Pennsylvania's Auditor General says that he has full confidence in the integrity of the Pennsylvania Lottery following a review in response to an investigation in 2017 that found some Pennsylvanians have claimed lottery tickets with seemingly improbable frequency.
In an interview Monday, Eugene DePasquale said he met with lottery officials and was given a thorough tour of its operations and an explanation of its security procedures.
DePasquale added that the lottery told him it had investigated the most frequent claimants and found no wrongdoing.
"I have a high degree of confidence from the beginning of the system to the end that it's a clean system," DePasquale said.
But statisticians said that many of the winning patterns PennLive identified remain improbable, raising questions about the thoroughness of the review.
Pennsylvania's most frequent winner, Nadine Vukovich, claimed 209 prizes from instant tickets worth $600 or more between 2004 and 2016, winning $348,000.
Investigations into unusually frequent prize winning in other states have sometimes found those wins rooted in fraud; from retailers secretly stealing winning tickets from prize claimants to schemes where people cash in winning tickets on behalf of others to avoid having debts deducted from their prizes.
Philip Stark, a statistician at the University of California Berkeley, analyzed Pennsylvania lottery data. He found that luck or frequent play couldn't plausibly explain how more than a dozen frequent winners in Pennsylvania — including multiple retailers — could win so often.
In the case of Vukovich, a Mechanicsburg veterinarian, Stark found that even if every Pennsylvania resident spent $7.8 million on lottery tickets, there would still be less than a 1-in-10 million chance that any of them would win as many prizes as she did.
Asked on Monday how the Pennsylvania Lottery explained the frequent winning, DePasquale said lottery officials maintained that the claimants were simply frequent players. He said they gave an explanation for why the 2017 investigation's analysis was incorrect but he couldn't remember their argument.
In a follow-up email asking for more information, Susan Woods, a spokeswoman for the Auditor General's office, said her office had no further comment.
In an interview on Wednesday, Stark said while it was always possible there was a bug in the computer code used to run his calculation, he had checked his calculations thoroughly.
If anything, Stark said, his calculations were likely underestimating the improbability that certain claimants, like Nadine Vukovich, could win so often.
"I don't know how to explain how small these numbers are," Stark said. "Even if they had spent those millions and millions of dollars, even if 12 million people had spent one million dollars, it would still be incredibly unlikely that any of them won this much."
Stark added that his method, developed with two other mathematicians in 2014, was well-vetted and based on long-standing statistical principles. That same method exposed suspicious behavior in the Florida Lottery that led to major security reforms.
Three other statisticians consulted this week said Stark's methodology was sound.
Bill Notz, a statistician at Ohio State University, who examined frequent lottery winning in Ohio several years ago, said Stark's method was more sophisticated than his own analysis and particularly conservative.
"In other words, when his calculations show that a probability is very small, the true probability is even smaller than the value he calculates," Notz said.
Ronald Wasserstein, the executive director of the American Statistical Association and a former statistics professor, also said he supported Stark's method.
Based on his review, Wasserstein said, the frequency of prize claims by a number of Pennsylvania's most frequent claimants couldn't be plausibly explained by luck.
"I would dare the lottery to find a competent statistician who would agree that it's just frequent play," Wasserstein said.
Jan Hannig, a statistician at the University of North Carolina at Chapel Hill who has analyzed frequent winning in the North Carolina Lottery, likewise supported Stark's analysis.
Hannig said that, among Pennsylvania's top ten most frequent winners, he felt the frequent claims of a few could plausibly be explained by gambling addiction.
But the frequent prize claims of at least five of them, he said, were clearly suspicious.
"There's no way this is by luck," Hannig said.
In response to questions, the Pennsylvania Lottery reiterated Wednesday that it believes the state's most frequent claimants are simply frequent players.
Asked whether it had changed security procedures in response to the 2017 investigation, the Pennsylvania Lottery declined to provide details.
"New measures are implemented on a regular basis, but we do not discuss specific measures for obvious security reasons," said Gary Miller, a spokesman.
