Computerized lottery drawings the culprit in flawed results
Dave Andres, an insurance underwriter from Altadena, California, thought he alone was in the habit of scanning hundreds of winning numbers in the lottery's least-popular game, Daily Derby.
He assumed no one else noticed that for 100 days running this winter, none of the winning times in the horse race-themed game had numbers that were repeated.
He knew that players had to not only pick the right horses but also the last three digits of the winner's finishing time. And he saw that those numbers were always like 3.70 and 4.23, and never 1.33 or 4.24 — even though the statistical probability is that duplicate numbers should appear 27% of the time.
But Andres was not alone. An equally sharp-eyed Walnut man who noticed the same aberration described it in a letter to lottery officials last month. That led to the discovery of a computer software glitch that had denied 650 players a chance to win the game's grand prize over six months.
"To whom it may concern. Something's been nagging me about the outcome of the last 150 or so Daily Derby race time draw results," began the handwritten April 18 letter from a man lottery officials refused to identify and can reach only by mail.
California lottery officials quickly investigated, eager to plug any breach in the integrity of a state-sanctioned enterprise that last year saw a record $2.9 billion in sales. The problem was fixed last week.
"This was taken seriously immediately," said California Lottery Director Chon Gutierrez. "It went from the marketing people to our security people, and they came up to us late on [April 29] and they said, 'We've looked at it and think there's a lot of merit to it.' "
As a mea culpa to players, officials plan to take about $250,000 from their advertising budget and boost prizes in the Daily Derby game over the next couple of weeks.
Gutierrez said the lottery from May 18 to June 1 would triple the race time prizes and augment the grand prize to roughly $300,000. Officials also plan to send the Walnut letter writer a thank-you gift.
"This guy identified a real problem for us," Gutierrez said.
There are two ways to win the $2 Daily Derby. Players can pick the top three horses in each race or pick the correct time it takes for a horse to win. Getting it all right nets a player the grand prize, which averaged $300,000 over the six months of the computer glitch and now stands at roughly $100,000.
California lottery officials traced the problem to November, when a contractor was replacing the 7-year-old computer hardware and software that draws the winning numbers each day.
The programmer inadvertently copied a line of software code that kept horses from appearing in more than one winning spot per race. The code guarantees that horse No. 8, for example, doesn't win both first and second place.
But when the code was mistakenly applied to the race time, it blocked the computer from picking any number that repeated.
"Now what I'd really like from you guys is a probability figure on this bizarre streak we've been getting for the last 159 Daily Derby race time draws," wrote the Walnut man.
Andres, who didn't alert lottery officials to his discovery, said, "I just didn't think anybody else would be out there wasting their time with stuff like that."
News of the glitch, Gutierrez said, has hurt lottery worker morale, especially in the unit that oversees the drawing of numbers.
"They're just morose," he said. "They don't know what to say."
The problem took just a few minutes to fix. Rebuilding the trust of Daily Derby fans may take more time. Though it generates only $200,000 of the lottery's $62 million in sales each week, lottery officials say its players are unusually loyal.
From November through April when the software problem existed, 650 of the 2,349 people who won the trifecta by picking the top three horses in the correct order had no chance of winning the grand prize because they picked a race time with a number that repeated.
California lottery officials said they did not know how many people had no chance to win the average prize of $50 for picking the race time alone. Because it is a parimutuel game, all prizes are based on how many people play.
Officials are trying to determine the names of the players affected — they already know the time and place each player bought a ticket — and are also reviewing other lottery games for software problems.
The Daily Derby trouble comes as the California lottery is preparing to join an interstate jackpot game called Mega Millions, a move aimed at boosting jackpots and the roughly $1 billion each year that the lottery generates for California public schools.
"The honesty and the integrity of the lottery is paramount to us," Gutierrez said. "Without it there can be no successful lottery. Whenever issues surface involving questions of integrity, they are dealt with instantly, aggressively and thoroughly. That was done in this case."
Gutierrez said he hoped the situation did not lead to lawsuits.
"We are dealing with this programming error in a forthright and direct fashion," Gutierrez said. "If litigation arises, we will deal with it appropriately. We believe we are treating our players fairly."
This isn't the first time the lottery has tried to make amends with players. In 2002, embarrassed by revelations that 11 of its 137 popular Scratchers games had continued after all grand prizes had been awarded, the lottery took $2 million from its administrative budget to give away $1 million in prizes. The prizes were awarded randomly through a one-time contest that allowed anyone to enter by mailing their name, address and phone number to the lottery.
Lottery officials said they did not know about the Scratchers problem until players filed a lawsuit.
Andres, who plays the Daily Derby occasionally but prefers the Daily 3, said there was not much else the lottery could do.
"It was clearly just a mistake," he said. "It seems like they're doing something to make amends."
