The Billion Dollar Bookie Beating Algorithm
- Calum McGrath
- Mar 9, 2020
- 4 min read
Updated: Mar 30, 2020
“In gambling the many must lose in order that the few may win.”
George Bernard Shaw
Hong Kong has a long-standing affiliation with gambling. With horse racing being by far the most popular outlet for punters. Run by the Hong Kong jockey club founded in 1884 as a refuge for upper-class Brits who wanted the feeling of England in their new colony. The club changed over time into a state gambling monopoly. It's two courses Happy Valley and Sha Tin are packed twice a week during the racing season.
Happy Valley, Hong Kong.
But the pinnacle of the Hong Kong betting scene is a bet which requires gamblers to pick the top 3 horses, in any order, for 3 consecutive races. The bet is known in Hong Kong as the Triple Trio, it contains over 10 million possible outcomes. The bet is treated as a lottery in Hong Kong, where the winners are presented with large checks on T.V.
In 2001 there was a huge jackpot that could be won. Roughly 100 million dollars H.K. One million people placed a wager on the Triple Trio for that race. That's equivalent to about one in seven of the population of Hong Kong at the time (Kit Chellel, 2018). When Bobo Duck romped home at Sha Tin the large jackpot had been won. 100 million dollars. But the jackpot was never claimed.
Who would leave a jackpot that large on the table?
To answer that question, we'll have to go to Las Vegas; home to many successful gamblers and also home to even more unsuccessful gamblers. One of those gamblers was a 23-year-old college dropout. Working for $3 an hour and making around 40 dollars on the tables a night he wasn't quite as ineffective as your average gambler. His name; Bill Bentor, and his secret to consistent profits at the table; counting cards.
Inspired by Edward Thorp's "Beat the dealer" Bentor packed up and moved to Vegas. In 1980 after months of lacklustre profits from Thorpe's techniques, Bentor was introduced to the man that would change his life, Alan Woods. Woods was the leader of a card-counting team in Las Vegas. He was once an insurance actuary with a wife and two children but left it all behind to travel the world as a gambler. Wood's team would pool their money and split their winnings thus reducing the risk of a spell of bad luck emptying a gamblers bank account.
After 4 years of high rolling with Woods, their luck rang out. Casinos around the world became wise to card counters. Woods, Bentor and the rest of their crew became banned from casinos around the world. They needed a new game.
Woods was interested in horse racing, and between them, they decided to target Asia and in particular the Hong Kong Jockey Club. They were aware there was this huge horse racing market they might be able to tap into. One of the reasons that they chose Hong Kong is because it's almost like a laboratory for studying horse racing. In Hong Kong, there are two racecourses that dominate. There's a huge amount of betting activity. There's a pool of about a thousand horses and hundreds of races a year. Making ideal conditions for mathematical and statistical methods.
Woods went to Hong Kong, and he came back with all the results of racing in Hong Kong going back ten years. They manually fed the results into a very basic early computer to try and create a statistical model. It took them about nine months to manually input all of the data. Their plan was to use these measures of horses' quality and come up with a prediction of each horse's, in a given race, their chances of winning. Then they would compare that to what the odds on the board implied those horses of chance of winning and what they wanted is a horse where the odds imply it has a very low chance, but their model says it has a pretty high chance.
In his first year of betting, they lost $150,000. During the slump, Woods and Bentor fell out over ownership, and both men went their separate ways. Bentor continued to work on his model. He discovered that in the Hong Kong Jockey Club, the odds are determined by the action. So the number of people betting on each particular horse, and the sums they bet, determine the odds. Eventually, he came to incorporate the public odds into his model, and that was a huge breakthrough (Kit Chellel, 2018).
Bentor's model began to become successful. He hired some employees to start calling in bets for him. He bought some new computers and even rented out an office space. Bentor's gambling operating started resembling a modern business. Then one he gets a call for the jockey club. Preparing to be told he was to be banned from the betting ring, the call instead took an unexpected turn. The Hong Kong jockey club make a commission on every winning bet, so instead of banning Bentor, they wanted him placing even more bets. They offered him a state of the art computer called Customer Input Terminal. The computer allowed him to places bets in a matter of a second. So it allowed him to make even more bets. More bets meant more profits for Bentor.
The amount of money that was coming out of the pool and going to Bentor and other computer teams was massive. Bill algorithm got too good. The jockey club became scared that if the Hong Kong betting public became aware that so much money from their betting pool was going to foreign-run computer teams that they may be discouraged from betting. As a precaution, they cut off Bentor.
Bentor, who had already made just under one billion dollars, still had a point to prove. He wanted to prove that out of all the computer algorithm his was the best, and that he had truly bet the bookies. Bentor wanted the triple trio. He assembled a team, and they began manually putting on bets at betting machines around Hong Kong. Bentor and the team put on over 50,000 bets and gambled over 1 million Hong Kong dollars. Bentor won. But he never claimed the prize, Bentor was satisfied knowing that his algorithm had beat the bookies. The 100 million dollars went to charity.
References:
Kit Chellel, 2018 https://www.bloomberg.com/news/features/2018-05-03/the-gambler-who-cracked-the-horse-racing-code
Another great article from cmgcrat, keep up the good work!