Pennsylvania's most frequent winners of $600 or more
- 209 wins ($348,048) - Nadine Vukovich of Mechanicsburg, Pa. (4/2/2004 to 12/22/2016)
- 162 wins ($1,319,556) - Neil Perretta of Chalfront, Pa. (1/14/2004 to 12/14/2016)
- 131 wins ($273,163) - David Cirocco of Pittsburgh, Pa. (1/15/2000 to 5/5/2015)
- 127 wins ($256,150) - Colleen Dunne of Phoenixville, Pa. (9/13/2009 to 12/28/2016)
- 122 wins ($174,277) - Steven Seibert of Corapolis, Pa. (1/12/2004 to 5/19/2016)
- 111 wins ($338,833) - James Hilton of Malvern, Pa. (2/20/2001 to 12/31/2016)
- 110 wins ($606,110) - Austin Tucker of Pittsburgh, Pa. (3/14/2005 to 2/12/2016)
- 110 wins ($227,550.50) - William Sestito of Philadelphia, Pa. (7/15/2008 to 3/18/2016)
- 106 wins ($1,748,295) - Angelo Scanzello from Conshohocken, Pa. (11/16/2005 to 12/27/2016)
- 104 wins ($262,500) - Harry Turner from Portsmouth, Va. (1/6/2000 to 2/23/2008)
Don't know what to think of this.
Now some crazies will find some of these individuals and cause them bodily harm. The red flag is flying compliments of the Pennsylvania lottery or perhaps target is a better word.
Be curious as to what other LP members have to say, I'm really @ loss for words right now.
If it looks like a duck, quacks like a duck and walks like a duck then it is a DUCK.
There is a lot of mendacities here. Lying is not good for our Country nor it's people.
How is it that 7 times lottery winner,
Richard Lustig has never been investigated or has he?
Not sure if he was investigated, but let me cover some points about some of the wins he had. I am not sure if he is a retailer either.
Win 1: $10,000 - Jan 1993 (scratch-off ticket)
Win 3: "Wheel of Fortune" Holiday trip to Los Angeles - June 2000 (valued at $3,594.66) (scratch-off ticket "2nd chance drawing")
Win 4: Elvis Holiday trip to Memphis - Oct 2001 (valued at $4,966) (scratch-off ticket "2nd chance drawing")
Scratch wins have better overall odds of winning then most "jackpot games". PB and MM have about a 4% chance of getting your money back or more. In Illinois, 20$ scratchers have 30-40% chance of getting your money back or more. This way you "lose" less money playing them. Also 2 wins were second chance drawings. Not as many people enter them. Also if he digs thru trash, he could enter more.
Win 2: $13,696.03 - Aug 1997 (Florida Fantasy 5)
Win 5: $842,152.91 - Jan 2002 (Florida Mega Money)
Win 6: $73,658.06 - Nov 25th 2008 (Florida Fantasy 5)
Win 7: $98,992.92 - Aug 9th 2010 (Florida Fantasy 5)
Florida Fantasy 5 has an overall odds of 13%, and he might have a "system". This guy has 17 wins over 17 years, not as much as the guys in the linked article. What I did find unusual is that he sound like a regular "big" player, why aren't there more $600-$1000 wins?
Now my comments for the winners in the original article.
For the retail owners, do I think many can get their wins by "discounting" or flat out theft, YES.
But let us consider other possibilities.
If a person (retailer or not) wins $600+ in a year, they can write off their losses up to the winning amount. By getting a tax break on your losses, the amount you actually lose is less, therefore your risk is less. With the lower risk, you can gamble more.
Second, for retailers and scratchers, lets talk about "card counting". Just like in Blackjack, you count the cards to increase your chances of winning. I just want to talk about the 20$ scratchers. Here in Illinois, there is a 30-40% chance of winning, anything. In each deck, there is at least one 100$ ticket. If you are a retailer that cashes all tickets up to $600, you will know the "big" prize came up in a stack. Also with a 1 in 3 chance of winning your $20 back (or more), sit and watch, if no one wins after 3-4 draws in a stack, go for it because you will have a BETTER CHANCE then anybody else. Yea, if I was a retailer, I would start by buying a couple books to check for patterns like average number of losing tickets between any winner. How often there are 2 winners in a row, etc.
Then I will get a notepad and track each ticket sold with a single letter code for win, loss, walkout, and "Big Win", then buy accordingly. Doing that you can easily double your odds of winning "something" and increase your odds of a bigger win because you will not waste money on a book if somebody won big already.
Of course if a retailer cashes in tickets not sold at his store, he probably is doing something illegal.
Were any of the frequent winners retailers? If it's in the article I missed it.
Perhaps it shows it is possible to win with systems. Some of these winners might have used systems. I see there are some strategy books even for instant scratch lotteries.
Nadine's winnings occurred between April 2004 and December 2016. That is a total of 153 months. She had 209 wins. You're talking on average, 1.3 wins a month over 12 years.
