Quote: Originally posted by rcbbuckeye on December 13, 2009

This subject is a little over my head. But I can't see how a RNG can be completely random AND be controlled. I pick my own numbers, although at times I will use LP's RNG to pick my numbers. BUT.... I still choose the combos I will play from the RNG.

I choose the numbers to fit what my research of past results has been over the life of the game in its current matrix. Then we see a drawing like last night's PB. 12-13-14-35-41. That is a random drawing, but I could never bring myself to play numbers like that. I'm beginning to think we are hurting ourselves because we have our idea of what a winning combo "should" look like.

Some things don't change even when the matrix change. I usually allow three consecutive numbers if the combination fit the profile I'm looking for because even if the winning combination doesn't have all three, it may have two of them and I would hate to miss out on a 4of5 because I rejected a triple.

* you don't need to buy every combination, just the winning ones *

I have been working on using the logistic map or other maps to be an RNG for about a year now.

I greatly feel they might somehow be used with past lottery numbers to choose winning numbers.

I would really like to see other people ideas-code or whatever on the topic. Sometimes other peoples view helps.

Jadelottery's pick 3 and pick 4 program and information has been giving me ideas, along with the logistic map for another type of program.

Any help or ideas you come up with may be even better then mine. I have been wondering if anything mechanically loaded off prior genuine winning numbers could choose the next.

Wouldn't attempting to associate the numbers drawn with any kind of number sequence require the false assumption that the order of the numbers drawn is known and that each number drawn is independent of the others. With a typical pick 5/6 game, 5/6 numbers would be drawn from the same lot and then published in sorted order so the draw order within each group of 5/6 numbers is unknown and of course each number within each set of 5/6 is exclusive of the other numbers i.e.: there would be no repeats within each grouping of 5/6 numbers hence you can't really fit any kind of sequence to the individual numbers. Wouldn't it be necessary to reduce the 5 or 6 numbers drawn into a single unique number before attempting to fit it to a function or sequence?

I guess one way would be to treat each draw as a separate digit of a value. With the Texas Two Step, you have four numbers drawn from a field of 35 and a bonus ball drawn from a separate field of 35. If you hash this into a number with each draw as a base 35 digit, you would wind up with certain numbers being excluded so you would have to hash it such that the first draw is a base 35 digit, the second a base 34 digit, the third a base 33 digit and the fourth a base 32 digit, the bonus ball would remain a base 35 digit, once this is converted to a standard numbering system such as base 10, you could then attempt to fit it to a function whether by fourier transforms to detect cyclic natures or wavelet or fractal decomposition. Regardless, it seems to me that it's critical to reduce the numbers drawn into a single value before proceeding with any reasonable analysis, even a simple binning of the numbers drawn into the 35 possibilities is flawed because the second number drawn is constrained to be exclusive of the first, the third exclusive of the first and second, and the fourth exclusive of the first second and third together with the fact that it's unknown which number was drawn in what order.

Mind you, trying to hash it by saying the first number is a base 35 digit, the second a base 34 etc still requires the order to be known which it isn't so some other way of representing the draw as a single number must be found. Perhaps if say the Texas Two Step example was D1, D2, D3, D4, B then maybe we could hash it into a single value by using D1 as a base(35-3) digit, D2 as a base(35-2-D1) digit, D3 as a base(35-1-D2) digit, D4 as a base(35-D3) digit and B as a base35 digit. Then there's the question of order of significance again complicated by the unknown order of draw.

Simply treating each number equally ignores the only known constraint on the numbers being drawn and it seems that would be the wrong way to go in a predictive system. Even a cyclic checksum would serve better to develop a match to a function or sequence.

Quote: Originally posted by jwhou on December 29, 2009

Wouldn't attempting to associate the numbers drawn with any kind of number sequence require the false assumption that the order of the numbers drawn is known and that each number drawn is independent of the others. With a typical pick 5/6 game, 5/6 numbers would be drawn from the same lot and then published in sorted order so the draw order within each group of 5/6 numbers is unknown and of course each number within each set of 5/6 is exclusive of the other numbers i.e.: there would be no repeats within each grouping of 5/6 numbers hence you can't really fit any kind of sequence to the individual numbers. Wouldn't it be necessary to reduce the 5 or 6 numbers drawn into a single unique number before attempting to fit it to a function or sequence?

I guess one way would be to treat each draw as a separate digit of a value. With the Texas Two Step, you have four numbers drawn from a field of 35 and a bonus ball drawn from a separate field of 35. If you hash this into a number with each draw as a base 35 digit, you would wind up with certain numbers being excluded so you would have to hash it such that the first draw is a base 35 digit, the second a base 34 digit, the third a base 33 digit and the fourth a base 32 digit, the bonus ball would remain a base 35 digit, once this is converted to a standard numbering system such as base 10, you could then attempt to fit it to a function whether by fourier transforms to detect cyclic natures or wavelet or fractal decomposition. Regardless, it seems to me that it's critical to reduce the numbers drawn into a single value before proceeding with any reasonable analysis, even a simple binning of the numbers drawn into the 35 possibilities is flawed because the second number drawn is constrained to be exclusive of the first, the third exclusive of the first and second, and the fourth exclusive of the first second and third together with the fact that it's unknown which number was drawn in what order.

Mind you, trying to hash it by saying the first number is a base 35 digit, the second a base 34 etc still requires the order to be known which it isn't so some other way of representing the draw as a single number must be found. Perhaps if say the Texas Two Step example was D1, D2, D3, D4, B then maybe we could hash it into a single value by using D1 as a base(35-3) digit, D2 as a base(35-2-D1) digit, D3 as a base(35-1-D2) digit, D4 as a base(35-D3) digit and B as a base35 digit. Then there's the question of order of significance again complicated by the unknown order of draw.

