Sangoma-number divination

Patrick123

Patrick123

Member
Hi Springbok, thanks I managed to download, the file.
Using all 3850 drawings, balls 3, 8 and 46 stood out with the 3 & 8 pair at 71% ( current skip 23 - all time high of 25).
Using 500 drawings I get the following coming through:

3 at 61%
6 at 47%
8 at 62%
9 at 59%
15 at 53%
27 at 28%
29 at 60%
31 at 31%
32 at 53%
40 at 61%
46 at 81%

Regards
Patrick
 
S

Springbok

Member
Hi patrick 123
I am glad you are using the 49s. It is fast moving. Also you can compare your predictions to mine. You don't need all the past draws just a hundred or so. There are errors in the blue square history but as long as they are not in the last hundred or so drawings that is ok. Always update on win-on-numbers. They are reliable. ARe you predicting according to n+4 (mark1) or using(mark2) to find the strongest pair?
 
Patrick123

Patrick123

Member
Hi Springbok,
Yes I'm using the mark2 version. I'll use a trimmed down file though I've noticed the more lines I use, the less noise(false positives) I tend to get. I want to test and see what is the optimum minimum number of lines to use as well as what impact the bonus ball may have on the predictions.
Regards
Patrick
 
Patrick123

Patrick123

Member
Using 500 draws, including the Bonus Ball:
Ball - Alpha - Beta - Delta - Ave %
----------------------------------
6 38 44 38 41
9 47 47 45 46
15 73 73 78 75
17 30 30 30 30
20 52 54 52 53
27 51 56 46 52
29 33 36 36 35
31 44 52 44 50
32 40 43 40 41
40 63 80 66 68
42 60 60 -- 60
44 -- 36 36 36
46 40 45 49 45
49 30 29 26 29

I used 80 sets with all three Alpha, Beta and Delta set to analyse pairs only.
the -- is where the program never received any value.

To complete the picture, I analysed all the draws, only 9 and 40 stood out at a skip of 17 max of 20.
Regards
Patrick
 
S

Springbok

Member
I always backtest as much as possible(time permitting). For example I have about 230 real draws on the 49s. I cut the latest 100 and put into a result file. I test an idea on the next 4 drawings starting from the bottom. Then cut them and place the bottom 4 on top of the real drawings. Test and then repeat the process and so on. The initial parameter is always within 4 drawings. You start off with a theory and you test it and see what happens. Actual real time results are always superior to theory. There is no doubt that number pairs are influenced by their history. I will say that 80% of number pairs are influenced by their history at a given moment. Why number pairs and not single numbers? They leave a bigger footprint or blip on the record. A triple group has a bigger footprint, a quadriple has an even bigger footprint. I will expand these ideas when we examine number pairs sitting on their median and hitting in the last draw to evaluate their chances in hitting in the next draw. Anyway, let us await the results of today's draws and see whatsup.
Although I am using 200 draws, there is the idea of short term trends say using only last 50 draw which I may look at in the future. Gee whiz, these things can get complicated. Fullhouse's original grid in comparison is nice and simple.
 
Patrick123

Patrick123

Member
I agree with you, that is why I added the 'Skip the first X then down load the next Y results' in the program, it saves a bit of time cutting and pasting. As you increase from numbers pairs to trips, quads (the program goes up to 7's) much more shows up on the radar, and the figures and percentages come closer together, the secret as you say is then to isolate the ones that really stand out.
 
S

Springbok

Member
Had a quick check on the lunchtime result. Omega16 and kilo23 did not hit. This has me worried as we enter unknown territory with these two fellows. Have they broken free from the past and run wild and free? If I had more draws I would check their history over the last 300 or 400 draws to get an indication of their future intentions. If I was betting on these two I would seek further confirmation of their true intentions before I laid down any money. Maybe its time added more draws to my real number file.
 
Patrick123

Patrick123

Member
Hi Springbok, yep, 46 did me proud, and relating to the Bonus ball equivalent, 15 showed its colours, as to whether this was beginners luck, I hope not but rather that mark2 is starting to shine.

Regards
Patrick
 
Patrick123

Patrick123

Member
Ran through details for teatime(highest values only):
3 = 69%
8 = 50%
9 = 51%
40 = 58%

1200 draws - 80x3 number sets - pairs only ( thus equivalent to 240 sets)

Regards
Patrick
 
S

Springbok

Member
I downloaded 800 draws from the blue square database. Had problems with the accuracy of blue square. Lines 128 and 638(thereabouts) are missing a number. Check the latest 15 lines against win-on-numbers. On 22 July bluesquare has a combination that was never drawn as it is not reflected on win on numbers. Blue square has a different combination. Delete the false combo and put in the correct one from win on numbers.
 
Patrick123

Patrick123

Member
Thanks, I'll do so, I've laughed off evaluating the numbers with the bonus ball included, I've found there are about 15 odd tainted results there, but have not compared actual draw values.
 
