Distinct of History Differences in Columns

Postby Jerzy » Mon Jan 28, 2008 7:09 am

Joe,

Thank you Joe.

New filters require new terminology to describe their charcteristics and their function. The development of new terminology is not easy. A disagreement over the use of terminology is possible and even friends disagree sometimes.

The name I proposed for this filter was just a suggestion. Give this filter a name that you think is the most appropriate.

Jerzy
Jerzy
 
Posts: 811
Joined: Mon Sep 18, 2006 5:26 am

Postby Joe » Mon Jan 28, 2008 7:14 am

I started working on that filter, Named "Occurrence of Distinct HDs", we can change the name if any one suggests good name to this filter, thanks
Joe
 
Posts: 1227
Joined: Mon Sep 18, 2006 6:51 pm
Location: Petaling Jaya, Malaysia

Postby Jerzy » Mon Jan 28, 2008 7:50 am

Joe,

That's fine. Thank you.

Jerzy
Jerzy
 
Posts: 811
Joined: Mon Sep 18, 2006 5:26 am

Postby Picsou » Mon Jan 28, 2008 11:07 pm

Hello all

Here some simulation on my loto 6/49 game

you can see the values of all the 11 differents columns of distinct history differences and the results of how many ticket to play to have the jacpot garantee.
it's work in all the cases!!!!! i have tested it one by one.

how to read it:
for example if you select value on the 1rst draw on the 16/01/2008, when you applied these 11 values on filter , you have to play only 6 tickets to win the future draw (jackpot)...(second draw of 16/01/2008) and so...

You can see values go from 3 to 6 maximun (only one exceptional 2)
you can see that if you have only numbers 5, 6 you have must tickets to play. (600 to 950)
if you have a lot of 3, 4 and some 5 and 6 filter can give few tickets to play (1, 2, 6, 14, 20) >> so to priviligied this values

In mathematics you have 3^11=177147 possibles combinaisons of differents filters.
The problem is simple: if you have the 11 goods values on the good places, YOU WIN JACKPOT.

Now, how to have the good values and at the good places???
only statistics and observations.???

First:
you can decided to play some "hard combinaison" that give only one ticket to play >>> EX: 10 "hard combinaisons (max 3 and 4, and some 5,6), >> give 10 potentiel jackpot tickets, play only these filters always (tickets will change at each future draw) and see.

Second: make statistic of values for each differentscolumns and make at each draw a different combinaison.
you can decided to play or not fonction of the numbers of tickets.

I want to use VRA analys to predict future value for each colunmn, but you must have a minimum of 500 data for each column
Go to very interisting site and download VRA analys http://zarnia.250free.com/index.htm

My question, is here somebody that can write a program to have all the values of 11 columns for a loto datafile of thousands draws.
It's very hard to make simulation draw for draw!!!! (time)

if this can be put in EL?? or only a simple basic or excel routine.
with this, i can after use VRA prediction.

I hope you have understood that i mean, (sorry for my ENGLISH)

Thank's for your answers

cdt
You do not have the required permissions to view the files attached to this post.
Picsou
 

Postby Picsou » Mon Jan 28, 2008 11:10 pm

Hello

an EXCEL file is better to read!!!

cdt
You do not have the required permissions to view the files attached to this post.
Picsou
 

Postby Picsou » Thu Feb 07, 2008 8:57 pm

Stan

is it possible to have these information on statistic on wining number for each draw?
a column with 5 6 3 5 5 5 6 4 5 5 6 for exemple

cdt
Picsou
 

Postby stan » Wed Feb 13, 2008 10:36 pm

[quote=Picsou:1202410630]
Stan

is it possible to have these information on statistic on wining number for each draw?
a column with 5 6 3 5 5 5 6 4 5 5 6 for exemple

cdt
[/quote]

let me read these posts first but i guess it shouldn't be a big problem:)
Expert Lotto Team
User avatar
stan
Site Admin
 
Posts: 6338
Joined: Thu Sep 23, 2004 1:01 pm

VRA is unreliable for segments.

Postby tax » Sun Feb 17, 2008 7:14 am

Hi Picsou,

The VRA program you gave us the link to is great and thank you for that. While attempting to predict drawn number might be somewhat reliable, it is extremely unreliable in predicting the next segment combinations using your suggested method of transforming each segment combination into a number, for example 6-0-0 = 1, 5-1-0 = 2 etc, or even if you group segment combinations into 1 group, for example 6-0-0, 5-1-0 and 5-0-1 = 1, 4-2-0, 4-1-1 and 4-0-2 = 2, etc especially if you load a cvs into VRA with 11 numbers per raw, each number being equal to the transformed segment combination. why? because while numbers drawn in a ticket are mutually exclusive, that is if number 20 is drawn, it cannot be drawn again in the same ticket, the same is not true for the segments. for each draw, segment combinations are not mutually exclusive. you could get 6-0-0 in col -2 and again in col -5 for example.

So asking Stan to program EL to transform each segment combination into a number is quite useless in this context.

In the spreadsheet attached you will find the example I explain below.

I have loaded the segment combinations for 10 draw in the worksheet called `current`. The worksheet called `ranges` shows all segment combinations in ascending order with a number in the column next to it. The 1st group has been given the name `range` while the 2nd group shows the same segment combinations but grouping them. This 2nd group has been given the name of `grouped_range`. I will use these names in my excel vlookup formulas to change my segment combinations in each columns in the `current` worksheet to what you want Stan to do for you.

