In recent years, there seems to have been a widespread move against backing more than one horse in a race. I suspect it's all to do with the so-called 'pure mathematics' that suggests bettors who go for more than one horse in a race will never end up winning.

How come, then, that I know several punters who have made what are known as 'Dutch Books' throughout their betting lives and have made them pay off? I guess it's a matter of how serious you are about the way you compile your selections. Any staking approach that is pinned by poor selections is always going to fail.

Dutch Book betting gives the careful punter a good chance of keeping ahead of the game, but it's a staking approach that has to be used sparingly. The user needs to pick the right races.

The whole thrust of the idea is that you are really taking some insurance on your bet. Instead of relying on one horse in a race, you rely on three or even four. The profit margin depends on the prices of your selections.

I know one chap who concentrates on major Group I and Group 2 handicaps for his Dutch Book betting. These are mostly wide betting races and he often secures remarkably good odds about his selections.

What, then, are you really doing in following the Dutch Book staking method? You are making a 'book' against the bookmaker. Instead of tackling him with one runner, you embrace three or four and look for as good a profit as you can get, given the prices available.

Let's have a look at an example: You have narrowed your choices down to four horses. They are at 5/1, 7/1, 10/1 and 14/1. The required number of units to bet on each is determined by adding ONE to each of the odds and dividing them into 100.

The bets, then, would be:

HORSE A: 5/1-17 units
HORSE B 7/1-13 units
HORSE C 10/1-9 units
HORSE D 14/1-7 units

This is a total of 46 units. If any of this four happens to win the race, you will make a profit of 54 units.

Of course, the ACTUAL odds are 1.17 to 1 (46 out, 100 back, profit 54). This gives you a clear idea of what you need to do via your strike rate in order to make a long-term profit.

I know many of you will say that you cannot afford to bet in such high figures. Okay. No worries on that aspect. You can use a simple bit of maths to work out your bets to whatever the size your stake is going to be.

Let's say you have narrowed the chances down to three. They are at 4/1, 4/1 and 10/1. Betting to the odds, as outlined above, the bets to take out 100 units would be 20, 20 and 9, a total of 49 units.

But you have only 20 units to bet. Your actual stakes, then, are determined by that percentage figure of 49. For Horse A, with a percentage of 20, the calculation is: 20 divided by 49 equalling 0.4081.

This is the proportion of the stake to be placed on Horse A.

The actual bet is calculated like this:

Horse A = 20 x 0.4081, equals 8.16, which is rounded off to a bet of 8 units.
Horse B is also at 4 / 1, so the bet on it is 8 units as well. It stands to reason, then, that the remainder of the money, 4 units, goes on the 10 / 1 chance.

What if you had say, 12 units to bet. The process is the same. Divide Horse As percentage of 20 by 49 and then go: 12 x 0.4081 which equals 4.89, which is a bet of 5 units rounded off. The same bet is for the other 4/1  chance, while the remaining 2 units goes on the 10/1 chance.

What you must remember is that it's impossible to 'Dutch' if prices are too short. The tighter the prices, the lower the profits, until you reach a situation where a profit cannot be made.

For example, if one of your fancies is odds-on, then it's almost impossible to make a Dutch book that will return you a profit, unless your other choices are real longshots.

Most professionals will work to a reasonable profit medium. They say never bet more than 75 or 80 units. For a total bet of 80 units, the profit will be 20 units, but this translates to odds of 1/4 ON if a winner is backed.

This same Dutch process can be used in different races. Imagine that you have the following selections:

Race 1 at 3/1
Race 4 at 7/1
Race 7 at 10/1
Race 8 at 9/1

The percentages for each are 25, 12.5, 9 and 10, a total of 56.5.

You can bet them according to the percentages and stand to make 43.5 units if only one of them wins. If you get two winners, the profit figure rises, and so on.

Let's say the 3/1 chance and the 10/1 chance both win. You get 100 units back on the 3/1 chance and 99 back on the 10/1 chance.

So your bet of 56.5 units has returned 199 units, a profit of 142.5 units, which equates to 252 per cent on your outlay. That is cooking with gas!

Another way of betting a few selections on a day is to follow the principle that has been referred to as the Triple Dynamite Plan.

On sound selections (not silly ones!) you'll need a bank of about 200 units, more if you can afford it. The ideal approach is to select three best races for the day.

Two horses are played on a flat stake to win in each race. The amount returned, if either horse wins, is split in half and played back on two more horses in the next chosen race.

Conservative bettors will aim for three successive winning plays. Once you've hit three in a row, you should (wisely) call it a day and start planning for the next day's betting.

Example: Let's say three winners were hit at 4/1 and you were betting in \$2 units. The return, including stake, from the first winner would be \$10. This is split into bets of \$5 on each of the next two selections.

The return for a winner would be \$25. This is halved for the third race action into two bets of \$12.50 each. The return on a winner would be \$62.50 for a profit overall of \$58.50.

This is a risky strategy in many ways in that you are bidding to score in three successive races. But, you have two chances in each race, making it a feasible operation.

Table of Prices and Exact Percentages

Price
Per Cent
Price
Per Cent
Price
Per Cent
1/1090.919/1052.638/111.11
1/990.001/150.008/111.11
1/888.8910/947.379/110,00
2/1588.245/444.4410/19.09
1/787.5011/842.1111/18.33
2/1386.576/440.0014/16.67
1/685.7113/838.1015/16.25
2/1184.627/436.3616/15.88
1/583.3315/834.7820/14.76
2/981.822/133.3325/13.85
1/4 80.009/430.7730/13.23
4/1578.955/228.5733/12.94
2/777.7811/426.6740/12.44
4/1376.473/125.0050/11.96
1/375.0013/423.5360/11.64
4/1173.337/222.2266/11.49
2/571.4315/421.0580/11.23
4/969.234/120.00100/10.99
1/266.679/218.18200/10.50
8/1565.225/116.67250/10.40
4/763.6411/215.38330/10.30
8/1361.906/114.29330/10.30
4/660.0013/213.33500/10.20
8/1157.897/112.50660/10.15
4/555.5615/211.761000/10.10

by Mark Merrick

PRACTICAL PUNTING - JULY 2000