Is it possible to predict Lottery number accurately based on historical data?

Saint

Saint

Valued Senior Member
Is it possible to predict Lottery number accurately based on historical data?
So that I can hit jackpot. :D
 
Since there is no way to make the choice of jackpot number truly random, in theory one can predict a Lottery number using historical data.
If the Lottery folks did their jobs right, you might need a million years of historical data to get a useful prediction (narrowing of odds sufficient to get a reasonable chance of hitting). But if the people pseudorandomizing the number overlooked something, or got lazy, you might not:
https://www.inc.com/bill-murphy-jr/...ng-over-over-for-55-weeks.html?utm_source=red
 
You can make it close enough to random for lottery purposes. If to predict the lottery it would take more time than mankind will exist and more computing power than exists...it's effectively random.
 
Saint said:
Is it possible to predict Lottery number accurately based on historical data?
So that I can hit jackpot. :D
Click to expand...

post the number of numbers needed to win and the total number of numbers to choose from.
then we can work out the probability of you winning.
it will be a very small chance.
i am guessing it will be roughly around 1 chance in 7 million
 
Predicting the winning lottery number is trivial - historical data not required.
Providing the lottery runs for long enough then all possible variations will result - but no one can state when any particular combination will turn up.
I have no idea whatsoever about the lottery in question, but (from what little I know of the UK one) it's 7 numbers chosen from 1-49 (could well be mistaken here). My prediction (anyone is free to use it, no reimbursement desired): 2,3,17,28,31,39,42.
Wait long enough and that sequence is bound to happen.
You're welcome.
 
I once asked a friend who kept track of which numbers were "hot" and which numbers were "due". I asked her why she didn't just pick 1,2,3,4,5. She said, "Are you kidding? Do you know what the odds are of those winning?"

"Yes. Do you?"
 
phyti said:
No, because the results are independent of time.
Click to expand...
They aren't independent of sequence.
Pseudorandom numbers are correlated - if the drawings are spaced in time, they are correlated in time. That correlation allows prediction with a much larger probability of hit than random. In theory.

It's just that the correlation is buried in a vast space. In theory.
If nobody screwed up.
 
The big lotteries in Australia don't use pseudo-random numbers. They use a physical machine that mixes a bunch of balls around, and draws them out at random. Is that not what happens in the US? Are lottery draws computer-generated?
 
James R said:
The big lotteries in Australia don't use pseudo-random numbers. They use a physical machine that mixes a bunch of balls around, and draws them out at random.
Click to expand...
That describes a pseudorandom number generator (!). A good one, to be sure, if the physical circumstances are well designed and monitored - I have no idea how much historical data it would take to fish out the correlations, but a hell of a lot seems to cover the matter.

(!) I see I have screwed up my terms, and posted error. "Pseudorandom" does not refer to any old approximately random number generated by a Newtonian deterministic process, but only to those generated by a computer algorithm. Apologies - I go to soak my head.

From Wiki, missing some " " qualifiers but otherwise clear enough:
Random number generators can also be built from "random" macroscopic processes, using devices such as coin flipping, dice, roulette wheels and lottery machines. The presence of unpredictability in these phenomena can be justified by the theory of unstabledynamical systems and chaos theory. Even though macroscopic processes are deterministic under Newtonian mechanics, the output of a well-designed device like a roulette wheel cannot be predicted in practice, because it depends on the sensitive, micro-details of the initial conditions of each use.
Click to expand...
Point is, one can in theory narrow one's choices and adjust one's probabilities using correlations found in historical data, thereby significantly improving ones odds of hitting the jackpot, in any lottery system that generates its numbers via Newtonian deterministic process. In theory.

In practice what allows such a breakdown is a glitch in the physical system - somebody overlooked something, got careless about maintenance, fudged the procedure, etc. Humans are not built to generate randomness, or comprehend probability. Most don't even floss their teeth. So the historical data approach is not as hopeless as the theoreticians claim.
 
