Welcome Guest
Log In | Register )
The time is now 12:25 pm
All times shown are
Eastern Time (GMT-5:00)

How to Determine If Results are Likely Not Random?Previous TopicNext Topic

Topic closed. 9 replies. Last post 3 months ago by JKING.

Page 1 of 1
PrintE-mailLink
Avatar
New Member
Paris
France
Member #122654
February 5, 2012
10 Posts
Offline
Posted: February 9, 2012, 6:30 am - IP Logged

General Question:  How can I determine if results are likely not due to randomness?

Specifics: The results of my Monte Carlo simulation shows that, out of a large number of trials, my results (based on only 100 actual bets) are in the top 30% and top 5% (winnings, matches, whatever). While that's nice, there are a lot better random results. So how can I convert the results of the Monte Carlo simulation into a statement such as "There's a 95%/99% confidence that the results are not due to randomness." What kind of results and what level of results are needed to create such a statement?

    Avatar
    New Member
    Paris
    France
    Member #122654
    February 5, 2012
    10 Posts
    Offline
    Posted: February 11, 2012, 8:18 am - IP Logged

    In answer to my own question, there are web page calculators that, given a list of values, will calculate value count, mean (average), variance and standard deviation. There's other web pages that will use those results to calculate confidence intervals for the mean for various alphas (90%, 95%, 99%). These sites have exactly answered my question.

      Avatar
      NASHVILLE, TENN
      United States
      Member #33768
      February 20, 2006
      866 Posts
      Online
      Posted: February 11, 2012, 1:26 pm - IP Logged

      Read up on the Chi-Square and T-Squared in statistics.

        RJOh's avatar - chipmunk
        mid-Ohio
        United States
        Member #9
        March 24, 2001
        13924 Posts
        Online
        Posted: February 14, 2012, 2:49 pm - IP Logged

        General Question:  How can I determine if results are likely not due to randomness?

        That's like asking  when some one calls out heads how can you tell if they flipped a coin to come up with it.

        * that which happens most *
        * is most likely to happen again *
         

        Thumbs Up

          sully16's avatar - sharan
          GO TIGERS
          MICHIGAN
          United States
          Member #82269
          October 28, 2009
          8818 Posts
          Offline
          Posted: February 26, 2012, 9:31 am - IP Logged

          The Bell curve,

          any day now.

            JADELottery's avatar - facepuke
            The Quantum Master
            West Concord, MN
            United States
            Member #21
            December 7, 2001
            2441 Posts
            Offline
            Posted: February 27, 2012, 12:21 pm - IP Logged

            Presented 'AS IS' and for Entertainment Purposes Only. Use at your own risk.
            Any gain or loss is your responsibility.

            Order is a Subset of Chaos.
            Knowledge is Beyond Belief.

            The Name Anagram
            name - Douglas Paul Smallish
            amen - US God plus Islam Allah
            mean - Jehocifer

            JADE Quintrains

              garyo1954's avatar - garyo
              Dallas, Texas
              United States
              Member #4629
              May 2, 2004
              711 Posts
              Offline
              Posted: February 27, 2012, 12:47 pm - IP Logged

              LMAO....I Agree!

                Avatar
                bgonçalves
                Brasil
                Member #92592
                June 9, 2010
                662 Posts
                Offline
                Posted: March 2, 2012, 9:22 am - IP Logged

                Hello, lorso. As in all probability and statistics is 80/20. The question is how the concept could apply 80/20 lottery, Maybe join monte carlo with the concept of pareto80/20! That is the bell curve with 80/20

                  Avatar
                  bgonçalves
                  Brasil
                  Member #92592
                  June 9, 2010
                  662 Posts
                  Offline
                  Posted: March 2, 2012, 9:53 am - IP Logged

                  Hello, using monte carlo, and concept Pareto (80/20) could be such a lottery 49/6 taking advantage a conjuçao (associçao) favorable pairs and trios, six positions in the lottery 49/6 or other lottery because we need to see who is coming in who draws It's not who came out with whom, male or female! Ahhhh, but even numbers and odd   In twos and threes.

                    JKING's avatar - Kaleidoscope 3.gif

                    United States
                    Member #5700
                    July 13, 2004
                    983 Posts
                    Offline
                    Posted: March 3, 2012, 5:30 pm - IP Logged

                    Hmmmm..

                    1,2,3,4,5

                    1,2,3,4,5

                    And the odds of 5 numbers repeating are? Odds of 5 numbers in sequence? Odd of 5 numbers in the same decade?....

                    1,2,3,4,x

                    1,2,3,4,x

                    And the odds of 4 numbers repeating are? Odds of 4 numbers in sequence? Odds if 4 numbers in the same deacade?

                    In other words, if you get results that are 1 in million chance happening repeatively over a small amount of draws...there is a problem.

                    Welcome to the California Fantasy 5.

                    The most valuable things in life are not things.

                    My predictions at the LP are simply a testing ground of my latest hairbrained ideas, please don't take them seriously.