Sangoma-number divination

[jokerh1977]

Springbok,
Yes thank you. I'm learning a bit more now. I've done back testing (using Patrick's baby) and had very interesting results. I run Trips median due, then sort z-a. For the past 10 draws, if I had had chosen JUST the top trip only, I would have had 1 out of 5 numbers 5 times. Faaaar above what I should normally end up with statistically.

Now, I am in a unique position due to the "newness" of my 5/39 game. I tried placing the real draws on top of a clean sim-5 file as you suggested, but I don't wind up with anything usable. I guess for my game I just have to use the real draws. Not a problem though. You are the man that started this ball rolling. I hope you have been winning big on these!

Patrick,
What can I say but "good job"? You have really got a winner here, thanks for the gift. If, dare I say "when", I win I will gladly send a thank you gift to Springbok and yourself. Actually, you should post your paypal for those that would like to donate to the cause. Heck, I'll be the first to donate.

You have done so much work on this project so I'm hesitant to say this. Springbok started me thinking... what if your draw probability tab analyzed all the pairs AND all the triples and then broke that down into a number by number weight? Is something like that feasible? I'm thinking that the current draw probability button just analyzes the numbers in the pairs? As far as I've seen my triples are far more accurate than my pairs. Does this make any sense?

 

[Patrick123]

Hi jokerh1977,

Thanks for the vote of confidence,

The draw probability does do the weighting of all the triples as well. The first few columns are ball analysis, the next few, the weights for pairs, the following are weights for triples, and the last are the total overall weights. Not all columns may show but at the bottom of the grid you should have a scroll tab to see these hidden columns.

The Predict Numbers Tab, allows you to select a bunch of numbers and view only the details where those numbers are included. e.g.
From the Draw probability button, you sort the numbers by the weight of the 'Trip % N+x>=H' column, and choose the top 4 numbers.
You then go across to the Predict Numbers tab, and click those numbers in (the buttons turn red), now clicking on the 'Pair Analysis' or 'Trip Analysis' button, all the pairs or trips that include one or both of those numbers are now shown.
By sorting the appropriate columns, you can start identifying the potential 'opening batsmen' and 'fence-sitters' that could come up with those numbers.


Regards
Patrick

 

[barge]

Patrick123 Sprinbok

Hi Patrick123

Thanks for the latest update.
Do you have a feel for how many days a prediction might last? Or is it immediate (the next two draws), or would you like feedback concerning this from those of us using the prog?

Many thanks

 

[turtle0747]

Winnalotto Fla

Hi Patrick123

Have created a csv of FLA 6/53 and it loaded. Of course, now need a quick lesson. Am sure you have gone over this before. Can you create the steps I should go thru to come up with some numbers? Not too handy with computer talk.

Thanks
Turle0747

 

[barge]

turtle0747

Hi turtle0747,

The best way to understand the program is to follow this thread from the start in order to learn how it developed along the way. I don't think you can easily use it without having some idea of how it developed, and this is all here in the thread. Good luck!

 

[Patrick123]

Hi Barge,
I think possibly Springbok, may be able to answer your question easier than I, but what I've noticed, is that by noting your predictions on a draw by draw basis, you can see the numbers 'bubble up' to the top of the queue where they stay 'ripe' until they are drawn. Thus most likely any number that missed the draw, is likely to turn up within the next draw or two.

Hi turtle0747,
At this stage I must agree with barge, the program is still very dynamic with changes taking place on a day to day basis. I'll start putting together a basic instruction manual on the features that are now stable, but please don't 'hold your breath' as it will take a little while.

Regards
Patrick

 

[Patrick123]

Hi All,
Another addition to the program. I have removed the Analysis type from the import screen, and added the option on the Ball Analysis screen.

Here you select what Analysis value you want in the 'DP Perc' Column then select your analysis, the first three are basically described above. The new ones:
Rebal Perc - The percentage of times the ball rebalanced itself -> Last draw = 0 and the two previous hits were above the median.
Double Hit Perc - The percentage of times, the ball hit the next draw after it had just been drawn.

The best way to use them is to either sort (a->z) on the 'Last Draw' column then and compare the percentages or sort (z->a) on the 'DP Perc' column and check those with a zero in the last draw(skip) column.

These should allow us to home in on the potential 'opening batsmen' and 'fence-sitters'.

Regards
Patrick

 

[Patrick123]

Another three analysis types have been added:

Double <= Medn - The percentage of times there was a hit and the two previous skips were less or equal to the median.

Hit M Hit Perc - The percentage of times there was a hit then a skip above the median then a hit.

