Melbourne Cup: winning formula

Monday, 2 November, 2009

UNSW mathematicians have crunched the numbers on two decades of Melbourne Cup race results to provide insight into how professional gamblers consistently back the winners - and have devised a winning strategy at the track.

Each year the Melbourne Cup encourages us all to take a chance on the "race that stops a nation". However in 2009, with many punters feeling the pinch, risk-taking probably seems like a bad idea - until you look back through time.

"For many, horseracing seems an extremely risky way to multiply their money," says Professor Anthony Dooley, head of the School of Mathematics and Statistics at UNSW.

However, as many horse-racing experts already know, the mathematical principles behind predicting outcomes in racing are very similar to those used for devising successful stock market investment schemes.

For both gamblers and investors, a successful approach involves working out how to maximise financial reward while minimising financial risk.

According to Professor Dooley, the best strategy for financial success at the track relies less on trying to pick the winner and more on identifying the group of competitors that is most likely to include the winner.

To find out how this theory applies on the turf, UNSW mathematicians analysed Melbourne Cup race results since 1989. From this historical data they devised a winning strategy called the Melbourne Cup Trifecta 50, or MCT50.

The formula has two simple rules. First, eliminate all horses from the line-up that have odds longer than 50-1. Second, place $1 on all the possible Box Trifecta combinations for the remaining starters.

According to this study, had a punter used this strategy every year since 1989, by now they would have realised a profit of $40,885 - a return of 154%.

After an initial investment of $4896 in the first year, profits from the winnings would then have been sufficient to fund all subsequent bets.

In terms of conventional financial products, this outcome is equivalent to having invested that initial layout in 1989 in an account with a compound interest rate of 11.23% for twenty years, or at today's interest rates until about 2045.

 "Like any strategy, this formula is not fool-proof," cautions Professor Dooley. "However, historically it seems to be every bit as safe as investing in the share market, and it certainly provided a greater return than traditional bank account investments."

Punters considering using the MCT50 should be warned. In any given year, this formula will only have a one in three chance of actually returning a profit. Furthermore, to realise the profit calculated in the analysis, an individual would have had to stick it out for the whole 20 years since 1989 to reap the full rewards (or exit after fewer years for a less fruitful profit).

"Superannuation funds have already familiarised Australians with thinking about investing and saving for the long term - so a two-decade wait could well seem acceptable for many - and a lot more fun," Professor Dooley says.

Even so, Professor Dooley urges a strong note of caution for anyone considering this exact strategy on Tuesday.

"For the MCT50 approach to work in the future a scenario statistically similar to the past is required," he advises. "If everyone starts to use the same cut-off approach the odds worsen, and pay-outs fall dramatically."

"This kind of formula is a fairly straightforward calculation and it's an approach used by many of our leading racing experts in their own work," Professor Dooley says.

"Gamblers would tell you that if you're on to a good thing you need to keep your winning formula secret. However, part of our role as mathematicians is to understand and communicate the patterns that help explain and predict everyday activities - including the Australian love for a 'flutter'.

"If you really want to back a winner, invest in your own long-term learning so you can accumulate practical information that will help inform your own daily decisions - in both banking and betting."

Media contact: 
UNSW Faculty of Science - Bob Beale - 0411 705 435 -