Welcome Guest
Log In | Register )
You last visited December 3, 2016, 8:56 pm
All times shown are
Eastern Time (GMT-5:00)

Convergence

Topic closed. 11 replies. Last post 3 years ago by Lottonomics.

Page 1 of 1
51
PrintE-mailLink
RL-RANDOMLOGIC's avatar - usafce

United States
Member #59354
March 13, 2008
3962 Posts
Offline
Posted: June 26, 2013, 7:58 am - IP Logged

Hi all

I wanted to ask if anyone has tried this idea for P-5, P-6 games.  I suppose it would work for the bonus

games also with the exception of the bonus ball.  Working with filters over the years I have found a few

that converge every so often.  Let's say that I have 6 filters that are history based which when all set

to single values they produce 1 line.  These filters are calculated in such a way that the line generated

will be different game to game because each new draw added to the database changes the variables

for the next line generated.    Lets say that all the filter values converge to the there most logical values

a few times each year or around 2 or 3 times out of 350 games.   There are no user settings and all a

person has to do is keep the database updated.  One click generates one line.  The user has to play the

one line every game and expect to loose around 347 out of every 350 games played but on the bright

side it should produce 2 or 3 JP's a year.   Such a method could not be released to the masses because

the set generated would be the same for everyone playing that game.  If 50 people were using this to play

the same game then you will have to split the JP between 50 people which kind of defeats the purpose.

Each variable has a 1 in 10 chance of showing in the next draw but using the historical data to assign

probabilities to each and then select the most probable for each filter.  As you might think most days the values

that hit are all over the place some from the low end and some from the high end of the pool so to say.  We

play for the game where all the most favorable all converge on the the same line or game.  When they do

we have the winning set.   Such a system is very easy to back-test using real draws and I would think that

if it hits 2 or 3 times out of every 350 drawings going back a few years then it should also hit 2 or 3 times in

the next 350 games.  One time would be enough for most people but I am sure no one would refuse a second

or third.  It would require playing every day because we have no clue when the drawing will converge to the

most probable values all showing at the same time. 

RL

Working on my Ph.D.  "University of hard Knocks"

I will consider the opinion that my winnings are a product of chance if you are willing to consider

they are not.  Many great discoveries come while searching for something else

USAF https://en.wikipedia.org/wiki/Prime_Base_Engineer_Emergency_Force

  US Flag Trump / 2016 & 2020  

    WIN  D's avatar - q05Q0
    Stone Mountain*Georgia
    United States
    Member #828
    November 2, 2002
    10491 Posts
    Offline
    Posted: June 26, 2013, 1:01 pm - IP Logged

    Hi all

    I wanted to ask if anyone has tried this idea for P-5, P-6 games.  I suppose it would work for the bonus

    games also with the exception of the bonus ball.  Working with filters over the years I have found a few

    that converge every so often.  Let's say that I have 6 filters that are history based which when all set

    to single values they produce 1 line.  These filters are calculated in such a way that the line generated

    will be different game to game because each new draw added to the database changes the variables

    for the next line generated.    Lets say that all the filter values converge to the there most logical values

    a few times each year or around 2 or 3 times out of 350 games.   There are no user settings and all a

    person has to do is keep the database updated.  One click generates one line.  The user has to play the

    one line every game and expect to loose around 347 out of every 350 games played but on the bright

    side it should produce 2 or 3 JP's a year.   Such a method could not be released to the masses because

    the set generated would be the same for everyone playing that game.  If 50 people were using this to play

    the same game then you will have to split the JP between 50 people which kind of defeats the purpose.

    Each variable has a 1 in 10 chance of showing in the next draw but using the historical data to assign

    probabilities to each and then select the most probable for each filter.  As you might think most days the values

    that hit are all over the place some from the low end and some from the high end of the pool so to say.  We

    play for the game where all the most favorable all converge on the the same line or game.  When they do

    we have the winning set.   Such a system is very easy to back-test using real draws and I would think that

    if it hits 2 or 3 times out of every 350 drawings going back a few years then it should also hit 2 or 3 times in

    the next 350 games.  One time would be enough for most people but I am sure no one would refuse a second

    or third.  It would require playing every day because we have no clue when the drawing will converge to the

    most probable values all showing at the same time. 

