Per the GIF and the hot-cold gradients posted earlier in this thread, here are my theories as illustrated below:
- Hot trends level out or begin to go cold.
- Middling trends can go either way.
- Cold trends level out or begin to get hot.
What I'll do is calculate a trend, then weight the numbers toward their theorized trajectories.
For tracking and metrics, I'll weight the numbers for odd in odd months, for even in even months.
This means Jan, Mar, May, Jul, Sep, Nov trend calculation will get:
- hot squiggly trend stays nominal, following the black arrow and staying flat (etc.)
- middling trend gets a +1
- cold trend gets a +1
And Feb, Apr, Jun, Aug, Oct, Dec trend calculation will get:
- hot trend gets a -1
- middle trend gets a -1
- cold trend stays nominal
Or, I could determine the trajectory of each number (patent pending), apply the number shown below, then simulate that's how many of each ball is in the "tumbler". It's easy enough to simulate a tumbler with any number of balls in a spreadsheet. Let's say that a number is trending hot. I could give it a weight of 4 on odd months, but a weight of 3 on even months.
Each adjustment is from a mean calculation that I came up with (patent pending), and the adjustment is an attempt to simulate the changing red to blue, or blue to red gradients as shown above.
Patent pending lol. You already know my sample size is cast in stone, cannot be changed. If I get this figured out, you'll have the entire spreadsheet method, no cost, no question.
Keep in mind that I don't try to calculate the 5 numbers most likely to hit. I try to calcluate the 5 numbers least likely to hit and wheel the rest according to some weight calculation. I have 9 versions of this thus far. This is revision 10 that I'm discussing today.
Early on the first day of each month, I'll post my last month's prediction statistics as calculated on the prediction board to see if this has any merit. I will continue this through 2021 then have a new method developed by 2022. I will post the metrics on May 1 of the April predictions.
What do you think?