Like lottery officials, Andres said the computer glitch hurt some players but helped others. It improved the odds of winning for those players who picked race times with numbers that didn't repeat. Their odds of picking the race time correctly dropped from 1 in 1,000 to 1 in 720. Andres was one such player.
"When I became convinced it was busted," he said, "I didn't use the duplicative approach."
Still, he didn't win. After 10 years of playing, his biggest lottery win was $666 last year on Daily 3.
"I'm not so good with numbers," Andres said, "it's just that after a while you look for patterns."
I bet that a lot of avid Daily Derby players did notice something and are upset at these guys for speaking up.
This story is a great example of how the lotteries are makign a HUGE MISTAKE by going to computerized drawings. We must do everything we can to STOP THEM.
There are so many other things that can go wrong, just like this. The whole system of independence and scrutiny is getting the shaft with the mire of computerized drawings.
CA requires the cash/annuity choice when you PLAY instead of after you win. I could see CA (or NY, or TX) drawing its lotto by computer, with the software programmed so that only players who chose annuity had a chance to win the jackpot.
The only concern with balls is that a number might be MIA, and all that would mean is that another number would be accepted. Computers are very secretive about what they're actually doing. So nobody knows if the computer is actually drawing random numbers.
Had the computer program been making a mistake that were in the players favor, it would have been discovered and fixed before going on line. Someone always know what a computer is doing, because it is following a set of instructions.
RJOh
Had the computer program been making a mistake that were in the players favor, it would have been discovered and fixed before going on line. Someone always know what a computer is doing, because it is following a set of instructions.
RJOh
RJOh,
Normally I would say that's the case, but to me, it appears that the computer programmer who programmed the RNG for the Derby game was not very good. They probably did not test very well either. I think the combination of those two things (bad programming + poor testing which did not ferrit out the problem) led to the glitch.
I can't get over what a classic case this is, in showing how terrible computers are for drawing lottery numbers.
I can't believe there is not more outrage over this.
I have contacted Governator Ah-nold about the fiasco a few minutes ago. Godspeed.
computers should not be used in any lottery game....
This story is a great example of how the lotteries are makign a HUGE MISTAKE by going to computerized drawings. We must do everything we can to STOP THEM.
There are so many other things that can go wrong, just like this. The whole system of independence and scrutiny is getting the shaft with the mire of computerized drawings.
I agree about the computerized drawings. THe problem is that the Powers That Be have the power. About the only thing the players can do is hit 'em in the pocketbook....by not playing.
the powers that be will keep doing it as long as the players keep letting them get away with it by still playing......
Normally I would say that's the case, but to me, it appears that the computer programmer who programmed the RNG for the Derby game was not very good. They probably did not test very well either. I think the combination of those two things (bad programming + poor testing which did not ferrit out the problem) led to the glitch.
I can't get over what a classic case this is, in showing how terrible computers are for drawing lottery numbers.
I can't believe there is not more outrage over this.
First, how many people that play the lottery even follow it enough to know this happened? Remember, those of us here on LP are not "casual" players for the most part. We pay more attention to all things lotto. I feel 90% or more of the ticket buyers do not. Heck, if you see how many of them come into the stores looking for the clerks to check their tickets, it shows you they don't have a clue as to what was drawn...they don't really care...let alone do they follow up on the lotto news. Therefore, what outrage?
Secondly, as you state poor programming and lack of testing of software caused the problem. Sounds to me like Microsoft was involved. (Sorry but I had to spit that out -- I have been piqued at MS for years for doing the exact same things with their monopoly software)
NOOOO!!!!! It has absolutely 100% NOTHING to do with Microsoft! Microsoft has nothing to do with someone programming the Derby game whatsoever. It is very important not to introduce an unrelated element in the argument, because people will use it to try to get rid of their fault.
NOOOO!!!!! It has absolutely 100% NOTHING to do with Microsoft! Microsoft has nothing to do with someone programming the Derby game whatsoever. It is very important not to introduce an unrelated element in the argument, because people will use it to try to get rid of their fault.
Todd, that was my personal shot at MS. I was not saying that MS was involved. My point is that even MS does the same things...i.e. poor programming and lack of pre-testing (which used to be called Beta Testing in the early PC days. Poor programming is rampant these days because programmers have so much RAM and ROM to play with that they aren't forced to be disciplined.
I've been "into" personal computers since the early 1980's when the C-64 came out. Those days, you not only had to program your own software (as there wasn't much on the market) but you also had to be efficient; there simply wasn't much RAM or ROM available.
Now all these technological "advancements" are making it easy for sloppy programming. And in the name of money and greed software designers use the endusers to discover all the glitches. I personally find it very sad.
