Welcome Guest
Log In | Register )
You last visited December 9, 2016, 3:58 am
All times shown are
Eastern Time (GMT-5:00)

How to calculate these types of probability?

Topic closed. 13 replies. Last post 5 years ago by SergeM.

Page 1 of 1
PrintE-mailLink
KnuckleHead's avatar - box

United States
Member #73037
April 3, 2009
147 Posts
Offline
Posted: February 3, 2012, 2:36 pm - IP Logged

Hello all,

I'm attempting to figure out if I can create a formula to:

1) Calculate the probability of the next number being either Odd, Even, or Same.

2) Calculate the probability of the next number being either Higher, Lower, or Same?

> From the below table of numbers (newest numbers are at the top - oldest to the bottom):

> [The next number in the sequence was a 7 - IE. 7 6 1 2 7 7 etc.]

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

AND

3) Is there a way to calculate the probability of which catagory the next number may come from?

I use low, medium, high for these numbers.

Low being 0, 1, 2, 3

Medium being 4, 5, 6

High being 7, 8, 9

 

Does anyone know how to do these types of formulas? Thanks ahead of time for any suggestions.

          The only DUMB question is the one question you DID NOT ask...

    garyo1954's avatar - garyo
    Dallas, Texas
    United States
    Member #4549
    May 2, 2004
    1720 Posts
    Offline
    Posted: February 3, 2012, 4:08 pm - IP Logged

    Based on the short term it would be easy to total each category and divide by the number of draws.

    Random being, well, random, gives you only a 10% chance of any number in a given position.

    However, a number going higher or lower, or staying the same, depends on the number. If the number is 7, then you have a 70% chance (0,1,2,3,4,5,6) it will go low, a 10% chance (7) it remains the same and a 20% chance (8,9) that it will go higher.

    Depending on the number of draws, the biggest drawback is the 'bulge' on the low digit totals. This is Texas Evening:

    The reason you have a disparity is 40% of the total is designated as Low, (0,1,2,3) versus 4,5,6 and 7,8,9.

    Might be an interesting idea to see how the other categories do as well. Wink

    Good Luck!

      KnuckleHead's avatar - box

      United States
      Member #73037
      April 3, 2009
      147 Posts
      Offline
      Posted: February 4, 2012, 11:12 pm - IP Logged

      Based on the short term it would be easy to total each category and divide by the number of draws.

      Random being, well, random, gives you only a 10% chance of any number in a given position.

      However, a number going higher or lower, or staying the same, depends on the number. If the number is 7, then you have a 70% chance (0,1,2,3,4,5,6) it will go low, a 10% chance (7) it remains the same and a 20% chance (8,9) that it will go higher.

      Depending on the number of draws, the biggest drawback is the 'bulge' on the low digit totals. This is Texas Evening:

      The reason you have a disparity is 40% of the total is designated as Low, (0,1,2,3) versus 4,5,6 and 7,8,9.

      Might be an interesting idea to see how the other categories do as well. Wink

      Good Luck!

      Evening garyo1954,

      Sorry not to have gotten back sooner...

      Below is my Texas Pick 3 Day game. According to my records, it has 1325 games in it since the last matrix changed on Nov. 12, 2007.

      TOTAL DRAWS - COLUMN COUNT - NUMBER FREQUENCY and PERCENTAGE CHART     
      Column Counts0123456789
      1st Column Count Number Frequency140134111120144145139112151129
      Percentage for 1st Column Count10.57%10.11%8.38%9.06%10.87%10.94%10.49%8.45%11.40%9.74%
      2nd Column Count Number Frequency136115135124144133119121138160
      Percentage for 2nd Column Count10.26%8.68%10.19%9.36%10.87%10.04%8.98%9.13%10.42%12.08%
      3rd Column Count Number Frequency134153130140143110120137129129
      Percentage for 3rd Column Count10.11%11.55%9.81%10.57%10.79%8.30%9.06%10.34%9.74%9.74%

      According to my table, in the 1st column, the number 8 has hit the most times. In the 2nd column, the number 9 has hit the most. And in the 3rd column, the number 1 is the winner. Playing the "most hits" (8 9 1) doesn't work. That's why I'm looking for another why to analyze the numbers...

      Based on the above table, I can see which number has hit the most times in which position. I was attempting to figure out how to use the "probability" functions to see if a different number popped to the surface other than what is apparent in the table. As for the 10% figure verses the 40% figure, if I understand correctly, each number has a 10% chance of being drawn (random). As to the L, M, H - it's just the catagories I placed the numbers into. I find it easier to grasp this way when I'm looking at the "historical table" draws.

