Beating the Lottery Odds - Part Four

When a retired couple were making millions from a math error in a lottery game, they suddenly found they had stiff competition.

Four-times winner of the Texas lottery, Joan Ginther. Image source

In Part 3, retired couple Jerry and Marge Selbee were working hard on exploiting a math error they found in a Massachusetts lottery. But they discovered someone else was manipulating the game to force them out.

Odds are statistics

Finding loopholes or statistical errors made by lotteries was not anything new. Joan Ginther won the Texas Lottery at least four times in 10 years. It is estimated she had winnings of $20 million at a cost of $3.3 million in tickets. A clue to this is that she has a Stanford PhD in statistics and lives in Nevada, making repeated trips to a small store in rural Texas. Sound familiar? No one has yet figured out how she did it and she's not telling and has gone to ground.

Mohan Srivastava. Image source:

In 1983, an MIT-trained statistician named Mohan Srivastava worked out how to correctly predict the numbers in scratch-off tickets in Canada—with a 90% success rate. But the rewards were not high enough to justify the amount of work that it would take to buy the tickets, sort the winners and redeem the prizes.

Of course, Jerry and Marge Selbee had been both open about what they were doing and were certainly willing to put in the hard work.

Serious competition

In late 2004, an MIT student by the name of James Harvey had found the flaw in the Cash Winfall game while researching lottery games. He convinced 50 students to put up $20 each to buy 500 Cash Winfall tickets for the February 7 roll-down drawing. They won $3,000, making a $2,000 profit.

Random Hall, MIT. Image source:

On the strength of that, Harvey put together a betting pool of 40 to 50 regular players, including professors. He recruited his classmate, Yuran Lu, to manage the group and to later help Harvey form the company Random Strategies LLC, named after their dorm, Random Hall. They would wager $600,000 each roll-down week. They opted to choose their own numbers, mainly to avoid duplicates. The lottery did not allow punters to print their own lottery slips so they had to spend weeks filling them in by hand.

At much the same time, a biomedical researcher at Boston University also discovered the flaw. Unlike the others, Ying Zhang was against the lottery, believing it to be exploitative—and researched the Massachusetts State Lottery to prove his point. But once he found that there was basically free money to be made, his greed took over. He too formed a betting club, Doctor Zhang Lottery Club Limited Partnership. They wagered between $300,000 and $500,000 on roll-down weeks. Zhang soon quit his job to concentrate on the lottery game fulltime.

It is not clear how legally the Zhang group operated. Stores mainly around where Zhang lived were found to have violated two rules: players had been scanning stacks of computerized betting slips, and a store where he operated had been extending him credit. The slips had been scanned before they’d been paid for.

Although well within its rights, the Lottery did not revoke the licenses of these stores, nor was any other action taken.

As the demand for tickets rose dramatically in the Cambridge region, store owners were sometimes frantically contacting the lottery to check if these large purchases were legal (they were), and to increase their stores' betting limits.

Image source

The freeze-out manoeuvre

With (at least) three betting groups exploiting the Cash Winfall math error, the payouts were being split, making it less profitable than it could have been. This got the MIT students thinking that if they could force an unscheduled roll-down by placing a huge bet, they could freeze-out the other players and have the pot to themselves. The lottery only announced a roll-down when the jackpot was very close to $2 million. By placing a bet that would push the jackpot over $2 million, the other players would be caught unawares.

On August 16, 2010, the opportunity arose to put their plan into action. Over three and a half days, they spent $1.4 million on 700,000 lottery tickets. That would easily push the jackpot from $1.6 million to over $2 million. The lottery—as expected—had no time to announce the roll-down, and the drawing took place without the participation of the Selbee and Zhang groups. The MIT group cleared $700,000 that day.

The lottery people, of course, noticed the strange pattern of play. Examining the data they correctly concluded that a large betting group had triggered the roll-down, but misidentified the culprits; they thought it was Jerry's group. But, once again, no action was taken except for writing a script to alert them of super-high ticket sales in order to announce a roll-down beforehand.

Jerry was not amused, “They took us out of the game intentionally.” It was time to play them at their own game.

The Christmas fightback

Jerry guessed that the most vulnerable drawing would be the one on December 27 when many convenience stores would be closed for the holiday and ticket sales would be slow. To take some of the guesswork out of it he asked his friend Paul Mardas to call lottery headquarters to find out if any stores were reporting high sales volumes.

When he heard back from Maradas that indeed, 5 stores were showing surges in sales, he jumped into his car and headed for Jerry's Place—alone. It was Christmas Day. On arrival, he immediately started printing 45,000 tickets, working way into the night, alone in the store.

While in the store, there was a knock on the door. The young man standing there introduced himself as Yuran Lu, coordinator of the MIT group. Jerry claims that Lu proposed that the two teams collude to take turns at the winning the pot. Jerry found the proposal unethical and shut the door in the young man's face.

As per their usual form, the lottery was again slow to react to the large bets forcing the roll-down. He was pleased that he had taught the MIT bunch a lesson, particularly as he'd made about $200,000 in profit.

Everything returned to normal for the next six months. But there was trouble ahead: the lax attitude of the lottery official was about to catch up with them.

---

This will be concluded in Part Five

Previous posts in this series: Part 1, Part 2, Part 3

References:
Inc.: Here's the Story of the Stanford PhD Who Allegedly Gamed the Texas Lottery (and Won $20 Million)
ABC News: Who Is The Lucky Four-Time Lottery Winner?
Huffington Post: Jerry and Marge Go Large
Don't Waste Your Money: This Couple Won $27 Million Using Math To Hack The Lottery

Also posted on Weku, @tim-beck, 2019-01-07