bgonÃ§alves Brasil Member #92564 June 9, 2010 2124 Posts Offline

Posted: August 8, 2011, 10:32 am - IP Logged

hello jking e rl=

"the interesting thing is to observe that this percentage that does not usually occur and also occurs also has a standard and this standard is even more fixed than the default that is known to occur more often."

It is obvious no?What else happens in the lottery is if Miss the result. Then the standard mistake is much larger than the standard set.

United States Member #5599 July 13, 2004 1185 Posts Offline

Posted: August 8, 2011, 10:38 am - IP Logged

Quote: Originally posted by dr san on August 8, 2011

hello jking e rl=

"the interesting thing is to observe that this percentage that does not usually occur and also occurs also has a standard and this standard is even more fixed than the default that is known to occur more often."

It is obvious no?What else happens in the lottery is if Miss the result. Then the standard mistake is much larger than the standard set.

Hi,

Do you have a sample/example of what you are talking about?

You are a slave to the choices you have made. jk

Even a blind squirrel will occasioanlly find an acorn.

bgonÃ§alves Brasil Member #92564 June 9, 2010 2124 Posts Offline

Posted: August 8, 2011, 10:43 am - IP Logged

jking=

recourse is the stochastic processes as a way to treat quantitatively these phenomena, taking advantage of certain features of regularity that they present to be described by probabilistic models.

You can define a stochastic process as a set of random variables indexed to a variable (usually the variable time), being represented by {X (t), t}.Establishing the parallelism with the deterministic case, where a function f (t) takes well-defined values over time, a stochastic process takes random values over time.The values X (t) can assume are called States and to the whole state space X.

Exclusive analysis of past results do not allows the determination of a matrix of transition probabilities of States.

The fractal geometry, created by mathematician Benoit Mandelbrot, provides us with the idea of how to analyze chaotic phenomena: Fractals.In our case, a fractal's contain a fixed number of results of previous contests to contest that is being analyzed, since that is not so great of shape that contains the whole universe of dozens of modality, nor so small that the searched occurrence has a very long period.(Example: each contest will be compared with the previous ten contests.)

This is done, we have a selective count of all occurrences recorded.

To analyze the selective count are worth in the theory of Markov Chains, assembling the array of State Transition Probabilities, and the vector of Initial Population.

The product of the vector Population by transition Matrix in provides two results quite interesting:

United States Member #5599 July 13, 2004 1185 Posts Offline

Posted: August 8, 2011, 11:39 am - IP Logged

Quote: Originally posted by dr san on August 8, 2011

jking=

recourse is the stochastic processes as a way to treat quantitatively these phenomena, taking advantage of certain features of regularity that they present to be described by probabilistic models.

You can define a stochastic process as a set of random variables indexed to a variable (usually the variable time), being represented by {X (t), t}.Establishing the parallelism with the deterministic case, where a function f (t) takes well-defined values over time, a stochastic process takes random values over time.The values X (t) can assume are called States and to the whole state space X.

Exclusive analysis of past results do not allows the determination of a matrix of transition probabilities of States.

The fractal geometry, created by mathematician Benoit Mandelbrot, provides us with the idea of how to analyze chaotic phenomena: Fractals.In our case, a fractal's contain a fixed number of results of previous contests to contest that is being analyzed, since that is not so great of shape that contains the whole universe of dozens of modality, nor so small that the searched occurrence has a very long period.(Example: each contest will be compared with the previous ten contests.)

This is done, we have a selective count of all occurrences recorded.

To analyze the selective count are worth in the theory of Markov Chains, assembling the array of State Transition Probabilities, and the vector of Initial Population.

The product of the vector Population by transition Matrix in provides two results quite interesting:

the Final Population vector);

(b)) hope the next result.

Hi,

Just a couple things to consider:

Can random have a state?

Are random and chaotic really comparable?

Fractals can be described with simple mathematical expressions, are you expecting that true randomness can be descibed in the same way.

Just my opinion, but I think there are too many what if's, what abouts, and assumptions made, which make this approach questionable.

