Welcome Guest
Log In | Register )
You last visited December 3, 2016, 10:50 pm
All times shown are
Eastern Time (GMT-5:00)

Interest In Backtesting and Simulating Lottery Systems

Topic closed. 99 replies. Last post 5 years ago by lotterybraker.

Page 5 of 7
53
PrintE-mailLink

What are the Merits of Backtesting and Simulating Lottery Systems?

I think it would provide valuable insights. [ 41 ]  [64.06%]
I think it would be a waste of time. [ 6 ]  [9.38%]
I Don't Know but I would like to learn about it. [ 13 ]  [20.31%]
I don't know and I don't care. [ 4 ]  [6.25%]
Total Valid Votes [ 64 ]  
Discarded Votes [ 1 ]  
Avatar
Kentucky
United States
Member #32652
February 14, 2006
7295 Posts
Offline
Posted: May 6, 2011, 10:18 pm - IP Logged

If one observed certain relations between present drawings and previous drawings, how will back testing disprove what they observed? 

Observations such as half the time a drawing in a 5/39 game will have at least one number from the previous drawing is supported by calculating the odds and those of all 5 winning numbers hitting in the previous 15 drawings at least half the time even though only 70% of the numbers may be covered are more likely to be proved by back testing.

The only things back testing will disprove are things we already know aren't true.

You could back test 12 numbers in a 5/39 game and find out they won several jackpots, but that doesn't prove or disprove they will win another jackpot in the near future. Or you could back test the same 12 numbers using a simulated lottery and find they never won a jackpot and that would neither prove or disprove those numbers will win a jackpot in the near future.

If a system using 1 of 5 numbers repeating at a 50% rate was based on past drawing statistics, back testing only proves what is already know. Maybe someone can explain a useful purpose for back testing a system against a simulation using the same number of drawings because those results have no baring on the actual drawing results and what the system is based on.

    RJOh's avatar - chipmunk
    mid-Ohio
    United States
    Member #9
    March 24, 2001
    19817 Posts
    Online
    Posted: May 6, 2011, 10:50 pm - IP Logged

    You could back test 12 numbers in a 5/39 game and find out they won several jackpots, but that doesn't prove or disprove they will win another jackpot in the near future. Or you could back test the same 12 numbers using a simulated lottery and find they never won a jackpot and that would neither prove or disprove those numbers will win a jackpot in the near future.

    If a system using 1 of 5 numbers repeating at a 50% rate was based on past drawing statistics, back testing only proves what is already know. Maybe someone can explain a useful purpose for back testing a system against a simulation using the same number of drawings because those results have no baring on the actual drawing results and what the system is based on.

    It's fairly easy to find 12 numbers that have won several jackpots but what does it prove?
    For example in Ohio Rolling Cash5 (5/39):

    03 04 07 10 11 15 16 21 23 30 35 38

     07/03/09 - 11 16 21 30 35
     04/18/09 - 03 10 16 35 38
     03/30/08 - 03 07 15 23 35
     02/04/08 - 04 15 21 35 38

    also:

     MATCH 0 = 277 
     MATCH 1 = 766 
     MATCH 2 = 788   $ 788
     MATCH 3 = 370   $ 3700
     MATCH 4 = 62     $ 18600
     MATCH 5 = 4       4 jackpots

    1220 prizes = $ 23088 + 4 JP

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

      Avatar
      Kentucky
      United States
      Member #32652
      February 14, 2006
      7295 Posts
      Offline
      Posted: May 6, 2011, 10:55 pm - IP Logged

      I didn't say I believed, I said I observed.

      Using a lottery program I can check any/all winning numbers with any amount of previous drawings. I checked 2251 of 2266 drawings of Ohio Rolling Cash5 and found the following number of times the 5 winning numbers had been in the previous 15 drawings at least 1 time:

      5/5 = 1135
      4/5 = 828
      3/5 = 256
      2/5 = 30
      1/5 = 1
      number pool size of 15 drawings = 28-29  (28/39 = 70%  1135/2251 = 50%)

      When I checked the previous drawing of 2265 of 2266 drawings, I got the following:

      MATCH  0 = 1084
      MATCH  1 = 930
      MATCH  2 = 218
      MATCH  3 = 32
      MATCH  4 = 1
      MATCH  5 = 0
      (1181/2265 = 52%)

      The above are statistics/observations not beliefs.