    RL

    Hello RL

      I tried a search for a subject close to what your talking about that took place years ago on the Math forum ....with no luck.  Although it was not called convergence ...it certainly was taking place all around the subject.

       It had to do with waiting on those "Perfect Storm" conditions that take place just a few times a year on the pick 5 games.  ALL ODD or ALL EVEN  draw numbers were a big part of the needed conditions. 

       Perhaps Rjoh can remember the title or tag line. It was quite awhile back.    Maybe we said confluence instead of Convergence.  LOL

     

      The thrust of that subject was ...... a person playing on those particular days was better off ODDS WISE ......by like a factor of 5 ....or something along those lines.

       At the least on those days the odds are reduced if .....if you were waiting for those conditions and played that way.

      You would know you were closer ......and could prove it with the odds on those rare draw days

     

     

    The only real failure .....is the failure to try.                               

                                  Luck is a very rare thing....... Odds not so much. 

                                  Odds never change .....but probability does. 

                                                                                           Win d    

      jimjwright's avatar - Yellow 3.png
      Park City, UT
      United States
      Member #69864
      January 18, 2009
      993 Posts
      Offline
      Posted: June 26, 2013, 2:56 pm - IP Logged

      I have found something similar for the Pick5 games.  I have come up with 2 filters that will reduce a 5/39 game from 575757 combos to around 50.  These filters typically hit 1-2 times every 100 draws but are not really predictable when they will happen.  It's a complicated heuristic algorithm that factors in more than the previous draw.  So its more expensive than what you propose in converging to a single line to play.  I am in the process of coding the same heuristic for pick3 and pick4 in hoping I can back it off to just leaving 1% of the combos and hoping for a much higher hit rate.  Should know by end of week if I am chasing my tail.

      But to answer your question in general I think you can do what you propose but I don't know about a single line that would be impressive and something to shoot for.  Heres hoping you can accomplish it.

      Jimmy

        WIN  D's avatar - q05Q0
        Stone Mountain*Georgia
        United States
        Member #828
        November 2, 2002
        10491 Posts
        Offline
        Posted: June 26, 2013, 5:22 pm - IP Logged

        Here is a small part of the reference subject going pretty far back to 2002 .......wow eleven years ago. 

         http://www.lotterypost.com/thread/144598/710140?q=win+d&rp=search

         

         

        The only real failure .....is the failure to try.                               

                                      Luck is a very rare thing....... Odds not so much. 

                                      Odds never change .....but probability does. 

                                                                                               Win d    

          RL-RANDOMLOGIC's avatar - usafce

          United States
          Member #59354
          March 13, 2008
          3962 Posts
          Offline
          Posted: June 26, 2013, 5:37 pm - IP Logged

          JW

          To better understand let's take a look at the lexicographic index for a 5-39.  We have 000001 to 575757.

          Much like a pick-3 only with 6 digits.  A simple way to explain the process is to break up the six digits

          and analyze each position using my special formula, lol which calculates a probability for each digit in each

          position.  This is not your normal probabilities.

           

          For my  5-39 this gives me six possible choices for the first digit (0 to 5) and (0 to 9) for the remaining

          five positions.  The program selects one digit for each position that the formula says has the best possible

          chance of hitting in the next game.  Once the program has all 6 digits it converts them to a 5 number set.

           

           

          The program uses many different variables including the draw history to make the calculations so with every

          update the output changes.  Take the lexi 414045 = 09-12-17-24-37 but if I change one digit, let's say the digit

          4 to a 3 in the first position I get 314054 = 06-10-23-28-38. 

           

          The same set of probabilities can produce more than one line but not within the same game.  Think of it as trying

          to hit the probability instead of the value.  The probabilities selected fall within a small window but the digits they

          represent change game to game.  If all 6 lexi digits fall within the formulas predicted range then the lexi produced

          will be correct. 

           

          I am still running test against real data as it would have been at the time of the actual drawing and so far it looks

          good but I have been down that road before.   I will try to post some results as I am testing it against several 

          different games both ball droppers and RNG's using the same formula to calculate the probabilities for each digit.     