Perhaps, some of the other members will feel more comfortable with this approach. *S*

You are a slave to the choices you have made. jk

Even a blind squirrel will occasioanlly find an acorn.

United States Member #5599 July 13, 2004 1185 Posts Offline

Posted: August 8, 2011, 12:41 pm - IP Logged

Hi dr san,

Please don't feel put off by my not buying in to your approach. Because nobody has posted anything that truely is a winning system, all approaches must be respected. After be a member of the LP for a while, I've seen systems and different approaches come and go. God knows I've tried enough of my own. Just keep in mind that the path you are treading, a pure statistical approach, has been taken before several times. And it's because of the lack luster results it has generated in the past that I shy away from taking the same path. Maybe you will be the one to find a solution, I don't know. I just know at this point and time that I am going to try to solve our common goal in a different way.

Thanks for comments, keep posting, and best of luck.

You are a slave to the choices you have made. jk

Even a blind squirrel will occasioanlly find an acorn.

mid-Ohio United States Member #9 March 24, 2001 19830 Posts Offline

Posted: August 8, 2011, 1:06 pm - IP Logged

I have a program where I look at 10000-15000 combinations for each drawing of which I consider 150-300 for my final selection of 20-30 lines.

Lately I've been keeping track of those 150-300 to see if I rejected any winners and I've found I usually reject a match4 and a few match3's every drawing and have rejected what would eventually be a match5 2-5 drawings back twice.

I've been thinking about saving all 10000 to 15000 randomly picked combinations for one drawing to see if I even had a chance of picking a jackpot winner. It could turn out that even though I generated up to 15000 combinations randomly, there were never anything better than a match4 in the group for that drawing and in that case I would need to come up with a better way of generating random combinations to pick from.

* you don't need to buy more tickets, just buy a winning ticket *

Las Vegas, NV United States Member #44841 August 9, 2006 1749 Posts Offline

Posted: August 8, 2011, 1:42 pm - IP Logged

Hi JKING,

Hope you don't mind me jumping in here, but now that you've established the top most hitting numbers in (CA Fan5), how would you go about playing them? Are you wheeling them with one or more "key" numbers?

United States Member #5599 July 13, 2004 1185 Posts Offline

Posted: August 8, 2011, 2:18 pm - IP Logged

Quote: Originally posted by Feisty1 on August 8, 2011

Hi JKING,

Hope you don't mind me jumping in here, but now that you've established the top most hitting numbers in (CA Fan5), how would you go about playing them? Are you wheeling them with one or more "key" numbers?

Hi,

Everyone is welcome to jump in. *S*

I've createdd a system that has approximately 20 filters with little to no error as the foundation. As a secondary filter I've included the 80/20 top hitting numbers. After running all the combinations I get a summary report that shows what numbers have peak values by position. I then very the numbers with the best counts against valid combinations (combinations that don't exceed the min/max conditions of my filters). Sorry, it wasn't an easy answer for you.

I just started testing the 80/20 rule with my system for the first time yesterday. Bet three line and got one 2/5 hit. I'll run it for a while and see what happens.

You are a slave to the choices you have made. jk

Even a blind squirrel will occasioanlly find an acorn.

Las Vegas, NV United States Member #44841 August 9, 2006 1749 Posts Offline

Posted: August 8, 2011, 2:25 pm - IP Logged

Quote: Originally posted by JKING on August 8, 2011

Hi,

Everyone is welcome to jump in. *S*

I've createdd a system that has approximately 20 filters with little to no error as the foundation. As a secondary filter I've included the 80/20 top hitting numbers. After running all the combinations I get a summary report that shows what numbers have peak values by position. I then very the numbers with the best counts against valid combinations (combinations that don't exceed the min/max conditions of my filters). Sorry, it wasn't an easy answer for you.

I just started testing the 80/20 rule with my system for the first time yesterday. Bet three line and got one 2/5 hit. I'll run it for a while and see what happens.

That's ok, I knew it wasn't going to be an easy fix with a pool of numbers that large ... but you were able to filter down to three lines?