      Note: these figures only apply to Ohio Rolling Cash5 (5/39) and I can't say if knowing only these statistics have any advantage because they're probably normal for all 5/39 games  but with other statistics I believe one may be able to develop profiles of combinations that are more likely to win.  Only by winning more than normal would one know for sure.  All any back testing would show is if any past winning numbers had one the profiles used but we already that if we used that data to develop them.
       
      *that match 4 in a previous drawing happened back in '04.
       12/14/04 - 03 11 20 29 35
       12/13/04 - 11 20 29 35 37

      Any 5/39 game has 291,720 (50.67%) combinations with 1 or 2 numbers repeating. Back testing RC5 only proves what we already know and unless there is a flaw in the simulated lottery, the results would be about the same.

      "All any back testing would show is if any past winning numbers had one the profiles used but we already that if we used that data to develop them."

      I Agree! and checking the probability is faster than setting up a simulation too.


        United States
        Member #93947
        July 10, 2010
        2180 Posts
        Offline
        Posted: May 7, 2011, 12:08 am - IP Logged

        Any 5/39 game has 291,720 (50.67%) combinations with 1 or 2 numbers repeating. Back testing RC5 only proves what we already know and unless there is a flaw in the simulated lottery, the results would be about the same.

        "All any back testing would show is if any past winning numbers had one the profiles used but we already that if we used that data to develop them."

        I Agree! and checking the probability is faster than setting up a simulation too.

        It seems that you and RJOh are basically saying you don't believe there is anything to be gained from applying backtest or simulation techniques to proposed lottery selection systems.  Since you've declared often that the stats published by the lotteries for each game only tell you what you already know, that they will pay back 50% of the gross over the long haul, you seem to have given up any hope of determining in advance what your chances of succeeding with a system are.

        Before giving up, I really hope you will be willing to SERIOUSLY [re]study the information and links in this Posting:

        http://www.lotterypost.com/thread/229947

        --Jimmy4164

          Avatar
          Kentucky
          United States
          Member #32652
          February 14, 2006
          7295 Posts
          Offline
          Posted: May 7, 2011, 1:42 pm - IP Logged

          It seems that you and RJOh are basically saying you don't believe there is anything to be gained from applying backtest or simulation techniques to proposed lottery selection systems.  Since you've declared often that the stats published by the lotteries for each game only tell you what you already know, that they will pay back 50% of the gross over the long haul, you seem to have given up any hope of determining in advance what your chances of succeeding with a system are.

          Before giving up, I really hope you will be willing to SERIOUSLY [re]study the information and links in this Posting:

          http://www.lotterypost.com/thread/229947

          --Jimmy4164

          Is it impossible for you to answer a simple question?

          What is the value of back testing a system based on past performance against a simulated lottery?


            United States
            Member #93947
            July 10, 2010
            2180 Posts
            Offline
            Posted: May 7, 2011, 3:00 pm - IP Logged

            Is it impossible for you to answer a simple question?

            What is the value of back testing a system based on past performance against a simulated lottery?

            "What is the value of back testing a system based on past performance against a simulated lottery?"

            If you would take my advice and read the Wikipedia and MathWorks articles you might not have a need to ask a question.  At least, if you did, it might be a question that makes logical sense, and consequently, be answerable.

            I really hope you will SERIOUSLY study the information and links in this Posting:

            http://www.lotterypost.com/thread/229947

            Until you do, further discussion will serve no useful purpose.

            --Jimmy4164

            P.S.  Hint:  A Backtest does not require a Monte Carlo simulation component.


              RJOh's avatar - chipmunk
              mid-Ohio
              United States
              Member #9
              March 24, 2001
              19817 Posts
              Online
              Posted: May 7, 2011, 5:01 pm - IP Logged

              Is it impossible for you to answer a simple question?

              What is the value of back testing a system based on past performance against a simulated lottery?

