Pushing Numbers To Top Occurrences

Postby lottoboy » Thu Jan 22, 2009 5:30 pm

[quote=Jerzy]
lottoboy,

I am glad that you found the reason for losing the JP ticket and that you know how to use the HD filter.

But you should try to filter all possible rows in all columns. However, I need to say that the rows with 1 or 2 numbers in them (any numbers, not necessarily winning numbers) cannot be filtered with Numbers in a row setting 0 to 2. It doesn't make sense and it will not result in removing any combinations from the package. And the rows with 1 number in them cannot be filtered at all with this filter. You cannot enter Number in a column 0 to 0. The filter doesn't allow for that. I hope that Stan will remove this limitation.

This is one of the best EL filters. Of course, like with other filters, we are running a risk of losing the JP ticket if we enter incorrect settings, but some tickets with lower prizes should remain in the package.
[/quote]

Jerzy,

Thanks for your sincere reply! But could you let me know how can I "try to filter all possible rows in all columns" for details instead of filtering the 1st row only?

Also, I'm not clear your means about "the rows with 1 or 2 numbers in them (any numbers, not necessarily winning numbers) cannot be filtered with Numbers in a row setting 0 to 2. It doesn't make sense and it will not result in removing any combination from the package. And the rows with 1 number in them cannot be filtered at all with this filter. " Could you give
me an example for those in details? Thanks.

Best,
lottoboy
lottoboy
 
Posts: 738
Joined: Sat Dec 08, 2007 8:37 pm

Postby Jerzy » Fri Jan 23, 2009 2:57 am

lottoboy,

I edited my last post. One number rows can be filtered with HD filter if you use Remove and Numbers in a single column 1 to 1. And two number rows can be filtered with Numbers in a single column 0-1, Leave, or 1 to 1 Remove, or 2 to 2 Remove.

But if you try to filter one number rows with 0 to 1, Leave, or two number rows with 0 to 2, Leave, all combinations will be passed, nothing will be filtered out, because all combinations targeted by the filter will be within the passing range of the filter. Think about it. It is simple and obvious.

To filter any row in the column, make sure that the column is entered correctly and place the row number in the History Differences boxes. For the row 2: 2 to 2, for the row 3: 3 to 3 and so on.
Jerzy
 
Posts: 811
Joined: Mon Sep 18, 2006 5:26 am

Postby lottoboy » Fri Jan 23, 2009 6:15 am

[quote=Jerzy]
lottoboy,

I edited my last post. One number rows can be filtered with HD filter if you use Remove and Numbers in a single column 1 to 1. And two number rows can be filtered with Numbers in a single column 0-1, Leave, or 1 to 1 Remove, or 2 to 2 Remove.

But if you try to filter one number rows with 0 to 1, Leave, or two number rows with 0 to 2, Leave, all combinations will be passed, nothing will be filtered out, because all combinations targeted by the filter will be within the passing range of the filter. Think about it. It is simple and obvious.

To filter any row in the column, make sure that the column is entered correctly and place the row number in the History Differences boxes. For the row 2: 2 to 2, for the row 3: 3 to 3 and so on.
[/quote]

Jerzy,

Thanks for your sincere editing and detail answer!

Next, I will make some back-tests (the last 50-100 drawings) for finding reasonable settings of Numbers in a single column to my game in statistical. Before doing those, I want to confirm the tasks of my back-tests as below:

1) The typical count of winning #s of the first row in each col.;
2) The typical count of winning #s of each row in each col.;
3) The typical count of winning #s of each segment in each col.

Is it correct? If something has been lost or overdone??

In addition, I think we can design a plugin to finish those statistical tasks above by using program. How about your opinion? Thanks.

Best,
lottoboy
lottoboy
 
Posts: 738
Joined: Sat Dec 08, 2007 8:37 pm

Postby stan » Fri Jan 23, 2009 7:29 pm

[quote=Jerzy:1232549487]
Stan,

I believe that it would be very useful if you changed the lowest possible value of Numbers in a single column in the History Differences filter from 0 to 1 to 0 to 0. There are rows in columns with no winning numbers, and the best results with the History Differences filter can be achieved when each row in each column is filtered separately.

Could you do this for EL version 5 and for current version as well?
[/quote]

will do
Expert Lotto Team
User avatar
stan
Site Admin
 
Posts: 6338
Joined: Thu Sep 23, 2004 1:01 pm

Postby Jerzy » Sat Jan 24, 2009 2:51 pm

lottoboy,

We have to be careful when we are using the History Differences filter. This is a very promissing filter, but it can fail if we use wrong settings. But this applies to all filters.