Rebal Median - The percentage of times the ball that should have rebalanced did not, but then hit on or below the median being reached.

Regards
Patrick

 

[jokerh1977]

Geez Patrick! Yer adding updates faster than I can download them.
Ok well good!
My bad though, I see now that the analysis took into effect the trips as well. Looking into the future I see that there will be a "master analyzer" that takes into account all the various statistical outputs you are adding and gives a summary per number.

To all those that are now joining Springbok and Patrick's quest. Please read all previous posts. You'll learn so much. Test out the program and then re-read if you have too. As Patrick has stated this is a dynamic idea/program that is morphing... the saga continues...

 

[Patrick123]

Hi jokerh1977,
Thanks, that's what I'm aiming at, but that's where I need everyone's assistance.

As you use the different methods, you might note that using this type of analysis and checking it against a certain type of scenario, you nearly always get a hit.
This is the type of feed back that I need, so that can put it into the program.

Not quite an update, but a bug-fix, where the Detailed percentages were not coming through. I've uploaded the fixed program.

Regards
Patrick

 

[Patrick123]

Hi Springbok,
Looking at the power of 10p, using a single number, would be difficult, but I think the power of 30p would make it easier.
Simply play the top 3 numbers that get predicted, chances are, one of the three numbers will hit, then spread the winnings over all three top numbers of each draw, repeating the process. Not such a quick jump to the top, but a slightly steadier one with each number supporting each other.

The other option is to 'lend' only 10p from the winning number to kick-start the losing numbers again.

Regards
Patrick

 

[Patrick123]

Power of 10p

Looking at my suggestion in the previous post, I would use the second options and I think it would be more frugal, only to play 50% of the winnings on the number, so that every time, even though the run does not have a winning streak of 6 numbers, it always produces winnings for you, if it has at least one win.

Regards
Patrick

 

[Springbok]

Hi Patrick
The 10p idea was a thought on sharpening up ones skills in single number picking using draadsitters, historical highs etc. You could could kick off with 5 single numbers and switch to the one that survives the first draw. Also the concept tickles my mind. Imagine someone interviews you and asked how you started your fortune. The reply would be "10p".

 

[barge]

Patrick123 Springbok

I loaded a dbase of 400 real draws, skipped yesterdays 2 draws, and got some remarkable results. There is a gap between the the first 19 or so hits, then a resumption of good hits further down the list. It doesn't seem to matter too much which sort method is used.....I would like to know if you can emulate this, and figure out a way of bringing all the hits closer together(!).

 

[Patrick123]

Hi Springbok,

Imagine someone interviews you and asked how you started your fortune. The reply would be "10p".

The 10p Millionaire!!! I like it.

Thanks barge, I'll have a look and see if I can track anything down that makes that pattern occur and try and bring them together.
I have a feeling that it's balls from the separate 'partitions' that are clumping together. e.g. two from the 'N+x>=H' section, two from the 'opening batsmen' section and the two from the double hit section. Something along those lines, and that is what we need to identify.

Regards
Patrick

 

[struxo]

Hi to all,
first thanx to Springbok for great idea and Patrick for great program.
Patrick, i wonder if you can make change in "Balls in draw". Now it's max 10 nos, can you make it to 20 so i can test it with Keno? Please!
Keep fighting folks, JP is just around corner

 

[Springbok]

Na Bloubul

Go to the link. You will see 49. click and you can copy and paste the results. Ek is stil in die "vreemde". Every day spent out of South Africa is a day wasted.I shall return soon to the beloved country.

http://www.bluesq.com/bet?action=go_lotteries

 

[Springbok]

Hello Patrick
I should rephrase the power of 10p, to the power of 20p. I have been running some simulations. Bringing into action the 6 number sets and 5 number sets can often isolate one or 2 numbers in the very next draw. Here is an example which I tested for lunchtime. These two sets were overdue:

alpha6(1) {1,2,3,4,5,6}
bravo6(6) { 1,2, 13, 20,37,40}

Notice the intersection at 1, 2. Number 2 popped up. It is better but more difficult to use trips but they can point to a single number. I think using different techniques and a bit of cunning could send 20p nuclear.

 

[barge]

Springbok

Rebalance won the day this pm, hits on the first six trips.
Guess who who used DPperc!

 

[Springbok]

To Patrick
Being exploring your software a bit more Discovering the setting of percentages for the trips. Now that is very very cool. You solved the problem of going through thousands upon thousands of trips to find what you are looking for. Most impressive. Well done.