              You are likely to know as much about creating a winning system as any other LP members including Jimmy so asking him or any other member a question about the value of back testing such a system using a simulated lottery is a waste of time.  If he had an answer he would had probably already posted it.  Besides, every thing Jimmy knows he has documented with a link to something he read so it's unlikely he've given much thought to the subject.

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

                Avatar
                Kentucky
                United States
                Member #32652
                February 14, 2006
                7295 Posts
                Offline
                Posted: May 7, 2011, 5:03 pm - IP Logged

                "What is the value of back testing a system based on past performance against a simulated lottery?"

                If you would take my advice and read the Wikipedia and MathWorks articles you might not have a need to ask a question.  At least, if you did, it might be a question that makes logical sense, and consequently, be answerable.

                I really hope you will SERIOUSLY study the information and links in this Posting:

                http://www.lotterypost.com/thread/229947

                Until you do, further discussion will serve no useful purpose.

                --Jimmy4164

                P.S.  Hint:  A Backtest does not require a Monte Carlo simulation component.


                "If you would take my advice and read"

                Someone's opinion who isn't a member of LP?

                "P.S.  Hint:  A Backtest does not require a Monte Carlo simulation component."

                Based on observation, RJ back tested RC5 and found that in slightly over 50% of drawings 1 - 2 numbers repeated and I did the math showing the probability that slightly over 50% of all the possible combinations have 1 - 2 repeating numbers. From that innocent exchange you said:

                "It seems that you and RJOh are basically saying you don't believe there is anything to be gained from applying backtest or simulation techniques to proposed lottery selection systems."

                Actually I said the probability is that slightly over 50% of the combinations have 1 - 2 repeat numbers and RJ proved in slightly over 50% drawings, 1 - 2 numbers DID repeat. We did the math and I asked you a simple question, but it seems you can only offer links as an answer.

                But you did offer an example of how your simulation works by showing "Of the 50,000 "People" who played for 13.7 years...". That's very valuable information for pick-3 QP players (all 12 of them), but useless information for "Backtesting Simulating LOTTERY SYSTEMS".

                You should my advice read this link:

                http://dictionary.reference.com/browse/backtesting

                  Avatar
                  Kentucky
                  United States
                  Member #32652
                  February 14, 2006
                  7295 Posts
                  Offline
                  Posted: May 7, 2011, 6:24 pm - IP Logged

                  You are likely to know as much about creating a winning system as any other LP members including Jimmy so asking him or any other member a question about the value of back testing such a system using a simulated lottery is a waste of time.  If he had an answer he would had probably already posted it.  Besides, every thing Jimmy knows he has documented with a link to something he read so it's unlikely he've given much thought to the subject.

                  I have nothing against simulations because they are valuable in testing betting strategies. But I see nothing valuable about testing any strategy with short term goals over millions of trials. If I had a pick-5 system that consistently matched 2 o 3 numbers with an occasional 4 or 5 number match, 21 or 28 trials is sufficient.

                  "Besides, every thing Jimmy knows he has documented with a link to something he read so it's unlikely he've given much thought to the subject."

                  What Jimmy showed so far is how 50,000 random bets would do over 1000 trials as if a "random bet" is the same as a system. At least everybody can agree he sure knows how to google and post links.


                    United States
                    Member #93947
                    July 10, 2010
                    2180 Posts
                    Offline
                    Posted: May 7, 2011, 6:36 pm - IP Logged

                    "If you would take my advice and read"

                    Someone's opinion who isn't a member of LP?

                    "P.S.  Hint:  A Backtest does not require a Monte Carlo simulation component."

                    Based on observation, RJ back tested RC5 and found that in slightly over 50% of drawings 1 - 2 numbers repeated and I did the math showing the probability that slightly over 50% of all the possible combinations have 1 - 2 repeating numbers. From that innocent exchange you said:

                    "It seems that you and RJOh are basically saying you don't believe there is anything to be gained from applying backtest or simulation techniques to proposed lottery selection systems."

                    Actually I said the probability is that slightly over 50% of the combinations have 1 - 2 repeat numbers and RJ proved in slightly over 50% drawings, 1 - 2 numbers DID repeat. We did the math and I asked you a simple question, but it seems you can only offer links as an answer.