Remember that VRA wants numbers, not text, so loading a bunch of 6-0-0, 4-1-1 etc means nothing to VRA.

In the `VRA all 11 col` worksheet you will see the result of the vlookup. As you can see from row 1 I have 2 in column A as well as F for example. Since I do not have mutually exclusive numbers for each draw, VRA will not be reliable, not to mention finding the correct number of dimensions and delay to be used. good luck!

Another way would to treat each column separately (worksheet `VRA col 0` and `VRA col -1` are example). there you will find the same results as excel column A and B respectively fro the `VRA all 11 col` worksheet. Loading each column into VRA representing 1 column only at a time, however, defies the purpose of having dimensions and delay so it becomes rather useless.

I am showing the same technique by grouping segment combinations into groups for example 6-0-0, 5-1-0 and 5-0-1 = group 1. You will find the results in the worksheet `VRA grouped 11 col`, with each column treated separately in `VRA grouped col 0` and `VRA grouped col -1`. This again either defies the purpose of mutual exclusivity or defies the purpose of dimensions and delay.

So what can you do? Well, there is a solution, but EL cannot help, only Excel would. First step would be to transform a text value like segment 6-0-0 to 600 which can be exported into VRA. You will find an example and formula into the worksheet `VRA simple text to number`.

The formula is like =CONCATENATE(LEFT(current!B2,1),MID(current!B2,3,1),MID(current!B2,5,1))

The non mutual exclusivity problem still exit. Although I have numbers now, if 6-0-0 appears in col 0 and col -5 for example, the formula would give 600 for both columns. What can you do? You can add 1 to the result as you move 1 column to the right. An example is in the `VRA unique text to number` (although this is not totally unique yet). in the first raw, look at the formula applied in column A. The formula is like: =CONCATENATE(LEFT(current!B2,1),MID(current!B2,3,1),MID(current!B2,5,1)) + 0
For column B the formula is now:
=CONCATENATE(LEFT(current!B2,1),MID(current!B2,3,1),MID(current!B2,5,1)) + 1
etc etc.

Have I got ride of the mutually exclusive problem? not quite. Look at the `mutually exclusive columns` worksheet. The bolded ones (column J and K) are potential values that would conflict with values in another column (column A and B).

So what? why not divide those 11 columns into 2 groups of 6 each. Besides, expecting VRA to predict 11 future values in 1 go is crazy! For each group of 6 you can add 1 to the next column without incurring the mutually exclusivity issue, hence making VRA for more reliable in its predictions, first because you only predict 6 values at a time (from col 0 to -5 then col -5 included again to -10) and second because each group will have mutually exclusive numbers.

I have also provided an area in each group worksheet for you to type the results from VRA and the resulting segment combinations are re-formatted automatically for you.

Enjoy
Laurent
You do not have the required permissions to view the files attached to this post.
tax
 
Posts: 73
Joined: Mon Feb 21, 2005 10:23 am

Postby Picsou » Sat Feb 23, 2008 1:11 am

Hello Tax


I made a simulation to find all the groups of 6 numbers on my loto draws. It exist few recurrence of combinations in the time. This methode thus is to be abandoned.

Thank you for your help and the file Excel

Cordialement
Picsou
 

Postby stan » Sat Feb 23, 2008 9:57 pm

i've got the stats for distinct diffs count ready, they'll be in the next build i upload (in a few moments actually:)
Expert Lotto Team
User avatar
stan
Site Admin
 
Posts: 6338
Joined: Thu Sep 23, 2004 1:01 pm

Postby Picsou » Sun Feb 24, 2008 12:49 am

Hello Stan

Thank's for your addon

Cordialement
Picsou
 

Postby Joe » Sat Nov 22, 2008 4:18 pm

I am going to make some changes in this filter "Distinct HD in Columns" in order to use this filter for all games, only problem is after making the changes, you can not use the compound/group filters which are used the old version of this filter, you need to re-create the compound/group filters, if any objections please let me know,
Joe
 
Posts: 1227
Joined: Mon Sep 18, 2006 6:51 pm
Location: Petaling Jaya, Malaysia

Postby Joe » Sat Nov 22, 2008 6:51 pm

New Versions of "Distinct History Differences" has been added,
Filter can be used for all games.
Expert Lotto Pro license is required to use this feature.
Joe
 
Posts: 1227
Joined: Mon Sep 18, 2006 6:51 pm
Location: Petaling Jaya, Malaysia

Postby teoman » Sat Nov 22, 2008 8:15 pm

[quote=Joe]
New Versions of "Distinct History Differences" has been added,
Filter can be used for all games.
Expert Lotto Pro license is required to use this feature.[/quote]
Hi Joe,

Filter "Distinct History Differences" does not work as shown in the image attached

Thanks!
You do not have the required permissions to view the files attached to this post.
teoman
 
Posts: 191
Joined: Sun Apr 27, 2008 11:27 pm

Postby Joe » Sun Nov 23, 2008 4:24 am

[quote=teoman]Filter "Distinct History Differences" does not work as shown in the image attached[/quote]yeah, i'll check this, thanks for reporting bugs,
Joe
 
Posts: 1227
Joined: Mon Sep 18, 2006 6:51 pm
Location: Petaling Jaya, Malaysia

PreviousNext

Return to Plugin Development

Who is online

Users browsing this forum: No registered users and 3 guests

cron