S

Springbok

Member
The recommended kilo23 is out of it. It should not have been recommended according to my updated report. It had already exceeded its 200 day historical high and was not on it yesterday evening. It is a pity blue square cannot be more professional in compiling their data base. It is not rocket science to keep it accurate. The inaccuracy affected kilo23 badly. I am thinking of running a simulated set of say 1000 draws and putting the last 30 real draws(and verified) on top and see what happens.
 
S

Springbok

Member
This is the updated report for omega 16 based on the previous 400 draws. I hope it is accurate. It should hit teatime today. 100% probability.

* omega # 16 * Hits: 115 (31, 32)
------------------------
* Skips >>>>> 11 0 2 3 7 1 3 2 2 0 1 0 2 6 0 1 3 2 0 2 1 0 0 0 2 3 5 1 0 8 6 2 1 0 0 5 1 0 6 0 3 1 5 0 3 4 0 8 1 0 4 5 5 3 0 1 7 3 2 0 1 1 2 7 0 3 0 7 8 0 3 5 0 3 5 4 2 5 2 2 4 0 0 1 2 2 0 2 5 11 1 7 0 0 1 0 1 6 1 0 3 2 2 0 1 1 2 2 0 1 4 7 4 0 6

* Sorted Skips: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 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 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 4 4 4 4 4 4 5 5 5 5 5 5 5 5 5 6 6 6 6 6 7 7 7 7 7 7 8 8 8 11 11

** Skip Median: 2
 
S

Springbok

Member
You hit a number in the last 2 draws. No luck there, just hard facts produced by the computer. We don't guess. Leave that to the statistically unwashed. I was amazed that omega16 did not hit teatime. First time in 400 draws it broke the historical barrier. Check the report of 600 draws. It has got to hit lunchtime tomorrow. If it does not I will be tempted to eat my laptop and desktop for supper and you can call me Suzi Mtetwe:)

* omega # 16 * Hits: 165
------------------------
* Skips >>>>> 12 0 2 3 7 1 3 2 2 0 1 0 2 6 0 1 3 2 0 2 1 0 0 0 2 3 5 1 0 8 6 2 1 0 0 5 1 0 6 0 3 1 5 0 3 4 0 8 1 0 4 5 5 3 0 1 7 3 2 0 1 1 2 7 0 3 0 7 8 0 3 5 0 3 5 4 2 5 2 2 4 0 0 1 2 2 0 2 5 11 1 7 0 0 1 0 1 6 1 0 3 2 2 0 1 1 2 2 0 1 4 7 4 0 6 3 13 2 0 2 0 2 9 0 0 7 1 0 3 0 3 4 4 0 0 3 0 2 2 0 5 1 0 9 0 5 3 1 1 6 7 1 5 7 1 4 4 7 1 1 6 4 10 0 1

* Sorted Skips: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 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 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 5 5 5 5 6 6 6 6 6 6 6 7 7 7 7 7 7 7 7 7 7 8 8 8 9 9 10 11 12 13

** Skip Median: 2
 
E

endbox

Member
Hi all

Been reading your threads with interest..............Even though I'm familiar with Skips .exe , I've little knowledge of converting real drawings into the file fragments you mention
Could you tell me where I can obtain a program that will do this ?
Thanks , endbox
 
S

Springbok

Member
The conversion is done by a computer programme. At the moment I have 14 files as in alpha, bravo etc. Each file is called and the real draws are converted into the required format(ie alpha, bravo, charlie etc). The output is sent to Mdied where I analyse the results in the stats programme which gives an ouput which you can see for the pairs I am looking at. So you will need a programme to do this. You could download a free language like Visual Basic Express, Just Basic, Q basic, Python. Then you will have to learn it and write the programme. Or maybe a nice person out there would like to write it and post it on this board. Or you could explore Patrick123 's application which you can download. His method is slightly different from mine. But the fact is for the technique and method I use the real drawings have to be converted into the different formats. If this is not done it will not work.
 
Patrick123

Patrick123

Member
Hi Springbok, my numbers are in agreement with yours, I'm getting:
29 at 87%
8 at 73%
31 at 71% and
32 at 64%.

I am thinking of running a simulated set of say 1000 draws
Click to expand...
I've got a bit of a problem with this, theoretically we are using the same set of numbers, but these would have different characteristics though is would become less noticeable with the 300 draws on top but it might still be enough to influence the true behaviour of the numbers.
Regards
Patrick
 
Patrick123

Patrick123

Member
I forgot to correct my 22 July draw: here is the updated values.
6 - 62% - 40 of 240 hits
8 - 83% - 10 of 240
9 - 75% - 35
15 - 43% - 13
17 - 70% - 27
27 - 56% - 11
29 - 78% - 30
31 - 60% - 29
32 - 59% - 19
37 - 59% - 17
44 - 70% - 7

I would assume that the frequency of hits would also be an indication of the confidence level.

The definition of hits here, is how many times my program identified the number had the right criteria.

Regards
Patrick
 
You must log in or register to reply here.

Sidebar

Top