gsobier said:
Nick:
If you did this for the pattern of last 10 draws, how reliable would your list be for the last 1,000 draws?
What is the maximum hit count, and what would the averages be for hit counts for your list of less possible to hit?
Regards,
George
Hi George,
1) If you mean, testing the pattern of the
last 10 Draws vs the results of 1000 Draws then the answer is as I ported.
2) If you mean finding the patterns of 1000 draws and testing vs the results
then my conclusion is that we will have no pattern
but the Frequency Distribution of these 1000 draws.
There is not noticeable pattern for so many, in this case, 1000 Draws.
I hope that the above answers both questions, including the
reliability of hit counts.
But to understand a bit more of this particular algorithm
I provide below a pseudocode for it.
1. For the last n Draws, 10 in this case
2. Find some Rules You Specify, "Algorithm 02" in this
case which is described below.
3. Use the Rule:
If Draw 1 does not have any of the Numbers of the
Next draw then
What Comes up more frequently
Store that Frequency.
Repeat the test for 2 to 10 draws
4. From the resulting Frequency Distribution
Simulate the Normal Bell Curve using the appropriate
function using the Average of the Frequency and it's
Standard Deviation.
Repeat the test x Times and store the results
5 Display the Final Simulated Distribution
6. Select x Numbers from within 1 SD from that Distribution
So the less probable numbers which I posted belong to the two
tails of that Simulated Distribution.
If you have any more questions I'll be glad to respond.
Nick