Simply treating each number equally ignores the only known constraint on the numbers being drawn and it seems that would be the wrong way to go in a predictive system. Even a cyclic checksum would serve better to develop a match to a function or sequence.

1.)Wouldn't attempting to associate the numbers drawn with any kind of number sequence require the false assumption that the order of the numbers drawn is known and that each number drawn is independent of the others.

I believe curve fitting does require sets of numbers in sequences—even if there supposed to be independently drawn.

2.) Wouldn’t it be necessary to reduce the 5 or 6 numbers drawn into a single unique number before attempting to fit it to a function or sequence?

NO

Numbers can be reduced to sets—patterns can then be developed—by there Transference.

3.) Simply treating each number equally ignores the only known constraint on the numbers being drawn and it seems that would be the wrong way to go in a predictive system.

Most RNG’s are uniform by nature. Shuffling does not have to be uniform.

HAWK

*We may see something that isn’t there because of what we expect to see

Or conversely, we may not see something because we don’t expect to see it.*

I have been working on using the logistic map or other maps to be an RNG for about a year now.

I greatly feel they might somehow be used with past lottery numbers to choose winning numbers.

I would really like to see other people ideas-code or whatever on the topic. Sometimes other peoples view helps.

Jadelottery's pick 3 and pick 4 program and information has been giving me ideas, along with the logistic map for another type of program.

Any help or ideas you come up with may be even better then mine. I have been wondering if anything mechanically loaded off prior genuine winning numbers could choose the next.

I believe that to hit the gold number or win the lotto we don't need all that math, I have notice that method like the Robert from lotto logix give me better results than a complicated math notation and so on it is about distribution and when somebody discovery really something, just keep it, because all we know that we are tracking and wathever we try to do, somebody also try to put down, remember we are talking about millions no cents, so is better to find another way to comunicate something about discoveries.

Quote: Originally posted by montecarlo2012 on February 7, 2010

I believe that to hit the gold number or win the lotto we don't need all that math, I have notice that method like the Robert from lotto logix give me better results than a complicated math notation and so on it is about distribution and when somebody discovery really something, just keep it, because all we know that we are tracking and wathever we try to do, somebody also try to put down, remember we are talking about millions no cents, so is better to find another way to comunicate something about discoveries.

I believe mathmatics and number theory is the only way towards winning numbers!

Its all in how and what you use to make final decisions.

While somethings may come easy -most things with the lottery start out to be complicated.

HAWK

*We may see something that isn’t there because of what we expect to see

Or conversely, we may not see something because we don’t expect to see it.*

I have been thinking --wouldn't a jackpot random picker be better if you already knew what category to play "like 2 odd 3even" because a file of just those numbers would much smaller then using a file of all possible combinations.

I don't believe it would effect the randomness of set selection by RNG.

Still odds will be high but much smaller then original.

HAWK

*We may see something that isn’t there because of what we expect to see

Or conversely, we may not see something because we don’t expect to see it.*

I still believe that a structured RNG for use with the jackpot games can be useful.

Sometimes I know that were going to get 3even 2odd with an odd number in the center.

5, 8, 17, 18, 52, The odd center number here is 17. I 'am working on ways to filter down the numbers of combinations from which to limit an RNG to choose from.

HAWK

*We may see something that isn’t there because of what we expect to see

Or conversely, we may not see something because we don’t expect to see it.*

I still have a few more ideas to post in this thread ---however I don't know exactly where they exist on the web. So when I know where they exist I'll post them later.

When all you see of the drawing is a cartoon, how can you be sure that the drawing was fair?

By this I mean, could not the lotto people ask their mainframe to produce 10 combinations which would result in the least number of second and third tier prizes?

Would manipulation in any manner influence negtively any data you might collect?

Could they (the Lotto Czars) not pre-determine when they will have a jackpot winner? Could they also not pre-determine who will win?

Why do you think some states went to RNG's? Do you believe the reason was to save money?

Would not manipulation in any manner distort "What happen in the past will happen in the future"?

Some things don't change even when the matrix change. I usually allow three consecutive numbers if the combination fit the profile I'm looking for because even if the winning combination doesn't have all three, it may have two of them and I would hate to miss out on a 4of5 because I rejected a triple.

that's brilliant, expect the "unexpected"

ab actu ad posse valet illatio - from the past one can infer the future

Quote: Originally posted by GASMETERGUY on June 9, 2010

pick4hawk

When all you see of the drawing is a cartoon, how can you be sure that the drawing was fair?

By this I mean, could not the lotto people ask their mainframe to produce 10 combinations which would result in the least number of second and third tier prizes?

Would manipulation in any manner influence negtively any data you might collect?

Could they (the Lotto Czars) not pre-determine when they will have a jackpot winner? Could they also not pre-determine who will win?

Why do you think some states went to RNG's? Do you believe the reason was to save money?

Would not manipulation in any manner distort "What happen in the past will happen in the future"?

Gasmeterguy:

When all you see is a cartoon --how do you know the drawing is fair?

Well they are supposed to test the output so that it passes all statisical tests --the the input is supposed to be unbiased however some states I really wonder if it is so.

I do belive that manipulation occurs in some states.

As far as the past effecting the future --that would depend on whether they are using a crypto routine.

Sorry for the long time to reply--I've been busy with a flood problems.

I still have a few more things to add to this thread--hopefully soon.

HAWK

*We may see something that isn’t there because of what we expect to see

Or conversely, we may not see something because we don’t expect to see it.*