                    But you did offer an example of how your simulation works by showing "Of the 50,000 "People" who played for 13.7 years...". That's very valuable information for pick-3 QP players (all 12 of them), but useless information for "Backtesting Simulating LOTTERY SYSTEMS".

                    You should my advice read this link:

                    http://dictionary.reference.com/browse/backtesting

                    I said, "If you would take my advice and read..."

                    ...to which you added, "Someone's opinion who isn't a member of LP?"

                    Unbelievable. When you're unwilling to admit you don't understand what the links point to, wouldn't it make more sense to just quietly ignore the thread rather than keep spewing out evidence that you don't get it?

                    --Jimmy4164

                    P.S.  What is your goal posting in this thread?


                      Avatar
                      Kentucky
                      United States
                      Member #32652
                      February 14, 2006
                      7295 Posts
                      Offline
                      Posted: May 7, 2011, 10:06 pm - IP Logged

                      I said, "If you would take my advice and read..."

                      ...to which you added, "Someone's opinion who isn't a member of LP?"

                      Unbelievable. When you're unwilling to admit you don't understand what the links point to, wouldn't it make more sense to just quietly ignore the thread rather than keep spewing out evidence that you don't get it?

                      --Jimmy4164

                      P.S.  What is your goal posting in this thread?


                      "P.S. What is your goal posting in this thread?"

                      Did you forget you asked "What are the Merits of Backtesting and Simulating Lottery Systems?"

                      I flipped a coin between "I think it would be a waste of time." and "discard vote" just on a slim chance you would actually follow through on something you started. "Read this, read that, study this, study that" isn't discussion, but that's exactly how you answered when I asked the merits of whatever it is you are trying to prove or disprove.

                      You started the thread and disappointed the 12 people (I think it would provide valuable insights.) who already knew 50,000 imaginary QP pick-3 players are not system players.

                      Stick to what you know best; advising the Challenge players not to spend $3168 four times a week playing MM and PB.


                        United States
                        Member #93947
                        July 10, 2010
                        2180 Posts
                        Offline
                        Posted: May 7, 2011, 11:50 pm - IP Logged

                        A recurring theme expressed by several people here when discussing my postings is the fact that I make a lot of references to other people's work.  Their conclusion is that since I make these references, I must not have any knowledge of the subjects myself.  Just today, RJOh said, "Besides, every thing Jimmy knows he has documented with a link to something he read so it's unlikely he've given much thought to the subject."  I can't help but wonder how they learn new things themselves, if not through reading.  What I am forced to conclude are two things:

                        1) It's unlikely they have read many scholarly articles or journals because if they had they would know that source citations are not only expected, but often required.  (See examples at Wikipedia.)

                        2) They apparently are more impressed with people who "reinvent the wheel" rather than those who rely on the proven and accepted works of their peers and predecessors.

                        --Jimmy4164

                          RJOh's avatar - chipmunk
                          mid-Ohio
                          United States
                          Member #9
                          March 24, 2001
                          19817 Posts
                          Online
                          Posted: May 8, 2011, 1:50 am - IP Logged

                          A recurring theme expressed by several people here when discussing my postings is the fact that I make a lot of references to other people's work.  Their conclusion is that since I make these references, I must not have any knowledge of the subjects myself.  Just today, RJOh said, "Besides, every thing Jimmy knows he has documented with a link to something he read so it's unlikely he've given much thought to the subject."  I can't help but wonder how they learn new things themselves, if not through reading.  What I am forced to conclude are two things:

                          1) It's unlikely they have read many scholarly articles or journals because if they had they would know that source citations are not only expected, but often required.  (See examples at Wikipedia.)

                          2) They apparently are more impressed with people who "reinvent the wheel" rather than those who rely on the proven and accepted works of their peers and predecessors.

                          --Jimmy4164

                          I can't help but wonder how they learn new things themselves, if not through reading. 

                          As far as I know no system exists that can predict the outcome of a lottery drawing better than a random quick pick and all the articles I've ever read say the same thing so reading them again isn't going reveal anything new about the theoretical systems we are discussing.

                          It's unlikely they have read many scholarly articles or journals because if they had they would know that source citations are not only expected, but often required.

                          What's suppose to be scholarly about picking some numbers to play in a lottery at this point?