I love how these statisticians always claim the odds are so steep, people would have to spend millions of dollars in order to win so often. That may be true if you're only buying quick picks. But serious players study the games they play, and know their states well enough to know how to take advantage at certain times.
So the audit found in favor of the lottery and against the players.
What a surprise. I am guessing they were paid off to find in favor of the lottery.
at noise gate, So was it a torn meniscus or a strain?=0p
When Dr Fanton says its a torn meniscus, its a torn meniscus!
Was an outpatient procedure as well fellini, think it took less than 30 minutes. It's been nothing but marvellous ever since.
Why is it so hard to believe that a few people can find order in chaos....that's my approach, and it works better thàn birthdays or whatnot...
Wait, so they're saying that one would have to spend almost 8 million to win almost 350K?
What?
I fail to see how those wins are classified as unusually high. Just because they're different in the sense that they are more than average does not make them in any way in my opinion unusual or duplicitous. Statisticians aside, I fail to see how even a statisticians point of view is relevant. These are vague and general opinions and observations that have nothing to do with fact. In my humble opinion. What is the lottery saying, that people are just willing to throw their money away without any hope of getting anything, except once in a blue moon? Very strange. As the lotteries keep raising prices, and keep the payouts low and the wins lower. Very strange. And it says quite explicitly, no wrong doing found.
"The numbers aren't unpredictable, the human mind is unpredictable ", lol
Point well taken. Thank you for your excellent analysis of the situation @ hand.
NG, I do believe the winners referred to in the story would love your signature, as well as anyone that wins frequently. Statisticians be <snip>ed...
The store where I purchase my tickets sells a lot of winners!
Very small town.
Top 25 of lottery retailers in PA!
They have a special touch screen lottery machine there.
"Were any of the frequent winners retailers?"
The article clearly says there were "multiple retailers" among them.
"You're talking on average, 1.3 wins a month over 12 years."
That could be one of the reasons it would be suspicious. Lottery prizes may return 50 to 60% of the money spent on tickets, but in most games the majority of the money is paid out as small prizes. That means that for most games you'll get far less than 50 to 60% back as large prizes. Of course the PA lottery knows exactly what games and prizes all of those players actually won.
Those 209 prizes for the most frequent winner average out to $1665 each (only 1 of the 10 listed had a lower average prize), and they were all at least $600. It's possible that the majority of them were for pick 3 or pick 4, in which case you could expect to win $348k by spending $696k on tickets. That's an average of $58k per year, which would be possible with a veterinarian's income. If that was the case the lottery would know.
For scratch games or jackpot games it's a very different story. Looking at a few of PA's scratch games the best odds I saw for winning $1000 were 1 in 120,000 on for a $2 game. There's a $1 game with odds of 1 in a million. Factor in multiple prizes of $1000 or more and it takes an average of $20,000 or more in tickets sales for every $1000 in prizes. In that case wining $348k would require spending $6.96 million, on average.
It's certainly possible to win $348k by spending far less if you get lucky and win one or two really big prizes, but the number of prizes means that there had to be a lot that were over the $600 threshold but not over $3 to $5k. That means you'd have to beat steep odds over and over and over. That's why winning consistently takes luck out of the picture unless you spend a hell of a lot of money.
"even if every Pennsylvania resident spent $7.8 million on lottery tickets, there would still be less than a 1-in-10 million chance that any of them would win as many prizes as she did."
Without seeing some details about how he arrived at that conclusion my initial reaction was to be a bit skeptical, based on the "every resident" and "any of them" part, together with having seen so many quotes from statisticians who are basing the odds on buying only 1 ticket. So let's do a rough calculation.
Let's figure that you buy enough tickets for a theoretical 50% chance of winning. Using the numbers above that means spending $10k or more for a 50% chance of winning $1000. If you repeat that 209 times you'll have spent a bit over $1 million, and the chances of winning all 209 times is .5 to the 209th power. The chance of that is 1 in 82,300,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000. Or 1 in 82.3 trillion trillion trillion trillion trillion trillion.
Let's figure some people spend way too much on tickets, and knock that down a bit. Figure you've spent enough to have a 90% chance of winning, and repeat it another 208 times. That's .9 to the 209th, which is "only" 1 in 3,658,605,030 but you'll have to spend $3.97 million.
So even if I'm off by a couple of decimal places if each of the almost 13 million residents of PA (a wildly implausible idea, of course) spent $3.97 million on tickets there's much better than a 1 in 10 million chance that one of them would win that often, but it's still not likely.