Use the Back-test function of this filter for the 1st row in each column. The Numbers in a single column 0 to 2. In my 6/45 game, for a single column, this setting was correct in 95.53% to 97.53% of drawings, depending on the column. Of course, combined reliability in 11 columns would be approximately 80%. This is at least satisfactory. But if we filter additional rows in the same way, the combined reliability will be even lower.

However, all filter that we use cannot be set to be 100% reliable.

We have to be careful when remowing rows which contain only single numbers, because many of those single numbers are next draw winning numbers. And sometimes there are 3 or even 4 winning numbers in a row, particularly in a long row.

Run the Back-test in this filter for the 1st row and other rows of your interest in each column. The Back-test will show you a list of past drawings. And you will see the drawings in which the filter failed. To go back to these drawings, skip appropriate results in the Winning Numbers and open the WNH History Differences tab. Click on Update button. Go to the column in which the filter failed and see why it failed.
Jerzy
 
Posts: 811
Joined: Mon Sep 18, 2006 5:26 am

Postby lottoboy » Sun Jan 25, 2009 5:21 am

[quote=Jerzy]
lottoboy,

We have to be careful when we are using the History Differences filter. This is a very promissing filter, but it can fail if we use wrong settings. But this applies to all filters.

Use the Back-test function of this filter for the 1st row in each column. The Numbers in a single column 0 to 2. In my 6/45 game, for a single column, this setting was correct in 95.53% to 97.53% of drawings, depending on the column. Of course, combined reliability in 11 columns would be approximately 80%. This is at least satisfactory. But if we filter additional rows in the same way, the combined reliability will be even lower.

However, all filter that we use cannot be set to be 100% reliable.

We have to be careful when remowing rows which contain only single numbers, because many of those single numbers are next draw winning numbers. And sometimes there are 3 or even 4 winning numbers in a row, particularly in a long row.

Run the Back-test in this filter for the 1st row and other rows of your interest in each column. The Back-test will show you a list of past drawings. And you will see the drawings in which the filter failed. To go back to these drawings, skip appropriate results in the Winning Numbers and open the WNH History Differences tab. Click on Update button. Go to the column in which the filter failed and see why it failed.
[/quote]

Jerzy,

Thanks for your truthful advice!
Still one question, how many back-tests did you make for the strategy?
Thanks.

Best,
lottoboy
lottoboy
 
Posts: 738
Joined: Sat Dec 08, 2007 8:37 pm

Postby Jerzy » Sun Jan 25, 2009 5:43 am

lottoboy,

So far, because of time limitations, I back-tested only for the row 1 in all columns. Actually, the back-test function in the History Differences filter works fast, but if you do it a lot, it takes a lot of time.

I need to back-test for other rows, in the 1 and 2nd segment in particular, and I plan to see all History Differences tabs for the last 10 - 20 draws to get a better feeling and understanding of this filter.

The back-test in this filter gives you information on how many drawings in the game history would pass or fail with the given filter setting. The % of drawing that would pass seems to be the most valuable piece of information. And you can see on the list which drawings would fail and, if you want, you can analyse why it happened.
Jerzy
 
Posts: 811
Joined: Mon Sep 18, 2006 5:26 am

Postby Jerzy » Sun Jan 25, 2009 12:36 pm

lottoboy,

The rows longer than 8 numbers should be given special consideration. In my game such rows can contain no winning numbers at all, but sometimes they contain 3 or even 4 winning numbers. To avoid unnecessary risk and disappointing results the Numbers in a single row 0 to 3 could be a better setting than 0 to 2 for such rows when you are using the HD filter.
Jerzy
 
Posts: 811
Joined: Mon Sep 18, 2006 5:26 am

Postby lottoboy » Mon Jan 26, 2009 6:11 pm

[quote=Jerzy:1232879805]
lottoboy,

The rows longer than 8 numbers should be given special consideration. In my game such rows can contain no winning numbers at all, but sometimes they contain 3 or even 4 winning numbers. To avoid unnecessary risk and disappointing results the Numbers in a single row 0 to 3 could be a better setting than 0 to 2 for such rows when you are using the HD filter.
[/quote]

Jerzy,

Thanks for your sincere experiences about back-tests!
I will do my back-tests for my P-6. Anything about my results
will tell you later.