           

          RL

          Working on my Ph.D.  "University of hard Knocks"

          I will consider the opinion that my winnings are a product of chance if you are willing to consider

          they are not.  Many great discoveries come while searching for something else

          USAF https://en.wikipedia.org/wiki/Prime_Base_Engineer_Emergency_Force

            US Flag Trump / 2016 & 2020  

            WIN  D's avatar - q05Q0
            Stone Mountain*Georgia
            United States
            Member #828
            November 2, 2002
            10491 Posts
            Offline
            Posted: June 26, 2013, 5:50 pm - IP Logged

            This was a post that was fun and a blast from the past. At that point I'm not even sure I had actually played Pick 5 before. 

             Years ago we were playing with Pick 5 game angles and awkwardly poking around with ....wait and shoot strategies.  There might be one very salient point in there ...somewhere.....if any. LOL 

             http://www.lotterypost.com/thread/179047

             

             

            The only real failure .....is the failure to try.                               

                                          Luck is a very rare thing....... Odds not so much. 

                                          Odds never change .....but probability does. 

                                                                                                   Win d    

              jimjwright's avatar - Yellow 3.png
              Park City, UT
              United States
              Member #69864
              January 18, 2009
              993 Posts
              Offline
              Posted: June 26, 2013, 6:44 pm - IP Logged

              JW

              To better understand let's take a look at the lexicographic index for a 5-39.  We have 000001 to 575757.

              Much like a pick-3 only with 6 digits.  A simple way to explain the process is to break up the six digits

              and analyze each position using my special formula, lol which calculates a probability for each digit in each

              position.  This is not your normal probabilities.

               

              For my  5-39 this gives me six possible choices for the first digit (0 to 5) and (0 to 9) for the remaining

              five positions.  The program selects one digit for each position that the formula says has the best possible

              chance of hitting in the next game.  Once the program has all 6 digits it converts them to a 5 number set.

               

               

              The program uses many different variables including the draw history to make the calculations so with every

              update the output changes.  Take the lexi 414045 = 09-12-17-24-37 but if I change one digit, let's say the digit

              4 to a 3 in the first position I get 314054 = 06-10-23-28-38. 

               

              The same set of probabilities can produce more than one line but not within the same game.  Think of it as trying

              to hit the probability instead of the value.  The probabilities selected fall within a small window but the digits they

              represent change game to game.  If all 6 lexi digits fall within the formulas predicted range then the lexi produced

              will be correct. 

               

              I am still running test against real data as it would have been at the time of the actual drawing and so far it looks

              good but I have been down that road before.   I will try to post some results as I am testing it against several 

              different games both ball droppers and RNG's using the same formula to calculate the probabilities for each digit.     

               

              RL

              I understand, with the Lexi research you have done and are doing it will always reduce to 1 line.  My approach will never reduce to 1 line.  So yours is the superior approach if you can accomplish it.  I'm rooting for you !!!

              Jimmy

                RL-RANDOMLOGIC's avatar - usafce

                United States
                Member #59354
                March 13, 2008
                3962 Posts
                Offline
                Posted: June 26, 2013, 7:20 pm - IP Logged

                JW

                It all comes down to rather a back-test can be trusted. Most of the time I have found that it not a good

                means to test something like this.  The past hits could be just chance and the program might not hit again

                for a couple years.  Most people don't have that kind of commitment to something that might not work.  The

                program only takes a couple seconds to run but filling out the ticket driving to the lottery store and back

                would get old after a few hundred trips, that is unless it hits.  I need to test the lower prize wins to see

                what the average day to day play returns.  Superior? Only time and comparison can make that determination.

