Statistics can play a vital role in the way we assess races. But how can you use statistics to help your betting and your selection process? In this article, Martin Dowling examines the work of Dennis Walker, of the Rating Bureau.

Dennis devotes thousands of hours every year to investigating and examining racing stats and to developing revolutionary new computer approaches to racing.


Dennis Walker prides himself on the fact that he has studied just about every angle anyone can think up on racing. Through the use of computer analysis, which enables him to probe back years and years, he can swiftly uphold or demolish most selection theories!

Take, for example, the always-hot issue of first-starters and horses resuming from a spell. This is an area Dennis has looked at thoroughly.

He says: "I constantly read that we should never back a horse at its first start in a race against tried horses, and that we should never back a horse first or second-up from a spell, and that we should always wait until we know it is fit at its third start back in a sprint, or fourth run back in a distance race.

"I asked the computer to give me the facts, and instead of taking a normal approach and looking at each race, I looked at each horse and found some interesting statistics. My report covered all racing at metropolitan tracks in Melbourne and Sydney for exactly two years (some 3700 races) and I defined a spell as a break exceeding 42 days.

"Now, of 3683 races a total of 9.5 per cent were won by first-up horses. Another 528 races (14.3 per cent) were won by horses second-up from a spell. To put all this into perspective, of course, we need to know how many first-up and second-up horses contested these races.

"The total 3683 races were contested by 45,211 horses and of these some 5575 (12.3 per cent) were first-uppers, and 2260 (5 per cent) were having their first start, while another 7574 horses were second-up."

Dennis says punters need to look at detailed breakdowns of the statistics by age, distance, track condition, prizemoney, odds-range and age, to discover further clues to potential profitability.

"One thing I must point out is that no matter whether a horse is having its first race start, or is first-up from a spell, or seventh back from a spell, it MUST be fit to win a race! A lowgrade horse is capable at peak fitness of beating a good horse that is unfit.

"Racing is so competitive that no horse can win a race unless it is fit. Punters who venture to the track have an advantage over stay-at-home punters in that they can look at a horse's condition, and they can make sure it is not sweating up badly on a cool day.

"We must remember that many horses are entered for races in Australia when they are not fit. In fact, many trainers use actual races to bring their horses to peak fitness. Some of the more astute trainers know when their charges are close to fitness; they even know the weight for a horse when at its peak condition.

"This is a factor we, as punters, do not know. In Japan, for example, a horse is not allowed to race unless it is close to a given racing weight.

"I have read that more than 80 per cent of races are won by horses who had their most recent start within 28 days, and some 60 per cent of races are won by horses which had raced within the previous 21 days.

"Contrary to the thoughts of many losing punters, racehorses are not mechanical machines; they are highly-strung animals weighing some 550kg and maintaining peak fitness is not always possible. Fitness is, more often than not, either improving or declining at any given time."

Dennis has some interesting figures on the comparison between betting at the track and off-course - these are even more relevant in the light of the recent suggestion in NSW that bookies should be barred!

"There really is no comparison to being there at the track," says Dennis. "This is where you have the best of both worlds at your disposal. In fact, if you're keen and are prepared to work hard you can often sense a 'best price' situation and do even better.

"Let's say you intend betting $100 on your selection, which has opened at $3.90 on the tote and 3/1 (best) with the bookies, and it blows to 9/2 after 25 minutes, and you see a few large bets being placed.

"The 9/2 starts shrinking to 4/1 around the ring. After a quick check, you note the tote price is now $4.75 with only five minutes to the jump. You should take the plunge and grab the 9/2 as quick as you can. After all, unless there is very heavy tote support for other runners, you are unlikely to achieve the equivalent of $5.50 on the tote.

"If you wait until the final moment, the SP could be back where it started at 3/1."

Dennis says the stay at home punter has no choice  but to accept whatever the fickle finder of fate hands out on the TAB. But, he points out, punters using the Teletext service, with updated tote divvies, can monitor different totes and place a bet on the one offering the best return.

"Something punters may not be aware of is that the TAB price will always add to more in the long term than SP because the average takeout on win bets is less than 15 per cent, while the bookies' takeout is, in reality, above 20 per cent on average.

"On most country meetings the bookmakers' takeout could be as high as 40 per cent if he is both smart and able to lay every runner evenly. But the main point I want to demonstrate is the huge advantage you can enjoy when in a position to choose between SP and TAB by being there at the track.

"A quick look at a recent day's betting figures showed that a $1 bet on the winner of every race would have netted TAB punters a total collect of $276 compared to $297.55 on SP. But a $1 win bet on every winner at the BEST FINAL ODDS results in a collect of $341.90, some 15 per cent more than the best available to SP punters or TAB punters.

"You can realistically expect the margin in the long term to be around 10 per cent. Putting it another way, you will add 10 per cent more to your bottom line. How much more than 10 per cent depends on your interpretation of the movements in the market on course.

"It's hard work but this is one of the secrets known to professional tote players who carve out a living."

Using The Rating Bureau's Trakline and Formline Gold data base, Dennis is able to produce any set of statistics to help in the development of fresh lines of thinking. Many professional punters are using such data bases to develop systems and check out staking plans.

As far as Trakline and Formline Gold are concerned, Dennis's checks in 1994 showed the following: Top-rated horses won more often from inside barriers.

  • Those that drew wide barriers paid much better dividends.
  • Top raters drawn barrier 17 or wider had the best strike rate and paid well.
  • The younger the top-rated horse the better chance it had of winning.
  • Top-rated horses older than 5 had a very poor record and should have been eliminated.
  • Top-rated male horses were much more reliable than female horses.

You can check out these findings against your own selections. How do your top picks perform from inside barriers? Are you wasting your money on female runners? Or wasting it on older horses?

Can past results be a guide to what we can expect in the future? Yes, they can, provided the statistics are taken over a lengthy period, and not just a week or two.

  • Dennis Walker's statistics, published in his regular newsletter Pot’O’Gold, cover a wide range of important racing factors, like winning margins, weight carried over the limit, winning weights etc.
  • Did you know that more than 50 per cent of winners are likely to win by a length or less? That more than 75 per cent of winners score by 2 lengths or less? That more than 90 per cent win by 3.5 lengths or less?
  • In a test of 22,989 races, it was revealed that 19.6 per cent of winners won carrying the Limit weight.
  • Just over 90 per cent of the winners carried between the Limit and 5.5kg above the Limit.
  • Only 6.3 per cent of winners carried under the Limit weight, due to apprentice claims.
  • Almost 56 per cent of winners carried the Limit or up to 2kg above the Limit.
  • In a survey of some 11,743 horses in 1062 races it was found that 12.43 per cent of races were won by horses which last raced 7 days ago.
  • Almost 56 per cent were won by horses that last raced between 7 and 14 days ago.
  • Just over 80 per cent were won by horses which last raced between 7 and 24 days ago.
  • Less than 5 per cent of races were won by horses which hadn't raced for 90 or more days.

By Martin Dowling

PRACTICAL PUNTING - OCTOBER 1995