      Is there a formula that provides the type of info I'm looking for OR would it be some sort of VBA script?

      By the way, which program are you using that provides a printout like you've shown?

      Thanks very much for the reply.

                The only DUMB question is the one question you DID NOT ask...

        garyo1954's avatar - garyo
        Dallas, Texas
        United States
        Member #4549
        May 2, 2004
        1720 Posts
        Offline
        Posted: February 5, 2012, 12:52 pm - IP Logged

        Evening garyo1954,

        Sorry not to have gotten back sooner...

        Below is my Texas Pick 3 Day game. According to my records, it has 1325 games in it since the last matrix changed on Nov. 12, 2007.

        TOTAL DRAWS - COLUMN COUNT - NUMBER FREQUENCY and PERCENTAGE CHART     
        Column Counts0123456789
        1st Column Count Number Frequency140134111120144145139112151129
        Percentage for 1st Column Count10.57%10.11%8.38%9.06%10.87%10.94%10.49%8.45%11.40%9.74%
        2nd Column Count Number Frequency136115135124144133119121138160
        Percentage for 2nd Column Count10.26%8.68%10.19%9.36%10.87%10.04%8.98%9.13%10.42%12.08%
        3rd Column Count Number Frequency134153130140143110120137129129
        Percentage for 3rd Column Count10.11%11.55%9.81%10.57%10.79%8.30%9.06%10.34%9.74%9.74%

        According to my table, in the 1st column, the number 8 has hit the most times. In the 2nd column, the number 9 has hit the most. And in the 3rd column, the number 1 is the winner. Playing the "most hits" (8 9 1) doesn't work. That's why I'm looking for another why to analyze the numbers...

        Based on the above table, I can see which number has hit the most times in which position. I was attempting to figure out how to use the "probability" functions to see if a different number popped to the surface other than what is apparent in the table. As for the 10% figure verses the 40% figure, if I understand correctly, each number has a 10% chance of being drawn (random). As to the L, M, H - it's just the catagories I placed the numbers into. I find it easier to grasp this way when I'm looking at the "historical table" draws.

        Is there a formula that provides the type of info I'm looking for OR would it be some sort of VBA script?

        By the way, which program are you using that provides a printout like you've shown?

        Thanks very much for the reply.

        The people who have studied gambling theory would be better suited to answer the formula question. In my opinion, the formula you are looking for depends on how you see the lottery drawings.

        1) Are they all independent  and random?

        2) When a digit hits, does this increase the probability of it hitting again?

        3) When a digit hits, does this decrease the probability of it hitting again?

        There are good and valid reasons on all sides. For example one could use the idea that 62% of the draws contain one digit from the draw before to validate the idea that those balls have a greater chance of hitting. On the other hand, that doesn't mean it is going to ber the first number. And don't forget each digit will hit on average every third draw.

        So how do you see it?

        The program I used is a personal ditty I've used for years. It is simple so I can add a few lines to get a fast answer.

          KnuckleHead's avatar - box

          United States
          Member #73037
          April 3, 2009
          147 Posts
          Offline
          Posted: February 5, 2012, 4:39 pm - IP Logged

          The people who have studied gambling theory would be better suited to answer the formula question. In my opinion, the formula you are looking for depends on how you see the lottery drawings.

          1) Are they all independent  and random?

          2) When a digit hits, does this increase the probability of it hitting again?

          3) When a digit hits, does this decrease the probability of it hitting again?

          There are good and valid reasons on all sides. For example one could use the idea that 62% of the draws contain one digit from the draw before to validate the idea that those balls have a greater chance of hitting. On the other hand, that doesn't mean it is going to ber the first number. And don't forget each digit will hit on average every third draw.

          So how do you see it?

          The program I used is a personal ditty I've used for years. It is simple so I can add a few lines to get a fast answer.

          Afternoon garyo1954,

          To answer your questions...

          1) Since PowerBall, Mega Millions, and the Texas Lottery use some sort of "ball" machines and not "RGN" programs, then yes, I believe that each draw is independent and random.

          2) and 3) Based on answer 1), I don't see how it could "increase" OR decrease the probability if the draws are truly random.

          Therefore, is it possible to "track" the historical draw data to get an idea of the ball positions as they've hit in the past. Yes, the draws are random, but...., is there a way to track the hit positions in a column for each draw thereby producing a "probability" for each number based on the historical position hits? Some sort of formula/Vba that could track positions would probably produce a better chance on a "RGN" style game because with enough draws, it may be able to see the patterns produced in an "RGN" program.

          Your correct, this subject is definitely for a "gambling theorist", and once this could be figured out, a very good programmer... (of which I'm not either).

