In Excel, I have tracked Games out or Skips converted them to Frequencies and created large Charts that compared events side by side over past game blocks. I have looked at Followers, Skip Trends, etc. all of these processes produce a sequence of numbers that I can perform the same calculation for all over again. I don't know where the iterations end, probably never. It always proves that the Draws are Random. Every time I think the most frequent event will occur next it rarely does. My hope is that reducing the number of Event Types (in this case 5) gives me a 1 in 5 Chance of being right about the next event.
In the case of the sequence I asked about I know the following:
There are 2700 Events in the sequence Each Event Count and Percentage are as follows:
(0 45 2%) (1 299 11%) (2 792 29%) (3 1035 38%) (4 530 20%)
The sequence is a fixed process that follows a random event so It is a parallel random sequence. I know that parallel lines never intersect. I look at my Sequence as a "Poly line". In Autocad a Polyline is a line that has multiple vertexes and variable line widths between vertexes. It is cool to look at but is no better at stimulating me to make the right selection than looking at the numbers. Neither is very effective. It just gets more complicated.
The sequence has many cousins. The Game type S-D-DD-T-Q is a parallel sequence. The number picked in the draw order, the sum of numbers.....................endless.
So my question was is there any way to ignore the method of creation and simply forecast from the sequence itself. I know the answer is No. Parallel line do not intersect. I am looking for some Math that I do not know or some other method.
It might be interesting to get a group of LP members to look at some form any sequence (created by anyone, not just mine) and make a prediction, pool all of the predictions and see if the group can get it right.