This month we will continue our examination of the relationship between starting price (SP) and 'true' odds.

By examining nearly 200,000 starters during the two year period August 1, 1988 to July 31, 1990, I concluded last month that horses with SPs of even money won only 43% of the time. In other words a price of 11/8 was required to break even. We also found that horses starting at 6/4 won 35% of the time, meaning a price of 15/8 was required to break even.

Horses starting at 2/1 won 28% of the time meaning that you would need a price of 5/2 to break even and horses starting at 5/2 won 25% of the time meaning that you would need a price of 3/1 to break even.

For SPs at 3/1 and greater the disadvantage to the punter just got worse and worse. Accordingly, I concluded that punters should concentrate their betting on horses at odds of less than 3/1 since they only need to get a bit over a quarter of a point extra about these horses to break even.

This month we are going to look at whether the gap between the 'true' price and price available can be reduced even further. We will attempt to do this by first finding out whether the gap varies with total market percentage and second whether it varies from one track to the next.

What is total market percentage I hear you ask? Basically it is the figure that gives the bookmaker an edge over the punter. If you convert all the bookmaker's prices to percentage probabilities and add them all together you will usually get a figure in excess of 100 per cent. The amount of excess is the bookies theoretical advantage. For example if the market is framed to 115% then the bookie has a theoretical advantage of 15% (for
every \$115 accepted in bets, only \$100 is paid back to the punters).

Bookmaker markets are usually framed to between 110% and 130%. The actual figure is usually higher early in the betting and lower at jump time. Similarly it is generally higher for country races than for city races. The common factor is how confident the bookie is about (a) the market accurately reflecting a 'true' market and (b) the likelihood of securing enough bets to make a balanced book.

With a fairly accurate market the bookie can get away with a smaller theoretical margin whereas with a much more risky race a bigger margin is needed to cushion the potentially larger risk.

Examining even money horses I found that whereas they won 43% of the time overall, their strike rate improved to 47% in races with a market framed to between 100% and 110%. In races where the market was framed between 110% and 120%, their strike rate was 45% and in races where the market was framed to between 120% and 130% their strike rate was 42%.

The strike rate of even money horses dropped off quickly to 39% in races with a market percentage between 130% and 150% while for races framed to more than 150% the strike rate was a very poor 32%.

For horses starting at 6/4, my computer told me that although they won 35% of the time overall, they won 39% of the time in races with a market framed between 100% and 110%. In races where the market was framed between 110% and 120%, their strike rate was 37% and in races where the market was framed to between 120% and 130% their strike rate was 34%. As with the even money horses, strike rates fell off more dramatically as the market percentage increased beyond 130%.

It is now apparent that by confining ones betting to races in which the market percentage is less than 120%, the gap between the true price and actual price becomes smaller. For example, for horses starting at evens and 6/4 we only need to average 5/4 and 13/8 respectively to make a profit.

Coupled with the methods described last month for getting better than SP in the betting ring, this extra piece of information further strengthens the oncourse punter's position.

To benefit from this new information, the punter needs to be able to determine the total market percentage very quickly while standing in the betting ring. For the majority of you who probably aren't happy trying to do this in your head, I suggest a small pocket computer or programmable calculator. The program to use is quite simple and you can write to me at the address given below for details.

Before concluding this research I looked at the winning strike rate of even money horses at metropolitan tracks. They won 46% of the time in Sydney, 42% in Melbourne and 40% in Brisbane.

Further to this, I found that horses starting at 6/4 won 38% of the time in Sydney, 35% in Melbourne and 33% in Brisbane.

The conclusion to be made here is that the strike rates for evens and 6/4 horses is better than the national average in Sydney, around average in Melbourne and below average in Brisbane.

The reason for this can be readily related to market percentages in the various cities. Over a recent one year period I found that Sydney markets were framed to 116% at jump time, Melbourne markets to 122% and Brisbane markets to 128%.

Beaten margins was a topic we were going to start discussing this month so here's a few statistics based on metropolitan and provincial racing for the period August 1, 1989 to July 31, 1990.

The average beaten margin for horses finishing second was 1.5 lengths. For horses finishing in third position it was 2.9 lengths and for horses finishing in fourth position it was 4.7 lengths.

More on this next month.

YOUR LETTERS
Mr E N. of Burpengary, Queensland, has been making a 105% profit on turnover by using a modified version of a system presented by Richard Hartley Jnr. back in the April 1987 P.P.M. Space limitations prevent publication of full details here although the main rule of the system requires consideration of horses that came first, second or third at both of their last two starts.

The profit achieved by Mr F. N. is testimony to the fact that good recent runs are essential criteria for picking up and coming winners. A very interesting letter from Mr J. B., of Waikerie, S.A., further requires that horses must have finished within half a length at their last start (he claims that these horses are just as good propositions as last start winners). Both these gentlemen are on the right track as we will see from our beaten margins statistics next month.

Mr G. B. of Airds, N.S.W., wrote to me about two form factors he thought were important. The first, days since last start, has already been confirmed as an important form factor earlier in this series. The second, weight allocated by the handicapper, is a factor we will look at in a couple of month's time.

Click here to read Part 6.
Click here to read Part 1.
Click here to read Part 2.
Click here to read Part 3.
Click here to read Part 4.

By Neale Yardley

PRACTICAL PUNTING - JULY 1991