Many systems have a rule which states, “the horse must have won its last start”. This month we’ll examine this aspect of the winners of 660 races and see whether the rule is sound.

We’ll also include results for horses that ran second or third at their last start, and look at applying the different restrictions we have used previously.

Here are the results for those 660 races: last start winners won 142 races for a strike rate of 21.5 per cent or to put it into perspective, two out of ten races.

If we now include horses that ran second at their last start we find that between them, they won 262 times for a strike rate of 39.7 per cent or four out of 10 races.

When we include horses that ran third, the total becomes 330 or exactly 50 per cent. These results show that horses that filled any of the first three placings at their last start, win one out of two races.

If we only consider races where the favourite was paying between \$2 and \$3, we find there were only 262 of them. Last start winners were successful 56 times for a strike rate of 21.4 per cent or two out of 10 races. First or second won 114 times for a strike rate of 43.5 per cent or four out of 10, while all three placings won 146 or 55.7 per cent.

Not only have we been able to increase the strike rate from 50 to 55.7 per cent for horses that ran first, second or third last start, but also, by reducing the number of bets, we would have increased our profit.

There were 269 races with 10, 11 or 12 runners and these saw last start winners come home 52 times for a strike rate of 19.3 per cent. First or second won 102 or 37.9 per cent, while all three placings won 123 times or 45.7 per cent. It looks like we’re going backwards with that restriction.

Looking at only metropolitan tracks where there were 304 races, we find that last start winners won 76 times or 25 per cent of them. First or second won 134 or 44 per cent, all three placings won 163 or 53.6 per cent. With this restriction we’ve increased the last start winner’s strike rate to 25 per cent, or 1?4 of all races, this would be almost as many wins as favourites have on average but with the advantage of these horses paying better dividends than the favourites are likely to.

Non-metropolitan tracks had 356 races where last start winners were successful 66 times or 18.5 per cent, first or second 128 times or 35.9 per cent, while all three placings, 167 or 46.9 per cent. There’s nothing to get excited about there.

If we now turn our attention to races with seven or less runners we find there were just 59 where last start winners won 14 or 23.7 per cent. First or second won 24 or 40.7 per cent and all three placings won 37 or 62.7 per cent.

Summing up, the last start winner had an average strike rate of 21.5 per cent or two wins every 10 races. However, if we were to consider horses that ran into the first three places at their last start, then the strike rate is an average of 52 per cent or five wins every 10 races.

Using the above information to formulate bets is quite straightforward. We’ll start by just looking at all horses that finished first, second or third at their last start. In each race, jot down the TAB numbers of the horses that qualify.

About one minute before the race starts, write down the win dividend on offer for each of these horses. For example, these were the figures for the first race at Caulfield on Saturday 19th August: TAB number one was a last start winner and its dividend on offer was \$8.90. TAB five also a last start winner, was paying \$35.90. TAB numbers nine and 12 both ran third at their last start and their dividends were \$3.10 and \$5.40 respectively.

To reduce our outlay we will only bet on the three lowest paying horses 1, 9 and 12. Number nine was the eventual winner paying \$3.10. With one dollar on each horse, the profit would have only been 10 cents.

A different attack would be to put the least amount required on the two highest paying horses to show a profit and then work out the least amount to put on the lowest paying horse to make about 20 per cent if it should be the winner.

In the above example, doing it this way and outlaying an extra dollar on number 9, would have increased the profit by over \$2.00 to \$2.20.

By increasing the amount of the bet on the lowest paying horse, it’s possible that the total outlay will be more than would be covered if one of the higher paying horses had won. If this occurs, increase the bet on this higher paying horse then re-check the other amounts to see whether they need adjusting.

Let’s try the first race at Doomben that same day where only three horses qualified. TAB three and four had both won their last starts and were paying \$2.90 and \$6.50 respectively. TAB seven ran third last start and was paying \$15.00. Number three won and paid \$2.90. Two dollars worth of that would have shown a profit of \$1.80.

We’ll now look at the first race at Rosehill that day. Once again there were only three qualifiers who had all won their last start. Dividends on offer were \$6.70, \$14.30 and \$8.50. None of them won this time so we lost three dollars.

Moving to the first at Morphettville, we find three qualifiers. TAB 1 and 6 both won their last starts and were paying \$6.40 and \$7.00, while TAB 7 ran second at its last start and was paying \$9.50. Number one was the eventual winner of this race and with a dividend of \$6.40 we would have made \$3.40 profit.

This is starting to look too easy. How about the first race at Gold Coast then? No horses qualified here so there were no bets. Let’s move over to the west coast and the first at Belmont. TAB 1 had run 3rd last start, two had won and three had run 2nd. Dividends on offer were \$2.80, \$7.70 and \$3.00. Number two won at \$7.70 and we would have made a profit of \$2.70.

Good, another profit. Next, we’ll look at the first at Toowoomba where five horses qualified. Two of them were paying over \$16 so ignoring them, we look at numbers 1, 2 and 9. Their dividends were \$2.50, \$7.00 and \$7.40. Number 1 was the winner and if we’d put \$2 on it we would have made a profit of \$1.00. Strike! We can’t go wrong. Maybe I should be selling this system.

Let’s see if we have the same sort of luck with the Kiwi races. We’ll try the first race over there. Of four qualifiers, one was paying \$5.10 so it was left out. All the remainder had won their last starts. Numbers 3, 5 and 6 with dividends on offer of \$4.70, \$4.60 and \$5.00. Number 3 won and we would have made \$1.70 profit.

That’s looking good, now we will come back home and check the first at Kembla Grange where only two horses qualified. TAB 7 had run 3rd and number 8 had run 2nd at their last starts. Neither of them won this time so we lost \$2.

Now finally we’ll look in at Kilmore. Numbers two and three qualified but neither won so we drop another \$2. Summing up for the first race of the day at all tracks, we find we’d have made a profit of \$4.80. Maybe we could add a rule that states only bet if three or more horses qualify. In retrospect, had we done that, then the profit would have been \$9.80.

In some races, the dividends of qualifying horses can be rather low and trying to make a profit with a reasonable outlay becomes almost impossible. Being a cautious punter, I start to get cold feet if the total outlay for the race climbs above \$10. Therefore, we could make another rule that limits our outlay to less than \$10 or whatever figure you would be comfortable with.

In all the above examples I’ve used a minimum bet of \$1 but of course you could use whatever amount doesn‘t give you nervous twitches.

By Mr Money

PRACTICAL PUNTING – MAY 2007