In the end, it's wildly unlikely that even one of those people won that many prizes as a result of luck and probability, assuming they didn't do it by spending 2 or 3 times as much as they won on the limited selection of games that would actually return 35 to 50% of their money as prizes of $600 and up.
Yes, some people do spend a lot of money on the lottery. Many years ago I would stop at the 7/11 on my way home from work at night. The clerk was a friend and we would shoot the breeze for awhile. Anyways, there was a guy who would come in and regularly drop $200 at a time on scratch off tickets, buying 5 or 10 at a time. I couldn't say if he ever hit anything big or not, but he always took his winners with him when he left.
There is also a lady I've seen in a couple of stores who spends $125 - $150 at a time on Pick 3 and Pick 4 tickets. She does win regularly.
If you're spending that much on P3 - P4 you can cover a pretty good range of numbers and get a good amount of straights each week. At least on the P3 anyways. And if you can get 1 or 2 a month on the P4, great!
I don't like stories like these. This is a negative story. I've seen people drop $1,000 cash on a jackpot game. Vendors tell me all the time about people dropping $600 on a jackpot game one drawing. They also tell me they get $3 back! Look it's nothing new. People spend money on these games all the time. $200 on scratchers ain't nothing. Trust me, I see it all the time; from my own hand as well (lol). What I don't like about this type of story is that a person will become investigated for 'winning too much'. It's a game, people win people lose. Leave it at that. I am aware that some people buy winning tickets at reduced prices. There are many people who owe taxes and are afraid to face it. I understand that. Among other reasons as well. However, for a regular citizen player, this kind of thinking is unfair against the player. And there should never be a penalty just for beating the odds. That's what people dream of. That's their slogan. Dream. So whatever yeah they found no wrong doing. So there should be no suspicion.
This reminds me of that female Stanford statistician who won all those jackpots playing scratchers in Texas. She won it all legally. She just had her own method and also invested a lot in buying tickets that did not win.
Yes, if I remember right - her method was to buy ALL the tickets from the same store with the mind set that the jackpot winner will eventually be sent to the store. She worked with the store and bought the entire pack when they got it and then waited until they got the next one, so no one but her had access to those tickets from that store.
And it worked!!
"so no one but her had access to those tickets from that store. "
It wasn't directly about preventing anyone else from buying tickets for the game from that store. Buying all of the available tickets was about getting lots and lots of the undistributed tickets sent to that store in the hopes that the unclaimed prize that she thought was available would be in one of the packages that was sent to the store. I've got no idea how fast the tickets were selling in general, but let's figure 1000 per week throughout the state. If her store has only one pack and she buys all 100 tickets that's just 10% of the tickets for that week, but the lottery has to ship another pack to that store while other stores are still selling the stock they have. Maybe the lottery starts shipping more than one pack, or maybe one pack at a time is fast enough compared to other stores, but at some point most of the tickets being shipped are being shipped to that store. If the big prize is among the tickets available for shipment there's now a much greater chance that it will go to her store instead of a different store.
That gives her a good chance of being able to buy the winning ticket, but how much does it cost to win the prize? If half of the tickets have been sold and only 1 of 3 big prizes has been claimed the odds of finding one of those unclaimed prizes in the remaining half is much better than the overall odds. She probably believed that as a matter of statistics each ticket she was buying had an expectation value that was higher than her cost for buying the ticket. Because she was collecting annual payments from a previous prize each ticket she bought was a tax deduction, so her actual cost for $1000 worth of tickets may have been something like $750 or less. Factor in the small wins from $1000 worth of tickets and the cost goes down even further.
Of course there was always a chance that somebody had another big prize but hadn't claimed it yet, or it could have been one of the last tickets in a pack that was already sitting in a store 350 miles away, but she presumably figured the odds were in her favor and her total cost for tickets was less than her expectation value.
Of course even with a few big prizes and a couple of dozen modest prizes, and without accounting for any edge she had, the odds against her winning those prizes is still a small fraction of the odds against winning 209 prizes in the PA lottery.
I still find it unsettling that the State is complaining at how much a person has won. My opinion on this matter would be to complain about how much the State has won rather.
Thanks G- l like my signature too. Took me sometime to get one that "says all things worthwhile." There is another signature out that that says something like... " don't give up on your dreams just because someone gave up on theirs." Saw the movie 12 years a slave last night for the first time..
Ditto on the liking of the signature. BrainyQuote has oodles of quotes by Bernard Shaw.
Good luck to all on lottery wins this year!
Wtf I’ve been doing...thats insane but congrats to the heavy hitters in Pa.