Sangoma-number divination

[Patrick123]

Okay, the program will now accept text files as well, I've also added the 5/34 UK Thunderball text file to the zip file.

Once the file is opened, you are able to modify the check boxes on the 'Load Results' screen and check the draw probability as well as click on the 'Give Balls' option on the Predict Numbers screen to see the outcome without needing to re-import and analyse the info. All other settings require a re-import.

Regards
Patrick
ps. I'm still working on the save information (Alpha, Beta, etc) to a text file.

 

[Patrick123]

I now have the Save feature up & running on the Ball Analysis screen. It will save the currently selected option only. You have an option to remove/change the comma in the delimiter edit box to allow for a text file. The headers also gets added, so very likely you will need to remove this before importing into MDIED.

Regards
Patrick

 

[Patrick123]

I will start work on allowing the user to define their own Alpha/Beta pairings most probably on Wednesday only.

Regards
Patrick

 

[Springbok]

Hi Patrick123
Maybe hold on for a few days. I will be presenting ideas for the numbers in the last draw and numbers sitting on the median. Evaluate the ideas and see if they are worthwhile for inclusion in the programme.

 

[bloubul]

Hi Patrick123

Welkom hier by ons. Ek bly hier in Kaapstad. Kan jy my sê watter resolusie jou applikasie gebruik.
Dit lyk baie goed, kan nie al die skakels by kom nie.

Thanks

BlouBul

 

[Patrick123]

Hi Springbok,
I'll do so, I'll concentrate more on the neatening up of everything than the actual prediction side.

Hi Bloubul,
Dankie, ek het altyd gedog jy was hier in Pretoria. Die applikasie sal in 1024x768 hardloop.

Regards
Patrick

 

[Springbok]

Hi Patrick123
I am also experimenting with 3 number sets to hit within 2 drawings. So I use another series of alpha, beta etc . So far using sangoma with these groups of triplets looks very, very promising. A vast improvement over my previous crude attempts at using triplets. I am really amazed how numbers can talk. Using a mixture of 2 and 3 number sets could possibly bring 4 hits in a single line. But first we must have the power to hit one group in the very next draw. If you cannot hit one group then there is no chance of in getting to 2,3 or 4 strikes in the next draw. My next series of tests which I will post starting Wednesday will be a 2 number set to hit in the very next draw.

Hello Bloubul
Long time no hear from you. How are you doing with your techniques?

 

[Patrick123]

Hi Springbok,
You will see that by setting the 'Beta Balls' value to 3 and if the 'Max Alpha' value is still set as 50, you'd have 50 of these 3 number sets that you can view under 'Ball Analysis'.
By leaving the the 'Median Hit', 'Use Alpha', 'Use Beta' only checked, go across to Ball Analysis then Draw Probability, the balls that have values in their Alpha and Beta columns were weighted according to your N+4 > H method. I now want to add in a feature so that by clicking on them, you can get the groups that highlighted them.

Using this, the numbers 7 and 15 stood out above the rest if I analyse from 18th only for the 5/34 UK draw.
Analysing from the 14, the numbers 4 and 7 stood out.
Based on this, 14 and 15 now look like the High probabilities for the 25th.

As the ball sets are generated randomly at import time, I've noticed that now and again there is a false positive. By re-importing the same data about 2 or 3 times, only the numbers that stay consistently high, are the ones I'd recommend.

Regards
Patrick

 

[Patrick123]

Following this from about 4 or 5 draws back, one can actually see the numbers starting to form a queue to be drawn. The numbers 4 and 7 became quite dominant with 1,14,15,and 27 'bubbling under'. As you start reaching your final draw, The numbers 14 and 15 are now dominant with 3,27 and 31 'bubbling under'.

Regards
Patrick

 

[Springbok]

"the balls that have values in their Alpha and Beta columns were weighted according to your N+4 > H method"

Surely you mean N + 4 =< H

 

[Patrick123]

"the balls that have values in their Alpha and Beta columns were weighted according to your N+4 > H method"

Surely you mean N + 4 =< H

Yep, my apologies, my thoughts were ahead of my typing.

 

[Springbok]

Thinking about your programme. Do you really need to have the date aspect? As long as you know the draws are up to date why have a date. I tried to load the 49s into the programme but it does not recognise strings eg it cannot read jul(as in July). I think for pure convenience just loading the draws and nothing but the draws. Latest on top and going down in descending order.
The algorithm you gave had me worried. It could have lead to some unforseen results. According to what I read, a French Arianne rocket went off course and had to be blown up because in the computer programme someone had written (3, 4) instead of (3. 4) which sort of upset the looping in the programme that controlled the guidance system.

 

[Patrick123]

I agree, the date is nothing more than a convenience for the user, I'll give an option that it can be included or excluded.

Coming back to the n+4 <= H statement, the reason why I ended up with N+4 >=H is because I was checking using N = the last hit and H = maximum hit value instead of the 95% cut. I will re-evaluate my workings.

 

[Patrick123]

Nope, the way I'm understanding the prediction, is that if I take the last hit value n, and add 4 to it, I then get the maximum hit value, H, and if n+4 exceeds or is greater than H, I select it. This would imply take all the groups that are within 4 hits of their maximum hit value. Thus N+4 >= H is the correct naming. N+4<=H would imply ignore all values that are within 4 hits of their maximum value and take the rest.

Regards
Patrick.