                RL

                Working on my Ph.D.  "University of hard Knocks"

                I will consider the opinion that my winnings are a product of chance if you are willing to consider

                they are not.  Many great discoveries come while searching for something else

                USAF https://en.wikipedia.org/wiki/Prime_Base_Engineer_Emergency_Force

                  US Flag Trump / 2016 & 2020  

                  jimjwright's avatar - Yellow 3.png
                  Park City, UT
                  United States
                  Member #69864
                  January 18, 2009
                  993 Posts
                  Offline
                  Posted: June 26, 2013, 8:36 pm - IP Logged

                  JW

                  It all comes down to rather a back-test can be trusted. Most of the time I have found that it not a good

                  means to test something like this.  The past hits could be just chance and the program might not hit again

                  for a couple years.  Most people don't have that kind of commitment to something that might not work.  The

                  program only takes a couple seconds to run but filling out the ticket driving to the lottery store and back

                  would get old after a few hundred trips, that is unless it hits.  I need to test the lower prize wins to see

                  what the average day to day play returns.  Superior? Only time and comparison can make that determination.

                  RL

                  I have dabbled with the approach that say you arranged the numbers in some predetermined random order as a string.  Then after each draw you pull the numbers out of string and re-insert them back at various pre-determined points into the string.  The hopes that after doing this pseudo-random shuffling for say n draws you push the winning numbers to either one end of the string or the other end or maybe to the middle where you would play those numbers for the next draw.  Then hopefully this pseudo-random shuffling would result in additional hits in the future by following the same procedure.

                  I coded something and let in crank for a week in background trying various random sequences to seed the string and never found anything useful.  The problem being for smaller cash 5 games most draws have a number with a low skip count so you needed to re-insert the numbers in such a way to account for this.

                  Jimmy

                    Lottonomics's avatar - box

                    United States
                    Member #133657
                    October 5, 2012
                    82 Posts
                    Offline
                    Posted: June 26, 2013, 10:16 pm - IP Logged

                    JW

                    It all comes down to rather a back-test can be trusted. Most of the time I have found that it not a good

                    means to test something like this.  The past hits could be just chance and the program might not hit again

                    for a couple years.  Most people don't have that kind of commitment to something that might not work.  The

                    program only takes a couple seconds to run but filling out the ticket driving to the lottery store and back

                    would get old after a few hundred trips, that is unless it hits.  I need to test the lower prize wins to see

                    what the average day to day play returns.  Superior? Only time and comparison can make that determination.

                    RL

                    It sounds like a good idea if there are enough consistent results to back it up!

                    You said "Most people don't have that kind of commitment to something that might not work." To me it would be similar to the person that plays the same favorite combination every draw for years and years hoping that "today will be the day" without ever really knowing IF it'll ever hit in their lifetime.

                    With your idea it wouldn't be a matter of IF, it'd be just a matter of WHEN. I like it!

                      RL-RANDOMLOGIC's avatar - usafce

                      United States
                      Member #59354
                      March 13, 2008
                      3962 Posts
                      Offline
                      Posted: June 26, 2013, 10:58 pm - IP Logged

                      LN

                      The advantage of playing the same set would be that you could buy up to 14 plays at a time, at least in MO

                      they give you that option so you would only need to buy a ticket once every 2 weeks.  The above would require

                      you to purchase a ticket every day.   I would hate to play every day for 6 months and the one day I skipped the

                      numbers come in. 

                      RL

                      Working on my Ph.D.  "University of hard Knocks"

                      I will consider the opinion that my winnings are a product of chance if you are willing to consider

                      they are not.  Many great discoveries come while searching for something else

                      USAF https://en.wikipedia.org/wiki/Prime_Base_Engineer_Emergency_Force

                        US Flag Trump / 2016 & 2020  

                        Lottonomics's avatar - box

                        United States
                        Member #133657
                        October 5, 2012
                        82 Posts
                        Offline
                        Posted: June 26, 2013, 11:21 pm - IP Logged

                        LN

                        The advantage of playing the same set would be that you could buy up to 14 plays at a time, at least in MO

                        they give you that option so you would only need to buy a ticket once every 2 weeks.  The above would require

                        you to purchase a ticket every day.   I would hate to play every day for 6 months and the one day I skipped the

                        numbers come in. 

                        RL

                        Ahh, I didn't consider the advanced play vs having to fill out a different line every day. I guess I was in the mega millions mindset where it would only be 2 draws a week, doh! Although, if someone had a daily routine of always having to go somewhere that happens to sell lottery tickets I suppose it wouldn't seem like such a chore, haha.

                        The idea still has some promise though!