2 or 3-3.

But we can eliminate a lot of unwanted bets by making certain rational assumptions. The assumptions would be based on normally sound factors. Suppose the favourite will win ONE of the two races, then we would not need the combinations of 2-2, 2-3, 3-2 or 3-3 and we would be left with 1-2, 1-3, 1-1, 2-1 and 3-1.

It is obvious that you can save a lot of 'useless' staking cash by applying such an approach. But the approach, of course, has to be based on sound principles. The bets you cut out have to be correctly 'denounced', so to speak.

This feature is based on a series of booklets by 'Promath', published in Britain in the 1950s. A second article will be published in next month's December issue of E.R. PPM.

Click here to read Part 2.

Click here to read Part 3.

Click here to read Part 4.

**By Denton Jardine**

PRACTICAL PUNTING - NOVEMBER 1996