Cheating Isn’t Always the Explanation

I spend much of my time trying to find cheating where other people don’t suspect it. So when I heard about a strange happening in Powerball — 110 people picking five out of six correctly when statistically you would expect only 4-5 such winners — I began to fantasize about a big cheating scandal. It had the appearance of the scam in the Breeders Cup horse racing a few years back when an insider was able to craft fake betting slips after some of the races had run.

The answer to the puzzle, though, lies not in the human tendency to cheat, but rather, in the amazing willingness to follow orders. From the New York Times:

May 11, 2005
Who Needs Giacomo? Bet on the Fortune Cookie
Powerball lottery officials suspected fraud: how could 110 players in the March 30 drawing get five of the six numbers right? That made them all second-prize winners, and considering the number of tickets sold in the 29 states where the game is played, there should have been only four or five.

But from state after state they kept coming in, the one-in-three-million combination of 22, 28, 32, 33, 39.

It took some time before they had their answer: the players got their numbers inside fortune cookies, and all the cookies came from the same factory in Long Island City, Queens.

Chuck Strutt, executive director of the Multi-State Lottery Association, which runs Powerball, said on Monday that the panic began at 11:30 p.m. March 30 when he got a call from a worried staff member.

The second-place winners were due $100,000 to $500,000 each, depending on how much they had bet, so paying all 110 meant almost $19 million in unexpected payouts, Mr. Strutt said. (The lottery keeps a $25 million reserve for odd situations.)

Of course, it could have been worse. The 110 had picked the wrong sixth number – 40, not 42 – and would have been first-place winners if they did.

“We didn’t sleep a lot that night,” Mr. Strutt said. “Is there someone trying to cheat the system?”

He added: “We had to look at everything to do with humans: television shows, pattern plays, lottery columns.”

Earlier that month, an ABC television show, “Lost,” included a sequence of winning lottery numbers. The combination didn’t match the Powerball numbers, though hundreds of people had played it: 4, 8, 15, 16, 23 and 42. Numbers on a Powerball ticket in a recent episode of a soap opera, “The Young and the Restless,” didn’t match, either. Nor did the winning numbers form a pattern on the lottery grid, like a cross or a diagonal. Then the winners started arriving at lottery offices.

“Our first winner came in and said it was a fortune cookie,” said Rebecca Paul, chief executive of the Tennessee Lottery. “The second winner came in and said it was a fortune cookie. The third winner came in and said it was a fortune cookie.”

Investigators visited dozens of Chinese restaurants, takeouts and buffets. Then they called fortune cookie distributors and learned that many different brands of fortune cookies come from the same Long Island City factory, which is owned by Wonton Food and churns out four million a day.

“That’s ours,” said Derrick Wong, of Wonton Food, when shown a picture of a winner’s cookie slip. “That’s very nice, 110 people won the lottery from the numbers.”

The same number combinations go out in thousands of cookies a day. The workers put numbers in a bowl and pick them. “We are not going to do the bowl anymore; we are going to have a computer,” Mr. Wong said. “It’s more efficient.”

Of course, playing the same number as everyone else is the worst possible thing you can do in lotteries. If all 110 had hit 6 out of 6, they would have had to split the pool 110 ways! There was academic research in the 1970’s that showed that by playing unpopular numbers (e.g. numbers above 31 because regular players love to play their birthday) you could actually make profits in the lottery on some occasions. With the rise of quick picks (which choose numbers randomly for you), I don’t think this would still be true.

(Thanks to Nicholas Feakins for alerting me to this story.)


I like your page very much. But a mexican has copy your blog:

Ophir's Razor

Freakonomics. Catchy name.

The lottery, however, is a tax on stupidity. You are more likely to get hit by lightning 10 times on the same day for a week -- clouds or no -- than you are to win the lottery.

Personally, I'd rather buy a company's stock where all the corporate officers are in jail for less than a penny a share. Now THAT has potential (Comparatively speaking).


QUESTION... Did anyone do any research as to any other commonalities of this large group of 110 who chose to play this # in the lottery? And how many of this # was manufactured? (Says they put out about 4million per day (not of the same # each day?) If so, (or even if not) this might make this an interesting pick for a group for testing in other psychological factors. Get a standard psychological battery of test of each of the participats. Then from that and other questions that would concern numerancy and (inumerancy for that matter) Could show some interesting differences and similarities of each individual. Do you think perchance they might be a good study group to use to put to rest some of the questions and punditry surrounding tax on the stupid and poor (with a taste for kum pow chicken!..HA) ?? follow them as they age make a TV show out of them (documentary) not reality. (Well, would be closer to reality). Did they all play it from a similar (proximity) location? Men? Women? How will they spend it? Where many of them Chinese? (or oriental) None of them? Ha! Such odd pickings make one wonder what other Stats lie untrolled deeply hidden?


James VI

Part of my job at a newsagent in Australia is selling lottery tickets. In the past year, I can recall only two days that I paid out more than I sold. The next question shoud now be "How many idiots will use those numbers again, and how many will still use fortune cookies' advice?" But who am I to complain? The money people squander on lottery tickets is putting me through University!

Jim Voigt


I think we should tax stupidity as much as possible. Run them into bankrupcty, and give them a stupidity exception. That we we'll at least know who they are and we can keep our distance.

Of course, it won't help find people like me. I'm plenty stupid, but I'm also too darn lazy to go out and buy a lottery ticket.


I heard a radio interview (on CBC) with somebody from PowerBall -- I believe it was Mr. Strutt, but it's not that important.

He talked a bit about the "pattern plays" mentioned in the article. For example, lots of people play numbers that make a cross, or an X, or some other pattern on the card. He also stated that 1,2,3,4,5,6 is actually very common.

That surprised me, given the human tendency to believe that this sequence is "less random" than other possibilities. But perhaps these are actually (relatively) smart people trying to follow the "unpopular play" theory you mentioned at the end of your post, and inadvertantly choosing one of the most popular plays. Kind of funny.