I programmed the C-64 as well, but it sounds like I started a little earlier than you: I also programmed Commodore PETs and VIC-20s. I learned machine code in order to squeeze code in a small space (plus make it run faster), so I understand your point about sloppy coding.
We all tend to look down on the bloated coding of today, but an honest appraisal of programming back then would acknowledge that programs were just as buggy as today. People love to program, but they hate to test, and that's been true all along.
That "human" side of computers is one of the reasons I am so against computerized drawings. Why bring all the problems of computing into lottery drawings?
What if the California Lottery discovered that problem on their own? Do you think they would have told anyone? Or would they have quietly fixed the program, and people would have never have known that they had no chance of winning the game?
As computers become more prevalent in drawings, the problems will increase greatly. There are just too many points of failure. So sad that lottery directors are in such a rush towards that big, bad idea.
I programmed the C-64 as well, but it sounds like I started a little earlier than you: I also programmed Commodore PETs and VIC-20s. I learned machine code in order to squeeze code in a small space (plus make it run faster), so I understand your point about sloppy coding.
We all tend to look down on the bloated coding of today, but an honest appraisal of programming back then would acknowledge that programs were just as buggy as today. People love to program, but they hate to test, and that's been true all along.
That "human" side of computers is one of the reasons I am so against computerized drawings. Why bring all the problems of computing into lottery drawings?
What if the California Lottery discovered that problem on their own? Do you think they would have told anyone? Or would they have quietly fixed the program, and people would have never have known that they had no chance of winning the game?
As computers become more prevalent in drawings, the problems will increase greatly. There are just too many points of failure. So sad that lottery directors are in such a rush towards that big, bad idea.
Wow ! The PET and the VIC-20 -- brings back old memories. You know, for their time, those were darn good machines !
Agree the programs back then were buggy. But one back then didn't release the program chock full of bugs. The programmers used to ask selected users to Beta Test it to discover the bugs. Actually, I had a lot of fun beta-testing. Back then programmers used to often say something like Don't be afraid to hit any combination of keys you want to see if the program crashes. The programmers then realized that the program code wasn't completely smooth on first try. But they also had enough self-respect and also respect for the potential users not to rush it out for use.
And these were utility prgrams, WP programs, etc; not vital programs like an OS ! Seems like every week I'm being informed I have a new security patch to install for WIN XP here. That's really shoddy ! If you have an OS, and basically a monopoly on it, and a browser that is basically a monopoly (since I've run into sites where the pages won't fully recognize any other browser but IE whether I want to use it or not) then gol darn it, don't release it so full of holes ! MS has more than enough financial resources to do it right ; not as if their employees or Bill is/are going to not be able to buy groceries this week if they don't rush it right out there. Perhaps Bill has the wrong people programming.....it seems from the overt evidence that the hackers are better at using his code than his programmers are ! It is just a real sore spot with me.
As to the same shoddy workmanship in the CA lottery...No, if they had discovered this on their own, I certainly don't think they would have mentioned it to the public. Computers for lottery drawings are a terrible idea ! You can already see from thestates using them that there is often nothing "random" about RNG. Right now in fact, I think it is ruining the games...there are some really lousy and weird things happening in the games now ; and they fly against statistics. The proof is in the recent draw histories of the individual games. If I had a Magic Wand I'd make the RNGs disappear.
i agree i'd make COMPUTERIZED DRAWINGS poof gone...
I would agree that this is a classic example of why computer rng's are not always the best. I always thought it was fine for keno (which I don't play) but for large jackpot games it's a disaster waiting to happen. While people will probably ignore small news like this on a minor game eventually the same news will happen on a major state game like the California superlotto and that will be the real test when people find out the game was rigged to make losing tickets.
dvdiva: great points! Furthermore, who is to say it hasn't already happened?
Maybe it has.
A computer can make a missile turn on a dime and then hit a penny in the middle of a forest so how do we absolutely know a state lottery computer isn't programmed to pick a selection of numbers not picked by any player? We don't know!!!!
Delaware is a case in point. Despite protests to legislators its games are still being "picked" by computers. And its jackpots are SO low.
Conclusion: there is nothing random in computer selected numbers. Computers are just another way for a state like Delaware to rip off a player who actually thinks h/she has a "chance" to win. Some chance when the computer eliminates that very chance!
i play with balls and balls only!
A computer can make a missile turn on a dime and then hit a penny in the middle of a forest so how do we absolutely know a state lottery computer isn't programmed to pick a selection of numbers not picked by any player? We don't know!!!!
Delaware is a case in point. Despite protests to legislators its games are still being "picked" by computers. And its jackpots are SO low.
Conclusion: there is nothing random in computer selected numbers. Computers are just another way for a state like Delaware to rip off a player who actually thinks h/she has a "chance" to win. Some chance when the computer eliminates that very chance!
mike:
That's "s/he".