 

[Patrick123]

Okay, I've modified the program. On the load results screen,
I have the N+4>=H and the N+4<=H option.
I also have a skip date checkbox, with this ticked, the program expects the input format of draw #, ball1, ball2,...ballx

As the exercise, I did the following:
Num Balls - 34
Balls Drawn - 5
Beta Balls - 3
Uncheck Use Delta
Uncheck Use Actual
Skip the first - 2- draws

Then import the 5/34 UK File.

Switching to Ball Analysis then draw probability, you can see 4 & 7 stand a cut above the rest. (4 came in, 7 the following draw)

There might be the odd false positive, but repeating the process two or 3 times, shows the consistent balls standing out.

regards
Patrick

 

[Springbok]

You are quite right Patrick123. It will have to be the greater than sign(>). When
I check each number I mentally calculate that the +4 to the current skip must at least be equal to the historical high or greater than which is not reflected in the original algorithm. Sloppy writing on my part in translating my thought processes into an algorithm. If I was puttting it into a programme I would have eventually spotted the error when I was cross checking. But when doing the calculation mentally I was actually doing it with the equal to or greater than scenario. Thanks for spotting the error.

 

[Patrick123]

Glad that we're on the right track again, as you say, describing it compared to actually programming it on an unforgiving computer are two different things.
Now we are ready for the next phase.

Regards
Patrick

 

[Excellence]

Good day gentlemen. I once stated that this system would clearly be a profitable one. I would wish to withdraw that, at least just for a time. The criterion presented by me (H – 4 <= n < H) it is not the criterion of this system. I think Springbok did not read this whit attention when he, kind of, approved it. In the light of new developments of Patrick’s software I have comprehended that the criterion is in fact: n + 4 >= H! I cannot explain why should we think that, a pair that already reached a large skip for it should come out in the next 4 draws!? This thing probably happens in most of the cases, I do not disagree, but the cases in where this thing does not happen it is not very small.
Also, if you decide to play you will need a money management. A 4 step double bet progression could prove unhealthy for your bankroll. For every progression lost, you need to win at least one, but mostly two or more others. It also depends the moment when you win. I don’t know if this system has a winning percentage that could cover the losses.
Well Springbok, before I present any system of mine I run a lot of test, so that I can have results! I would love to see what results had this system in the past, and since you use more powerful tools than Excel, this should not be a problem. A win/loss statistic would be very interesting, so as the actual profit for an x system of money management. This are the only things that support your system, since I cannot have any explanation why should a pair whit n + 4 >= H. I wish you the best off luck friend, but in my opinion, this system needs improvements, because in this form, I would hardly believe that it is profitable. Again I would be very glad that I was wrong.

 

[Springbok]

End of this round of testing:
Alpha10 hit 1st draw
Delta17 hit 1st draw
Delta21 hit 2nd draw
Foxtrot16 no hits(fail)
Foxtrot23 hit 4th draw

In the last 3 rounds 14 number pairs were forecasted. 11 hit within the required 4 draws and 3 failed. 79% strike rate.

 

[Springbok]

A strike rate of 79% was not particularly good when we were expecting around 100%. If you were betting on all those number pairs you would have made a handsome profit in the end though. So let us refine n + 4. All the pairs we pick with the n + 4 scenario are not equal. When I have given probabilities of hitting in the next draw the higher probability numbers are pretty accurate. I have picked these 3 pairs for the next round. They have the highest probabilities from a group of 6. This is what we do. We take the current skip and see how many times that skipped was reached in the history. Then we find out how many times the current skip turned around and hit next draw and divide that by the total amount of times the skip was achieved. For example take Charlie7.Current ongoing skip is 8. The skip of 8 was achieved 6 times in the last 200 draws. It turned around 3 times to hit in the next draw. 3/6 = 50%. So it has a 50% chance on hitting in the next draw.
Delta3 has a 71% probability and delta7 a 50% probability. So the preferred pair is delta3 only because of its probability to hit in tomorrow's lunchtime draw(25 july). If not then teatime. An eye should be kept on the other two pairs to see how they fare.


* charlie # 7 * Hits: 53 pairing {7, 27}

* Skips >>>>> 8 1 1 0 1 3 0 1 10 6 3 0 9 1 10 5 2 3 4 0 2 4 8 7 8 6 1 1 2 0 3 5 3 2 2 0 1 2 0 2 4 0 0 2 3 4 0 0 1 4 0 1 1 * Sorted Skips: 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 5 5 6 6 7 8 8 8 9 10 10 ** Skip Median: 2

* delta # 3 * Hits: 57 {16, 32}
------------------------
* Skips >>>>> 7 1 1 3 3 1 1 4 7 0 0 0 3 3 0 4 4 1 5 2 0 2 5 2 6 1 1 4 1 1 4 1 0 0 0 1 1 0 2 1 1 9 5 7 0 7 0 1 0 8 7 5 2 0 2 5 0 * Sorted Skips: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 4 4 4 4 4 5 5 5 5 5 6 7 7 7 7 7 8 9 ** Skip Median: 1

* delta # 7 * Hits: 51 {17, 29}
------------------------
* Skips >>>>> 10 2 2 1 3 0 0 12 10 2 6 0 5 0 1 3 3 1 5 6 0 1 1 0 0 1 0 1 1 0 0 1 9 9 3 0 1 1 1 2 3 11 5 2 8 2 1 0 6 0 0 * Sorted Skips: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 3 5 5 5 6 6 6 8 9 9 10 10 11 12 ** Skip Median: 1