                          They apparently are more impressed with people who "reinvent the wheel" rather than those who rely on the proven and accepted works of their peers and predecessors.

                          The wheel have been reinvented many times in the last hundred years.  Where do you think the US auto industry would be today if they were still manufacturing wooden wheels.

                          There was a time when scholarly people thought the earth was flat but ideas and theories evolved as we learned new things and what is considered scholarly today may be nonsense tomorrow.

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


                            United States
                            Member #93947
                            July 10, 2010
                            2180 Posts
                            Offline
                            Posted: May 8, 2011, 12:54 pm - IP Logged

                            I can't help but wonder how they learn new things themselves, if not through reading. 

                            As far as I know no system exists that can predict the outcome of a lottery drawing better than a random quick pick and all the articles I've ever read say the same thing so reading them again isn't going reveal anything new about the theoretical systems we are discussing.

                            It's unlikely they have read many scholarly articles or journals because if they had they would know that source citations are not only expected, but often required.

                            What's suppose to be scholarly about picking some numbers to play in a lottery at this point?

                            They apparently are more impressed with people who "reinvent the wheel" rather than those who rely on the proven and accepted works of their peers and predecessors.

                            The wheel have been reinvented many times in the last hundred years.  Where do you think the US auto industry would be today if they were still manufacturing wooden wheels.

                            There was a time when scholarly people thought the earth was flat but ideas and theories evolved as we learned new things and what is considered scholarly today may be nonsense tomorrow.

                            RJOh,

                            Thanks for your thoughtful response.

                            It's hard to disagree with most of what you said above.  A [notable] exception is that I think it is foolhardy to cavalierly dismiss the accumulated knowledge of centuries from around the world.  Don't forget, the flat earth crew didn't land on the moon! Smile

                            I suspect the only thing we truly disagree on is whether it's possible to design a mathematically consistent, provable system, or not.  I also believe YOU would welcome any method that could test the efficacy of proposed systems to most people's satisfaction.  With this in mind, I would like to know what could be motivating the 2 or 3 people here that, to me, clearly, either have a psychological problem, or a financial interest, that causes them to fear the possibility their systems could be shown to be worthless.

                            How do you explain their bizarre behaviour?

                            --Jimmy4164

                              Avatar
                              Kentucky
                              United States
                              Member #32652
                              February 14, 2006
                              7295 Posts
                              Offline
                              Posted: May 9, 2011, 10:57 am - IP Logged

                              I said, "If you would take my advice and read..."

                              ...to which you added, "Someone's opinion who isn't a member of LP?"

                              Unbelievable. When you're unwilling to admit you don't understand what the links point to, wouldn't it make more sense to just quietly ignore the thread rather than keep spewing out evidence that you don't get it?

                              --Jimmy4164

                              P.S.  What is your goal posting in this thread?


                              "P.S.  What is your goal posting in this thread?"

                              I found an old pick-3 system circa 1980s that you can use to show us the merits of back testing simulating lottery systems.

                              8 7 9
                              1 4 7
                              3 8 0 
                              9 1 3
                              4 5 6 
                              6 2 1
                              2 3 7 
                              0 9 4 
                              5 2 8 
                              7 5 0

                              Following the system's instructions, I place the ten pick-3 digits in a gird in order of most preferred digits to least. While it does have a 1 in 100 chance of hitting a straight and a 1 in 9 chance to match a front pair, it's designed for box hits. Those 10 combinations have a 1 in 9 chance of matching any drawing boxed when three different digits are drawn. For practical lottery play 100 drawings or $1000 bet should be sufficient.

                              I'm assuming you'll simulate 100 random bets of $10 in each of the 100 trials and to make the payoff structure easier, use the KY pick-3 payoff of $100 to $1. If some of random bets matches a double digit box hit in one of the probable 27 drawings that's a $200 to $1 payoff. Show us if this system can't beat chance with 6 hits or can't show a profit by with 11 hits.

                              A frequency digit system player would probably change the order at least once every 10 drawings and a player that keys on repeating digits might change the order every drawing. But to make the testing easier for you, use the same 10 combos in all 100 simulated drawing.

                              Looking forward to you giving us an actual example of what you started over a month ago!