          Thank you for the lively discussion. It provides me with different ideas to search for and to think about.

                    The only DUMB question is the one question you DID NOT ask...

            garyo1954's avatar - garyo
            Dallas, Texas
            United States
            Member #4549
            May 2, 2004
            1720 Posts
            Offline
            Posted: February 6, 2012, 5:43 pm - IP Logged

            Afternoon garyo1954,

            To answer your questions...

            1) Since PowerBall, Mega Millions, and the Texas Lottery use some sort of "ball" machines and not "RGN" programs, then yes, I believe that each draw is independent and random.

            2) and 3) Based on answer 1), I don't see how it could "increase" OR decrease the probability if the draws are truly random.

            Therefore, is it possible to "track" the historical draw data to get an idea of the ball positions as they've hit in the past. Yes, the draws are random, but...., is there a way to track the hit positions in a column for each draw thereby producing a "probability" for each number based on the historical position hits? Some sort of formula/Vba that could track positions would probably produce a better chance on a "RGN" style game because with enough draws, it may be able to see the patterns produced in an "RGN" program.

            Your correct, this subject is definitely for a "gambling theorist", and once this could be figured out, a very good programmer... (of which I'm not either).

            Thank you for the lively discussion. It provides me with different ideas to search for and to think about.

            I am a big fan of historical data. I tend to use it mostly as a baseline while tracking recent trends.

            I might have something that would help. I post it later this evening when I get a chance to find the program and do an update.

            G

              KnuckleHead's avatar - box

              United States
              Member #73037
              April 3, 2009
              147 Posts
              Offline
              Posted: February 6, 2012, 6:48 pm - IP Logged

              I am a big fan of historical data. I tend to use it mostly as a baseline while tracking recent trends.

              I might have something that would help. I post it later this evening when I get a chance to find the program and do an update.

              G

              Afternoon garyo1954,

              That's what I'm attempting to figure out, "trends". How to construct formulas or subs/scripts for... "trends" in the data.

              Anything can be counted, but what good is having all this historical data if you don't know how to put it all together and analyze it properly?

              Thank you. KH

                        The only DUMB question is the one question you DID NOT ask...

                garyo1954's avatar - garyo
                Dallas, Texas
                United States
                Member #4549
                May 2, 2004
                1720 Posts
                Offline
                Posted: February 6, 2012, 10:26 pm - IP Logged

                Afternoon garyo1954,

                That's what I'm attempting to figure out, "trends". How to construct formulas or subs/scripts for... "trends" in the data.

                Anything can be counted, but what good is having all this historical data if you don't know how to put it all together and analyze it properly?

                Thank you. KH

                Can't find the stuff on this machine.

                But this is along the idea:

                Top part is the overall breakdown of repeating digits - First digit repeats back to first position 285 times, it moves to the second position 240, and moves to the third 234. Reads the same for the second and third digits.

                Bottom of the chart shows how many times the digit in each position remains the same, goes lower, or goes higher. But you might be better served to use a chart like this that shows the change on a draw to draw basis.

                I don't know. Maybe (hint-hint) WIN D has an idea??

                  Avatar
                  NASHVILLE, TENN
                  United States
                  Member #33372
                  February 20, 2006
                  1044 Posts
                  Offline
                  Posted: February 11, 2012, 1:59 pm - IP Logged

                  Afternoon garyo1954,

                  That's what I'm attempting to figure out, "trends". How to construct formulas or subs/scripts for... "trends" in the data.

                  Anything can be counted, but what good is having all this historical data if you don't know how to put it all together and analyze it properly?

                  Thank you. KH

                  KnucklyHead

                  You are dealing with random here.  You are leaving the world of probabilities.  There is no mathematical proof of anything while you are vacationing in the realm of random.  Therefore, anything you may think or say can not be proven nor disproven.

                  Having said this, you must make decisions as you see them.  Does a drawn number have a better chance of being drawn in the next drawing or less of a chance?  Makes no difference.  You must make a determination and then stick with it. 

                  If the last 5 numbers drawn resulted in an odd sum, will the next drawing result in an even sum?  Again, makes no difference.  You must make a determination and then stick with it.

                  As for me, I see no advantage in determining trends.  Trends come and go.  That's why they are called trends.  Usually if you have enough data to indicate a trend, you also have enough time passed to neutralize that trend.  I have made my decision and I am sticking to it.

