Coldest |
10 |
17 |
36 |
41 |
42 |
Cool |
4 |
5 |
6 |
7 |
12 |
14 |
18 |
19 |
20 |
21 |
22 |
23 |
24 |
32 |
33 |
35 |
38 |
44 |
45 |
48 |
Neutral |
1 |
2 |
3 |
8 |
9 |
13 |
15 |
16 |
25 |
27 |
29 |
31 |
37 |
39 |
40 |
43 |
46 |
Warm |
26 |
28 |
30 |
34 |
47 |
Hottest |
11 |
Extra ball
Coldest |
1 |
3 |
5 |
6 |
8 |
11 |
12 |
13 |
14 |
16 |
17 |
18 |
Neutral |
2 |
4 |
7 |
9 |
Hottest |
10 |
15 |
Depends on what one considers to be hot or cold.
I see two data points:
- How many times drawn in a given sample size
- How long since drawn (can be N times with trend/frequency adjustments)
The formulas to calculate a resultant pool to select from those two data points is constantly being tweaked to get better results.
Let's see how my cold to hot list above does on the next draw.
I don't want a system that carefully selects the few numbers "guaranteed" to drop. Hone a system so precisely that any movement off target guarantees a loss.
I want a system that excludes very few numbers, which increases my odds ever so slightly. If the system is a little off target, so be it. Still have a chance at something.
"Rifle vs. shotgun vs. blunderbuss methods of choosing numbers."
I play cold numbers. More accurately, I mostly avoid the hottest and "wheel" the rest based on their relative coldness weight. But, my system does occasionally pick a hottest number.
Funny saying "wheel" when it's possible to pick all the numbers, but to each their own I suppose.
I deliberately skew the hot-cold rank to prevent too many excluded numbers. That's why the graphs are not balanced around neutral. I am also not weighting by fifths, for example.
Top graph for the 1-5 ball: For hottest, there is (1) total number 11 in the randomizer. There will be (5) each of the 10, 17, 36, 41, and 42. Cool gets 4 each. Neutral gets 3 each. Warm gets 2 each.
The extra ball gets (1) each of the 10 and 15. There will be (3) each of the 1, 3, 5, 6, 8, 11, 12, 13, 14, 16, 17, and 18. Neutral gets 2 each.
A spreadsheet easily handles giving each number a weight, then picking a random number within that given range of weights.