Combination Groups Linked to WNH Differences - Statistics an

[PadawanLotto]

Hi Jerzy, remember the strategy The difference in 11 to 11 columns is at most -20/20 Leave? How many times did it work? Have you ever tried different difference combinations and Leave and they didn't work? It's not easy to find the correct difference range. Look into The difference in 11 to 11 columns is at most -20/20 Remove and other difference combinations Remove, most likely you won't hit the combination that removes the jackpot combination. Just a thought, consider that the last difference balance drawn will not be drawn in the next drawing, maybe a good place to start.

[Jerzy]

Hi Padawan,

You already published your idea of setting the WNH Sum/Differences filter to "the difference in 11 to 11 columns is at most -20 to 20" and "Remove" some time ago. I do not play my lotto often, but when I play I use this setting. Thank you.

How many times could it work? It is easy to find it out by using Back-test function.

In my game this setting "the difference in 11 to 11 columns is at most -20 to 20" passed in 81 drawings (6.89%) and failed in 1094. And with this setting it is possible to Remove/Leave several thousand tickets from the full package. For example, today I could Remove/Leave 612,281 tickets from 8,145,060.

Back-testing shows the dates of drawings when this setting was passed. In 2007 it happened 9 times, sometimes in consecutive drawings, and it happened more often than in previous years.

But if I play with Leave and get 612,281 tickets, what other filters could reduce this package to a manageable size, and leaving in it some prize winning tickets? I'm afraid nobody knows the answer, and it is rather a matter of chance.

Best Regards

Jerzy

[laurontario]

I studied entire history of Canadian 649 and noticed that the differences in Col 0-10 never repeats from one draw to another - well... actually it's about 2% from the totals, so, there are about 98% chances they won't repeat in the next draw.

I have therefore filtered the package for min/max = last draw difference - Remove. I noticed that this way I can reduce the number of tickets in the package with about 18-20% and still keep the jackpot ticket.
I can then work on the remaining tickets and filter more, setting new ranges, this time leaving the tickets in the package (according to the strategy part1+2 posted here a while ago).

The advantage is that if, let's say, at last draw I had a difference of +8 and now I predict a difference between +5...+12 (I wish to be so accurate in my predictions), the tickets that give a difference of +8 will be in my package, and I don't want them.

Current example:
After filtering the 14 mils for StDev and Sum ranges, I ended up with 4,188,801 tickets. I applied this strategy and I now have 3,394,221 (that is 794,580 tickets less and, if my ranges for sum and StDev were correct, I still have 98% chances to have the jackpot ticket in my combinations).
I can now start looking at the WNH charts and predict the ranges.

Any other ideas/thoughts?

[laurontario]

Meant to attach a screenshot too...

[Jerzy]

Hi Laurontario,

I use the filters set to Remove with the last draw exact settings. In most cases it works.

Do you filter WNH Sums according to their O/E pattern in columns? Sums in columns could be odd or even. There could be so many O/E patterns in 11 columns that they repeat very rarely.

Using the WNH Sum/Differences filter and "at least 3 - 9 columns must pass", Leave, I can remove last 50 - 100 O/E sum patterns (or even more), usually without losing the jackpot ticket. I do it when the package is already prefiltered with quick working filters and reduced to a small size (6/45 lotto).

Best Regards

Jerzy

[CARBOB]

[quote=Jerzy]
Hi Laurontario,

I use the filters set to Remove with the last draw exact settings. In most cases it works.

Do you filter WNH Sums according to their O/E pattern in columns? Sums in columns could be odd or even. There could be so many O/E patterns in 11 columns that they repeat very rarely.

Using the WNH Sum/Differences filter and "at least 3 - 9 columns must pass", Leave, I can remove last 50 - 100 O/E sum patterns (or even more), usually without losing the jackpot ticket. I do it when the package is already prefiltered with quick working filters and reduced to a small size (6/45 lotto).

Best Regards

Jerzy
[/quote]

Jerzy, would you please post an image of your WNH Sum/Differences settings. I must be doing something wrong, no combos left in package!! Thank you.
My image is attached.

[Jerzy]

Carbob,

You should select each of the 11 checkboxes situated vertically on the left of the filter (you should filter all the columns but with the current setting you do not filter any of them). After that click on Update sums, Leave and OK.

O/E pattern checkboxes selected correctly. Logical condition AND is correct. At least 3 to 9 columns must pass is OK.

I leave the checkbox for "The difference in ... to ... columns is at most ..." unselected. I never used this option with O/E sum patterns but you can try it and see what happens.

Before using this filter I usually Lookup for Min & Max Ranges.

Let me know how it works.

Jerzy

[laurontario]

[quote=Jerzy]
Do you filter WNH Sums according to their O/E pattern in columns? Sums in columns could be odd or even. There could be so many O/E patterns in 11 columns that they repeat very rarely.
[/quote]

Never thought about that. I'll try it and see how it works.
Thanks for the tip.

:wand: Damn it... Last night I only got two numbers out of six - invested $20.00... got nothing!
I got the sum range correct, got also the stdev range... have to see where the heck did I do wrong :motz: