Sangoma-number divination

[black prince]

Hi Springbok

You play a 5/30 game. Why? Is it not better to play a 5/25 game or even less. You should be able to delete numbers fairly safely. In the 5/34 part of the UK Thunderball I delete 10 numbers pretty safely. If you use Saliu's software, the Pot filters should handle this. You can use the zero filters in Winnalotto to help you delete numbers. What game do you play? I will download a data file and show you how.

Thanks for responding to my post. I actually have been eliminating 17 to 21
numbers for a 5/9 to 5/13 play. I'm now adapting filters that match my own
spreadsheet filters and using winnalotto for post predicting. I need to better
understand the winnalotto filter commands. If you could give me a few simple
examples and explain their logic this would further my effort. I want to use
FFG as a part of the filtering logic.

BP

 

[bloubul]

Hi black prince

May I have a copy of your spreadsheet, or if not, will you than post the lay out and the formulas please.
Thanks

BlouBul

 

[Springbok]

Goie dag, Bloubul

I downloaded the files. I was confused with RAR files. I thought RAR means Rhodesia African Rifles at first or it was RARE as in rare beef without the e. Eventually I checked on the internet and found out what RAR files are. I downloaded a RAR type installer and managed to open them. Will have a good look at them today.

 

[Excellence]

Hi jokerh1977,

In order to calculate the probability of a pair to hit you need to employ the Hypergeometric Scheme. I will write the formulae for Excel. The example is for 7 / 49.

(Combin(2 ; 1) * Combin(47 ; 6)) / Combin(49 ; 7) + (Combin(2 ; 2) * Combin(47 ; 5)) / Combin(49 ; 7)

For explanations:

Combin(2 ; 1) is obtained: 2 = how many numbers do we choose (in the case of a pair 2); 1 = how many numbers will be drawn from the numbers we choosed.

Combin(47 ; 6) is obtained: 47 = 49 – 2 (where 49 is the maximum number); 6 = 7 – 1 (where 7 is the number of balls being drawn).

Since we obtain 0,26778, using Pascal’s Distribution or the FFG you will find that you need 3 trials in order to obtain a probability of 0,60 ( the probability of a pair to appear within 3 draws), the first above 0,50. Therefore a skip of 2. Do not use this algorithm in pick3 games or something like that. For those one you need a different algorithm.

 

[barge]

Springbok

In the interests of research Skipping the first 4 draws,I loaded 50 draws from the 49s. Loaded a filter of 3 x 2/19 random wheel selections. Selected column filter High Skip,as it is less restrictive. I ran Pairs, then taking a leaf from your book sorted skip 1 from the top. On the visible grid (I didn't scroll down) there were 21 hits in the 4 skipped draws. Worth further research I think, although there was still a void, but maybe accounted for in this case by the filter.

Also, running the Hots filter, then Pairs, produced 4 Pairs. 3 each had a hit, the 4th was a double hit.

 

[Springbok]

Hi Barge

I am not sure what you are doing with the 2/19 random wheel. Please explain step by step what you do. I looked at Merseyworld and saw the wheel but cannot figure out what you are doing.

 

[Springbok]

Hi Barge again

This is my current approach to the 5/34 section of the Uk Thunderball game. There are two ways of approaching it through optimised spectrum picks. First is to delete 9 numbers and second is to delete 14 numbers. The first method looks like this-the numbers are for illustration purposes only.

1....2...3...4.....5
6....7...8...9...20
11..12..13..14..15
16..17..18..19..20
21..22..23..24..25

Two things can happen here. One winning number will come from each line or more frequently two numbers will come from one line and one number from 3 of the 4 remaining lines. One line will be redundant.

The second way is to delete 14 numbers.

1....2...3...4.....5
6....7...8...9...20
11..12..13..14..15
16..17..18..19..20

Here 2 numbers will come from one line and one number will come from each of the remaining lines. You take each line one by one and create 2 or three trips and run a filter to try and find the strongest trip This will(hopefully) reduce each line to 3 numbers. You then take a random walk through the lines creating 5 number combinations. Create as many as you want. In the end the combinations will be purged by Saliu's Pick532 and Past draws down to 2 or 3 combinations. First test yesterday's draw gave me 3 numbers in one combination. I was disappointed as I expected to win. I purged 5 combinations with past draws using a 150,000 D5 file which indicated that it would hit above the median on Wednesday.It did. In other words if any of my combinations occured in those 150,000 lines they were purged as in knocked out.To me purging is vital. Those people who indulge in full spectrum picks are crippled unless they purge with Saliu's software.
Notwithstanding my failure to win on Wednesday I am full of confidence for Saturday. Hope grows eternal within the human breast. It can be done and will be done.

 

[barge]

Sprinbok

Correction, my filter was 9 2/19 wheels.
I am experimenting, using the random wheels as a filter with various column filters. Highskips has produced the best results so far. Each 19 line wheel contains a number of potential winning pairs. With some testing this might increase our hit rate.
You were right about the Tball, I hit 4 out of 5, only 2 on one line unfortunately(of 5 lines), but I got the Tball, and a small profit. I will take a look at filters for the Tball, I am trying this lateral thinking thing

 

[barge]

Springbok

I used the 9 number reduction method for yesterdays draw as you suggested, the distribution occurred as you described. I will start purging next time......

 

[Springbok]

Hi Barge

When you write down the 5 lines have 3 pens handy-a red pen, a green pen and a blue pen. All numbers with a skip from 0 to 4 are written in red. All numbers with a skip from 5 to 9 are written in green and all numbers with a skip of 10+ are written in blue. If you take a random walk you do not pick exclusively from one colour alone. 3 to 4 of the numbers should be red, maybe one or 2 are green and maybe 1 is blue. A random walk through the lines is not totally random as random numbers have to obey certain rules.

 

[barge]

Springbok

Good idea, coloured pencils that is, my brain is too busy unscrambling pairs and trips plus filters to remember what was going on ten minutes ago!

 

[Springbok]

Hello Barge

I suffer from the same problem. Thinking about all of this I think in the end we will end up by being smarter people as a sort of spin off from "lottery studies". We learn strange things about statistics. I once contributed to a South African site and statistically worked out a person's chance of being murdered in South Africa vs a person living in Bagdad(Iraq). I was promptly banned from the site. unable to post there again.

 

[Springbok]

Delete numbers with care if you are using the anti filters, they are not gospel. just a couple from each. I am trying to work out how to find the redundant line in the five. Working with 4 lines causes less headaches and we rely on St Jude less.

 

[barge]

Springbok

All true,
Clearly the site which banned you were no brainers!

Someone mentioned having a method of predicting number repeats, we get a significant number, particularly on the 6/49s. If you read this, let us know your method please!

 

[black prince]

Median

Hi Patrick123

I’m doing some testing of past draws for a Pick5 (5/30) and I loaded
the history into Winnalotto. I compared the Median to Saliu’s UTIL532
FREQ5 report and noticed the Median is different between Winnalotto
and FREQ5 reports. Except for “0” as a current winning number your
program shows the Median as 1 greater then UTIL532. Here’s an example

Number………..Median (FREQ5)………Median (Winnalotto)

1……………………..4………………………….5…………….
2……………………..3………………………….4…………….
3……………………..4………………………….5…………….
4……………………..1………………………….2…………….
5……………………..7………………………….8…………….
6……………………..5………………………….6…………….

Which is accurate or is it possible you both are correct ?

BP

 

[black prince]

Median

Patrick123,

The Median is different from UTIL532-> FREQ5 report with many numbers
and the the same for a few. Is this correct?

BP

 

[Springbok]

Hello Black Prince

Your question on the median difference can be answered, I think, on where the count starts on the lines illustrated as follows:-
Saliu's count.........Winnalotto count
0...........................1
1............................2
2............................3
3............................3
4.............................5
5.............................6

So when you set say the max or min filter in Saliu's software you add +1. If the THREE median is 24 and you want to discount the previous 24 lines then you would set the filter at min 25-this takes into account the count starting at zero.

 

[Patrick123]

Hi Everyone,
I'm back from Lesotho for the weekend, but will be going there again on Monday until Wednesday.
I heard South African Lotto has started again. So I'm going to use the previous results that ended in I think Feb, as my starting base as I have no history. Hopefully time is not an influence in the way balls are drawn, but then it's the equivalent of me using UK49 as a base.

Barge, I remember you mentioning the system bombing with an access violation when you load historical draws. I was able to duplicate it. What seems to be the problem, is when you have the 'Skip first x lines' set to a value other than zero,(possibly you skipping the header row). As a work-around, delete your header row, and ensure that it is set to zero before loading the historical info. I'll try and get it sorted out as soon as possible.

Thanks for the reply to BP there Springbok. In many ways, you are more of an authority of the WinnaLotto SE than what I am

Regards
Patrick

 

[Springbok]

Ndaa Patrick

Vho vuwa hani?
I am familiar with the count with Saliu's statistical reporting. Some programs start at 0 and others start at 1.

So the South African lottery will restart. The question is are the previous draws valid for the reporting on lottery programs. To most lottery programs out there it would make no difference as they are useless But to Winnalotto SE and Saliu's software the question is vitally important. I think there are two answers to this question. One is right and one is wrong and I do not know which.

The first answer is that it would make no difference. Logically the continuum created by the previous draws will continue despite a 48 draw break. How would the machine know there was a break? After all it is a dumb piece of fancy plastic with the balls whizzing around it

The second answer is the most troubling as it defies the logic of the first answer. A lottery machine or machines sets up a morphic field(a la Rupert Sheldrake) and creates a certain continuum specific to that lottery. The continuum creates the morphic field which somehow "communicates" to the machine so that continuum carries on for that specific lottery. If this is so then would the 6 month break have broken this specific continuum and a new one would have to be started or will the previous continuum carry on. There is experimental evidence of this kind of "field".

I will take part in the SA lottery through the internet but will wait a few draws to see whatsup. Hell, we have enough problems with lotteries without having to worry about this one

Anyway, I will try and contact Sheldrake and ask his opinion and report back.

 

[Patrick123]

Ndaa Springbok,

Yep, that is the most important question. To me though, will the 4th dimension (time) have any impact on the results?
Until I get enough draws to analyse the numbers on their own, I'll have to make do with make do with SA's old draws as my statistical base.The experiments that you did using random drawings, gave relatively good results for the n+4>=H selection.
What the hang, I isolated 11 numbers, put them in a wheel that I generated, and will check the results. I may as well start getting the 'feel' of the SA numbers.