Why do winners win? As we all know, there are a thousand answers to that one question! But by examining the 'why' of previous winners we get the chance to discover patterns. We can find that certain categories of horse win races.

A few weeks ago, my colleague Brian Blackwell did some instant research on a day's racing results in Sydney, Melbourne and Brisbane. He made the following 13 discoveries:

• 10 of the 22 flat winners at Moonee Valley, Rosehill and Doomben has their last start 14 or less days ago.
• 18 of the 22 winners had their last start 21 or less days ago.
• 11 of the 22 winners who were beaten at their last start had finished 4 lengths or less from the winner.
• 14 of the 22 had finished 6 lengths or less from the winner.
• 6 of the winners had won their last start.
• 6 of the winners had finished 2nd last start.
• 2 of the winners had finished 3rd last start.
• 1 of the winners had finished 4th last start.
• 3 of the winners had finished 5th last start.
• In total, then, 18 of the 22 winners had finished in the first 5 at their last start.
• 12 of the 22 winners started at 5/1 or less at their last start.
• 19 of the 22 winners started at 9/1 or less at their last start.
• 14 of the 22 winners started at a LONGER price than they did at their last start.

Of course, some 'control' needs to be exercised over these figures in that the winners were not the only runners to fit the criteria. But, certainly, the stats do set the brain thinking, don't they?

Can they be used as a starting point for us when we study form? Knowing that a very high percentage of winners (18 of 22 in this case) ran in the first 5 last start, can we now confine our form study to those horses with the form figure '5' or lower at their last start?

We will miss some winners - but we will still be looking at a group likely to land 8 or 9 out of 10 wins!

If we examine things further, we see that 45.45 per cent of the winners had their last start 14 days or less ago. However, on the day a total of 147 of the 239 runners at the three meetings fitted the same factor, which is 61.5 per cent of the total runners.

A total of 190 of the 239 runners had had their last start 21 days or less ago, which is 79.49 per cent and at Doomben a total of 81 of the 83 runners fell into this category!

Winners last start? Well, there were 6 at the three meetings, which is a 27.27 per cent strike rate. But closer examination shows that 51 of the total 239 runners were last start winners.

These examples show that the task of finding winners in various categories will never be easy - but there is still room for the astute punter to make the most of the stats.

I believe the significant statistic is that one which says 18 of the 22 winners finished in the first 5 at their last start. I am convinced this statistic can be used to help us enormously in finding winners.

Of course, that is only the start of the story; more detailed long-term study needs to be undertaken, though I have the suspicion it will come close to verifying this one day's action.

But let's look at a recent race to see how we could have examined it with this theory in mind.

It's the Grange Hcp 1000m at Sandown on August 24. There were only three runners fitting the category, ignoring one which was resuming after a spell.

These were Planet Warrior, Damascus Road and Tristolier. Tristolier though, hadn't started since June 8 and knowing what we know about horses resuming from long breaks, we could put a question mark over him.

Planet Warrior had run 14 days ago and run 3rd at Sandown, beaten less than 4 lengths. Here was a runner who had some appeal. He looked a pretty sound 'factors' bet in regard to days since last run, beaten margin last start and last start placing within the first 5.

He could be earmarked as a definite contender. The other runner, Damascus Road, had run 4th last start, beaten 2.6 lengths. Those two factors are fine, but we see he is starting for the first time in 6 weeks. This is a worry.

But his credentials are certainly enough for us to pencil him in as a chance, not as strong a chance as Planet Warrior, but strong enough to have something on him. It's interesting, indeed, to see that Damascus Road won at 12/1 with Planet Warrior running 2nd at 4/1. That's an approximate 25/1 quinella.

Okay, that's only one race but I think it's a classic example of how simple statistics can enable you to zero in on value winners. Some other races that day defied the statistics, pointing up the inherent trap in any sort of racing statistics - the horses don't read!

One crystal clear point emerges from all this: Whenever we think about why winners win, we should take the time to analyse all the winners from a day's racing to find the 'positive' element in their form which could have alerted us to their winning chance.

The following are my own comments on winners at Randwick and Sandown on August 24:

RANDWICK

Race 1: Miss Solitaire: Handy form last time in; obviously prepared for a first-up win seeing as she was put in a 1400m sprint and not something shorter.