Best,
lottoboy
lottoboy
 
Posts: 738
Joined: Sat Dec 08, 2007 8:37 pm

Postby lottoboy » Fri Feb 06, 2009 7:53 am

Stan,

I met three problems when running the Filter of HD in WNH:

1) The results of both left tickets in the Package and the prize are very different or in the opposite direction in taking the setting of HD (rows) separately and directly. For example, the setting of HD such as 4-4, 5-5, 6-6, 7-7, and 8-8 separately and the setting of HD such as 4-8 once. Why?

Note: a) The other settings such as # count (0-6), Even # (0-6), Colunm
(-5), and #s in a single Col.(0-2) are same for both runnings
above;
b) The differences happen in both status: Leave & Remove too.

2) The final results of both left tickets in the Package and the prize are very different or in the opposite direction in taking the different setting of Column (0- -10) separately and totally. For example,the final result is A after ticked the setting of Column from 0 to -5 separately but the result is B when ticked the all settings of Columns from 0 to -5 in combined once in the same other settings. Is it correct?

3) There are 4 interesting results in my back tests of Pack-6:

For felting 0 Column, (I'm not sure what's happen in other columns)

a) The setting of HD: 7-7, the setting of #s in a single Col: 0-2;
47 tickets filtered out , The prize: 0/0/12 (JP/2nd prize/3rd prize);

b) The setting of HD: 7-7, the setting of #s in a single Col: 0-3;
5 tickets filtered out . The prize: 0/2/21;

c) The setting of HD: 7-7, the setting of #s in a single Col: 0-4;
0 tickets filtered out;

d)The setting of HD: 7-7, the setting of #s in a single Col: 0-5;
16 tickets filtered out , The prize: 1/2/22.

But there are 4 winning #s only in the total 5 #s on the 7th row when I checked the Tab of HD for the last drawing. Why?

I'm afraid that the running results above are abnormal.
Could you help me to check those? Thanks.

Best,
lb
lottoboy
 
Posts: 738
Joined: Sat Dec 08, 2007 8:37 pm

Postby lottoboy » Tue Feb 10, 2009 5:42 pm

Stan,

Could you give me an answer for my post above?

Also, the settings of both # count & Even #s seem in the filter of HD not working. For example, it's nothing for the running when I changed the setting of # count from 0-6 to 1-6, 2-6, 3-6, ...6-6 or 0-1, 0-2,.... the same things is in the setting of Even #s. Pls check those. Thanks.
lottoboy
 
Posts: 738
Joined: Sat Dec 08, 2007 8:37 pm

Postby stan » Sat Feb 14, 2009 9:26 pm

the filter results depend on 'history differences' range. if history difference of a ticket number is outside this range then it isn't evaluated at all and other filter conditions like 'even numbers' or 'number count' aren't considered for this number. that is the case for problems 1) and 2) in your post above

problem 3) - sorry, but i don't know what's going on there. perhaps you should check other filter conditions (even numbers and number count)
Expert Lotto Team
User avatar
stan
Site Admin
 
Posts: 6338
Joined: Thu Sep 23, 2004 1:01 pm

Postby lottoboy » Sun Feb 15, 2009 1:50 am

[quote=stan]
the filter results depend on 'history differences' range. if history difference of a ticket number is outside this range then it isn't evaluated at all and other filter conditions like 'even numbers' or 'number count' aren't considered for this number. that is the case for problems 1) and 2) in your post above

problem 3) - sorry, but i don't know what's going on there. perhaps you should check other filter conditions (even numbers and number count)
[/quote]

I still not clear what's the means about "if history difference of a ticket number is outside this range then it isn't evaluated at all and other filter conditions like 'even numbers' or 'number count' aren't considered for this number." Could you give me an example for it? Thanks.
.
lottoboy
 
Posts: 738
Joined: Sat Dec 08, 2007 8:37 pm

Postby lottoboy » Sun Feb 15, 2009 2:05 am

Stan,
my problem 3 is still for the filter of history differences.
the settings of both # count & Even #s in the filter of hd
seem to work not at all or don't have any use.
the running results don't have any change in any event of
changing their settings such as from 0-6 to 1-6, 2-6, 3-6, ...6-6 or 0-1, 0-2,....
lottoboy
 
Posts: 738
Joined: Sat Dec 08, 2007 8:37 pm

Postby m4rv1n » Sun Feb 15, 2009 12:45 pm

lottoboy,

#count and even # count counts #s & even #s of all column in the selected HD range.
m4rv1n
 
Posts: 51
Joined: Sat Jan 26, 2008 7:26 pm

PreviousNext

Return to Tips + Tricks

Who is online

Users browsing this forum: No registered users and 1 guest

cron