                    Avatar
                    bgonçalves
                    Brasil
                    Member #92564
                    June 9, 2010
                    2125 Posts
                    Offline
                    Posted: February 11, 2012, 4:51 pm - IP Logged

                    Hello, knuckl, so you can use 0,1,2,3 = high   Average = 4.5 6,7,8,9 = high,   To sort the endings, ok! The big problem as gasmu speak, are the trends and repetitions of a result to another, because one is projected to result, (single event), making impossible any prediction for the whole of the draw, since half if you can predict, for example a lotto 49/6 3-4 numbers can be predicted with confidence, then we have to tackle is the prediction of half of a lottery,   The other part is random, a lottery can see example 49/6 a 100 playing cards, you hit numbers 3-4 pretty easy, this is the way

                      KnuckleHead's avatar - box

                      United States
                      Member #73037
                      April 3, 2009
                      147 Posts
                      Offline
                      Posted: February 12, 2012, 8:55 pm - IP Logged

                      Evening all,

                      Reading the last 2 posts, I have to admit that I understand that the draws are "random". I track the Texas, Power-Ball, and Mega Millions games. I've watched the videos of the balls come out of the ball machines. But, without enough historical data to analyze, it appears that specific balls react differently based on the historical data. That's what I'm attempting to figure out how to analyze. Just because a table contains ball numbers with their associated draw numbers in it for a specific time period doesn't mean that those calculations by themselves are enough data to feel confident on the next drawn numbers. My experience has shown me that my data is incomplete and needs different and more intense analyzation.

                      That's what this post is all about... Understanding different types of analyzation concepts.

                      So, if anyone is interested/willing to share/discuss some of their different ideas with a "non-math geek" and a very "inexperienced" VBA creator, I'd be more than willing to attempt to comprehend/understand new ideas...

                      Thanks to all that have replied and to those that have taken the time to read this post. Hopefully, new ideas can be created...

                                The only DUMB question is the one question you DID NOT ask...

                        SergeM's avatar - slow icon.png
                        Economy class
                        Belgium
                        Member #123700
                        February 27, 2012
                        4035 Posts
                        Offline
                        Posted: February 27, 2012, 11:12 am - IP Logged

                        You have ten digits there. Your probability is 1/10.

                        If you play pick 3 you are likely to lose half of your money.

                        What is random?

                        I don't play pick 3, the payout is low.
                        -1 +500 = 499 with odds 1:1000
                        499*0.001=0.499

                        You lose 0.501 of your bet.

                          KnuckleHead's avatar - box

                          United States
                          Member #73037
                          April 3, 2009
                          147 Posts
                          Offline
                          Posted: March 1, 2012, 9:35 am - IP Logged

                          You have ten digits there. Your probability is 1/10.

                          If you play pick 3 you are likely to lose half of your money.

                          What is random?

                          I don't play pick 3, the payout is low.
                          -1 +500 = 499 with odds 1:1000
                          499*0.001=0.499

                          You lose 0.501 of your bet.

                          Hello SergeM,

                          Your statement of: "I don't play pick 3, the payout is low.", is very understandable.

                          It would appear that from your point of view, "low pay out" games are a waste of time. This is also understandable.

                          From my point of view, attempting to comprehend different theories of probabilities and combinations, then attempting to figure out how to create the program language to implement an idea, it's much easier to work with a smaller number pool. That being stated, I still attempt to apply what I've learned with the Pick 3 game to the other games I track. Sometimes it works, sometimes it doesn't...

                          But here's the difference, if my strange theories pan out, it's a little money in the pocket instead of no money in the pocket.

                          Which is more valuable, an empty pocket or a little in the pocket while you continue to test different theories?

                          I thank you very much for your reply, it's showing me that there's still interest in this post and that has got members thinking... This is good... Thank you.

                                    The only DUMB question is the one question you DID NOT ask...

                            SergeM's avatar - slow icon.png
                            Economy class
                            Belgium
                            Member #123700
                            February 27, 2012
                            4035 Posts
                            Offline
                            Posted: March 17, 2012, 8:25 pm - IP Logged

                            I analysed some of pick 3 and noticed that if things would go well, you would pay a lot of tickets and you would get just a bit over break even. If the case doesn't come out, you lose big money.

                            The only interest that I had in pick 3 is working out odds and tables in Excel.

                            Again, if I consider that you could win a pick 3 by some method, get break even consistently, then you might double your money at roulette. That I would do!

                            Technical problems like wheeling can be solved. Printing on the tickets is also possible, if you can handle the settings and the stinky printer.

                            The main part is the picking of the numbers and getting a payout that is higher than all the money that you put into the tickets. If you have a sure way to win, I gather the cash to play. For pick 3, for a dollar input, you would need to win a bit more than three times in thousand to make 50% net winnings.
                            The more you spend, the more you must win.

                            I might work out probabilities, but I really don't believe in that one.