Last edited:
Ah, okay. When I hear "pseudo-random numbers" I think of the algorithms typically used by computers to generate random numbers, as per the "random()" or "rnd()" functions typically available in most high-level programming languages.

I don't dispute that a chaotic physical system - like the lottery ball machines that are usually used - is deterministic. Nevertheless, it is practically unpredictable.
 
iceaura said:
They aren't independent of sequence.
Pseudorandom numbers are correlated - if the drawings are spaced in time, they are correlated in time. That correlation allows prediction with a much larger probability of hit than random. In theory.

It's just that the correlation is buried in a vast space. In theory.
If nobody screwed up.
Click to expand...

A computer can list all possible sequences of a set of symbols.
It cannot predict when a given one will appear.
Rare events can happen anytime.
 
Also, the lottery system has no memory, so every drawing is as if the first one. I.e. there is no relationship between drawings that would allow any prediction.
If the system had a memory, it allows for the possibility of a biased system.
 
spidergoat said:
If there is no when, it's not a prediction.
Click to expand...
predict
/prɪˈdɪkt/
verb
verb: predict; 3rd person present: predicts; past tense: predicted; past participle: predicted; gerund or present participle: predicting
say or estimate that (a specified thing) will happen in the future or will be a consequence of something.

Like I said: provided the lottery runs for long enough... :tongue:
 
phyti said:
Also, the lottery system has no memory, so every drawing is as if the first one. I.e. there is no relationship between drawings that would allow any prediction.
Click to expand...
It doesn't need a memory to produce correlations and frequency biases - any more than a coin flip, which likewise has no memory, does.
Last I checked, the coin flips at sporting events were running between 51/49 and 54/46 for whichever side was "up" immediately before the flip. It's because of the difficulty of spinning the coin exactly perpendicular to the horizontal plane.
phyti said:
It cannot predict when a given one will appear.
Click to expand...
What we need is a sufficient bias in the odds, a correlation or frequency imbalance. Not an exact and definite prediction. Every deterministic system will give us that - over time.
 
James R said:
Ah, okay. When I hear "pseudo-random numbers" I think of the algorithms typically used by computers to generate random numbers, as per the "random()" or "rnd()" functions typically available in most high-level programming languages.
Click to expand...

'Seeds' are used in the random function algorithms so you have to be careful that you don't re-use the seed or you will receive the same results. So if you use something like a time function as your seed you must make sure that you don't run it at exactly the same time or your results will not be random.

https://en.wikipedia.org/wiki/Random_seed
A pseudorandom number generator's number sequence is completely determined by the seed: thus, if a pseudorandom number generator is reinitialized with the same seed, it will produce the same sequence of numbers.
Click to expand...
 
what i have always wondered about is
1 a computer picks peoples random number choices
2 people pick their own numbers
3 people do not pick random numbers
people mostly pick sets of numbers attributed to numbers of special significance & people have preferences for numbers they like the sound of or the shape of and so pick those.

i use to watch people pick their own lottery numbers on occasion.
it was a rare occasion that someone would just pick random numbers
9 out of 10 people put considerable thought into the numbers they pick.

the funny thing
the computer random number selector usually always wins the 1st prize.
mostly because around 75% of all tickets are bought as the option to have the computer select them.

yet the computer is not tumbling balls in a bin and having them roll out one by one.(it would cost a considerable amount of money to make one of these to pick your ticket numbers but it would look really great and be fun for kids and family's
[for a 1 dollar donation to the national children's hospital you can all stand and watch the machine pick your lottery numbers] have you seen young parents trying to pause and make a special mentally stimulating moment for their small child while they take a moment to gather their sleep deprived mind?)

side note
i think all big lottery's should be run by a government agency where they put all the winnings(the winning pot profit margin) back into the community.
and i also think there should be maximum winner pots of around 1 average house price per winner and the winning pot be divided into those 1st tier winners then divided again into 2nd tier and so on until the pot is distributed.

i think selling massive(tens to hundreds of millions) lotterys(tickets) to working class people is like preaching communism to 10th graders
 
You must log in or register to reply here.
Back
Top