Race 2: Crows Before Dawn: Unsuited over 1400m at his last start but flew home over 1500m the start before that. Got in with a half kilo over the minimum. Top jockey riding.

Race 3: Wavertree: Had solid form last time in, and this included a 2nd to Gold Ace at Flemington last November. Had big reports about him going into the race.

Race 4: Port Watch: Hard to find any compelling reason for his win, though a win three starts back in a Canterbury Welter was probably the main clue.

Race 5: Aurum Heights: Had run two good races following a spell and came in with a 3rd, beaten 1 length, and a 19 day break. Top rider aboard.

Race 6: Paint: Had much to recommend, including Blue Diamond win and Golden Slipper 2nd. Was expected to race well fresh.

Race 7: Macrosa: Had run a half length 2nd last start, 21 days before. Comes from a top stable.

Race 8: Stand And Cheer: Top stable, but form had been ordinary. Was having 6th run from a spell. No real clue!

SANDOWN

Race 1: Asian Incline: Classy WA type who had raced well at the track before. Came in after 0.8 lengths 2nd 14 days earlier, so was a pattern galloper all right.

Race 2: Syncopated: Last start winner but having first run for 8 weeks. Top stable and had won fresh before.

Race 3: Ascorbic: Resuming from a spell; last start had been 1.3 lengths 2nd. Had no previous first-up winning form, though lightly raced.

Race 4: Damascus Road: Was 4th, 2.6 lengths, last start and got in with a light weight. Had winning city form three starts ago.

Race 5: Street Talk: Had run 8th last start but in better company and obviously appreciated easier field. Had solid win and place form at the track.

Race 6: Delsole: Could have been chosen had you taken a line through Moondah Stakes win three starts back. But you had to ignore last 2 failures, though there apparently was an excuse for his last start 7th at Moonee Valley.

Race 7: Abra Magic: Came in after a win 14 days earlier and had a good winning record at the track.

Race 8: Our Brigetta: Was a last start winner 21 days earlier and had good win and place form at the track.

You can see from these comments that each winner had something positive about its form - some had a lot going for them, with others you had to really search for the clue.

And 'search' is the operative word when it comes to tracking down winners. The easier you can make it for yourself, the better. This is often where systematic selection comes into play for many punters.

Friends of mine swear by mechanical systems because they are men who haven't got the time to sit down and sweat over formguides. They demand a quick path to selections. The use of selections gained by a system is the way they choose.

A couple are subscribers to Equestrian services, another has his own system, which involves very little work. It's a plan that simply looks at betting and weights, and picks out a selection after some simple calculations.

Says this friend: "I've often asked myself why winners win, and I have found there are so many answers that I feel the task of unravelling the mysteries is too much for anyone. My system gives me peace of mind. I don't have to worry about results, about how a horse performed last start, whether his jockey fouled up and all that stuff.

"I just tick off my system rules for each runner and within a few minutes I have my picks for the day. I think I'm doing better now than I ever have done before."

But for those who abhor or decry systematic selection processes, the task looms to decipher the past in order to make sense of the future. Looking at winners, and why they won, should really be considered a vital part of the approach to wider and more penetrating form study.

By discovering what made a lot of horses win a lot of races you can make a breakthrough for yourself. Your studies may point up significant areas that demand further consideration.

I know many punters are keen on dissecting first-up form. My colleague Brian Blackwell is an ardent fan of this approach, often snaring seemingly 'impossible to find' winners by looking deeply into previous first-up form.

To my way of thinking, though, the essentials to be remembered always are these:

• STRONG RECENT FORM
• UNPLACED LAST START BUT NOT TOO FAR AWAY FROM THE WINNER
• WINNER OR NOT BEATEN MUCH LAST START
• GOOD STABLE AND GOOD RIDER.

I am sure that when you examine winners and their form these factors will crop up more than any others. Sure, a lot of runners will meet the criteria, but it's still a giant step in the right direction.

Examine these areas of a horse's form and you will come across many winning selections. It's all a matter of dealing positively with the POSITIVES of form. Sift through carefully, weigh up class and weight, allow for problems in the run, and then MAKE YOUR CHOICE!

By Philip Roy

PRACTICAL PUNTING - OCTOBER 1996