If lottery employees cannot partake in the lottery, it is because of fear/suspicion of tampering, and the lottery is doing a smart (if you ask me) PR job.
The commissions battle tampering everyday with the step-by-step precautions they take. In the early 1980’s, in Pennsylvania, the daily lotto was rigged (balls were weighted down, so as to not pop us as winners). I will try and find a reference for you.
If lottery employees cannot compete, it is to minimize the chance or temptation of tampering.
Also, it is a PR thing. If a lottery official was to win and it was public, or a few of them won, all sorts of people (many who don’t understand the lottery) will complain that they either rigged it or had access to secret “systems”.
Keeping employees out is a way to let everyone maintain faith in an objective random process, with minimal chance for tampering.
Again, there is no system. The lottery orgs do not have the world’s greatest mathematicians locked away with the only systems known to win.
They are protected from losing by capping the number of times a number can be played ( I know that triple numbers in PA are routinely locked out after “x” number of people play them. That’s because they are so popular, if they ever came out, it’d be a big loser for the state.) For big games, it gets more complicated, but they don’t pay close to the odds, so they have a lot of padding. Plus, they know there is no system other than tampering.
Okay, this guy you cite is just finding patterns in previous lottery drawings and assuming that those patterns will repeat. I don’t doubt that you can find patterns. As I stated before, the number 444 came out more than other numbers in PA, and I think 777 came out “frequently” relative to other numbers. This was going on for a while…but you know what?..people have spent hundreds of days betting on 444 and 777 and they haven’t come out as often.
There is nothing special about the patterns of repeating numbers. Random doesn’t mean never repeating.
And this guy is just finding things that fit his model.
What he is laying out will never topple this fact:
The odds of any number coming out when you have three bins 0-9, drwaing one number from each is 1000 to 1. Nothing can predict that one combo has a better likelihood of coming out. Nothing.
If zero is the least frequently drawn ball over the past 2500 drawings, it means nothing. It is the type of crap that makes addicts out of the uneducated, and that would drag us into Great Debates.
I’m not sure I understand this sentence. If you don’t reverse-engineer an RNG exactly, then it seems like it would be pretty useless to you. RNGs are chaotic, so that small changes quickly lead to completely different patterns.
Obviously, there is no system. Just as obviously, lottery officials would love for you to believe there’s one. Some states have gone so far as to have lottery “cheerleaders” go to senior citizens’ homes and tell people “It could be a good day for 17 or a bad day for 17,” which is clearly crap.
I’m interested to hear PA cuts off sales of certain numbers–I’m not aware of this in any other state. It would lead to complaints of those who were prevented from buying their usual number when it hit. It certainly is not unusual for states to lose several million dollars on one draw every now and again. The day after TWA 800 crashed, 800 hit in CT and they took a bath. Maryland lost several million this year when 1111 hit on the Pick 4.
I’ve had a lottery official tell me that the only thing they were really worried about was the contractor (most state/national lotteries are run by either G-Tech or AWI, not the state itself–and that’s where the real money/corruption is) could tuck code for a specific winning keno number at a certain time, and it would never be caught. Of course, that doesn’t offer winnings as large as PowerBall/The Big Game, but those are done with balls.
“to some extent” means that you can, in fact, reproduce some characeristics of the RNG’s output without knowing what the actual RNG function is. In other words, you could, in theory, take a lot of data that came from a single RNG function, crunch the numbers in every way that it’s possible to think of, and write a program that generates numbers with those characteristics. Your actual function may be the same one that’s used in the original RNG, or it may be a huge hairy fuzzball of an ad-hoc program that works only because it shares some essential property with the RNG in question.
Note that the above is pure theory. In the real world, it takes a hell of a lot of data to model even general trends accurately. A RNG based lottery system would not only tend to have insufficient results (for a while yet), but would have the additional and obvious advantage that lottery results tend not to report potential seed data (time in milliseconds or what have you). On the other hand, the Keno story above does demonstrate taht it’s possble to break such a system, assuming it’s weak enough.
Incidentally, the balls used for the big drawings (PowerBall and such) don’t use Ping-Pong balls. They are hard rubber, balanced and matched, and run about $100 apiece for the trouble. Furthermore, there are three or more identical drawing machines readied before the drawing. The one to be actually used is randomly picked before the show. I hope they use a fair means to determine that, probably an electronic RNG.
I’m 99% certain that in PA they lock out certain numbers when they are played heavily. I am drawing on memory, becausue in my family store in the 90’s we had a lottery machine, and I remember 777 (called “triple 7”), 444, and 1111 as being locked out occassionally.
1111 stands out the most. I remember that being played heavily and being locked out often.
Nowe, even when they come out, the lottery loses significant money, but they have a controlled loss. I guess that they are protecting themselves from the theoretical catastrophe that could occur if a certain number sells through the roof and then comes out.
There is a dollar and cents issue, but there is an administrative nightmare in paying out to huge numbers of winners.
Jeez guys, I know a gooood bit about probability. I’m not someone who thinks they’re can be system for winning anything at a casino - I can count up to 6 decks in blackjack. I know my odds. Do any of you remember how the enigma was cracked? It is an unbelievable story of matching aspects of the data that noone would ever conceive could have something to tell via reverse engineering, but they did. So please dont respond that they are all hoaxes, cuz of course they are. And Philster, your still assuming the the original randomn generator is randomn. I was a commodity trader too, and lord knows i know about finding patterns when there aren’t any. So do me a favor when asking and assume we’re all intelligent. My question is to mathematicians - COULD A "RANDOMN NUMBER GENERATOR ALGORITHM SHOW STRANGE ATTRACTORS, IF YOU WILL, AND IF SO, WOULD THEY BE OF AN EXTENT TO REAP A BENEFIT OVER AND ABOVE THE RISK. I know that last part is a bit much to ask. Just your opinion.