Sangoma-number divination

E

Excellence

Member
Well 32 was drawn. I noticed that if you analyze the pairs and sort them by the probability you would have obtained 6 pairs whit 100%. The first two whit the highest Median Due where 41-43 and 32-41, in this order. Strangely 41 appeared in 4 of this pairs, but was not drawn. 32 on the other hand appeared in only 2 pairs: 32-41 and 5-32. I believe we should choose only the first two pairs whit the highest Median Due.
I also noticed after the draw that, in the Draw Probability table we have a column: Pair Rebal. If you had sorted the numbers after that column, in the first 8 you would have obtained 3 numbers: 1, 37 and 49.
Patrick, will be nice to have in the trips analysis the possibility to select two numbers in a trip, instead of only one, as it is now.
 
Patrick123

Patrick123

Member
Yes Springbok, you will find Excel to be quite a powerful program for this type of work. I saw 23 came in as the booster. As you discovered Excellence, there does seem to be a bit of correlation coming through. This is where I'm trying to match the original pair {32,41} against all their pairs and triplets. This then gives me a number or two which I then repeat the process until I have a complete chain of numbers. By doing this I'm hoping to build a sort of tree structure with each branch having a probability for the numbers to come in.
I'm sure I can add an extra filter so that we can select 1 or 2 numbers in a trip.
 
Patrick123

Patrick123

Member
Well I took {15,41} as my initial pair.
41 then links into {19,30,36} which then links into {13,25,31,38,48}.
15 short-cuts and links into {13,25,38,48}
I ran out of time, to follow the next links, hopefully I can get this automated if this shows promise.

Regards
Patrick
 
S

Springbok

Member
Hi Patrick. I was not sure which 2 of the top pairs would hit lunchtime, so I bet on all 3 of the common numbers which would guarantee a win no matter what. Just got back from the library with one of those shiny books on Office which has a section on Excel with nice pictures to illustrate. Will swot it up this evening. Excel looks nice on the screen. I also have it on my ultra smart phone which of course I have never used- excel that is. Maybe I can check out trips and pairs on the Tube:) It seems that we have a new algorithm as n + H = WIN . Here is a point to ponder-when the South African lottery resumes would we use the history from the previous lottery?
 
S

Springbok

Member
Hi Patrick. Got data files to load and actually run. A quick study of the nice shiny book explained clearly what to do. Move the data block two places to the right. Auto fill the first column with ascending numbers save and voila it loads. I assume I could fill the first column ( column A) with any old rubbish to fool the program that there is something there. The ignore date works well even though I do not record dates in my files. So I assume it automatically skips column B. Hopefully that is all I need to ever know about Excel and can return the nice shiny book back to the library tomorrow. Running a check of what the program recommends is identical to my program with pairs. Will check the recommended trips with my files tomorrow.
 
Patrick123

Patrick123

Member
Hi Springbok, Good to see that you got into grips with Excel, for manipulations of block or columnar information I use it quite religiously. Regards the South African Lotto, me feeling is to discard all statistical info built up & treat it as a brand new Lotto starting up, but it would be interesting to see if it continues/builds onto the stats & trends of the old Lottery.

I've added in the two features of the program, a 'No Key Field' check box, so you can have either one or both checked.
On the Ball Analysis screen, I've added in the extra number filter for the Trips Analysis. By leaving both numbers on the same number, will then give you all the combinations for that one number.

Regards
Patrick
 
S

Springbok

Member
Hi Patrick, good idea to give the option of leaving out the key field. It has no value. I wonder how sangoma would treat the digit lotteries which seem to be popular in America. Three or four slots and digits from zero to nine randomly drawn. Your program would have to include the zero.
Thinking again about the South African lottery. Assume the first draw takes place. There would be a gap of many weeks between the last draw in March. Question is, would this make a difference? Originally the SA lottery was held on a weekly basis and then went to twice a week.Would this have made a difference. A gap of 7 days vs a gap around 3 days? I have a gut feeling that the time gap would make no difference. The machine which spins the balls around and then shoots them out cannot have the "knowledge" of this time gap. Can it? If it does then there is something funny going in the world which we don't know about. I will run some tests on this in the weekend to see what happens. This question is starting to fascinate me.
Speaking of Excel, sure it does the job but its weakness is that it is a Windows program. Windows programs are slow. When you are analysing millions of combinations you must use good old DOS. Nothing can touch it for speed. To do this you avoid Visual Basic and go to QuickBasic or any language that runs directly in DOS. I will run a test. I will time what it takes to produce 7 million combinations in DOS and 7 Million using a Windows program. The difference is staggering.
 
Patrick123

Patrick123

Member
Hi Springbok, I agree, Windows has way too many other things going on all the time that steal the cpu cycles. I'm also waiting with bated breath for the SA lottery and if stats extend from the existing history, well then we can jump right in. Regarding the time gap, I agree, it is a fascinating theory whether the fourth dimension of time relates to the randomness of chaotic order, way past my levels of philosophy.
I've added in the Prediction section on the Program. At this stage it does not predict numbers, rather you can select the numbers that you are interested in, then look at those specific double, trip combos along the line of the Post-draw analysis. This should allow us to start picking up any relationship patterns with the other numbers.
Regards
Patrick
 
S

Springbok

Member
I have often looked the Florida lottery(5/36). Vernons runs bets on this. I have never ever succeeded in downloading their past draws. Seems to be done in a way to prevent this as far as I can see. Have a look and see if you have an idea on this. I have found Excel very useful in downloading the SA past draws. Piece of old takkie compared to my previous methods. Gee Whiz! learning how to manipulate Excel yesterday in 30 minutes is the steepest and fastest learning curve I have ever ascended.
 
Patrick123

Patrick123

Member
I'll look at the Florida (5/36) and see what I can come up with. That's the one thing of windows & thats the convenience factor for such things. I updated the program again, the sort feature was not working on the Prediction screen. Necessity is the mother of all learning:)

Regards
Patrick
 
Patrick123

Patrick123

Member
Okay, Back to the UK 49's.
The main numbers of interest here are {5,18,41,43}. Pairing & tripping them with the other numbers, I got {2,3,4,10,11,12,38} as having similar feel/values on the Predictions screen.
Regards
Patrick
 
S

Springbok

Member
Thanks for the Florida download. How did you do that? I would leave the lunchtime draw well alone. Too many likely candidates to profit from. I have put them into 8 pairings and saved the Mdied history to try and learn what the history is saying after the draw. When there are too many possibilities its probably best to hold fire. The fact is only one of the numbers will come in, even 2 on freaky occasions. There is another way of probably getting some idea and that is by using the ALL or POT filter in Lotwon. But the 632 program does not cover 7 number draws.
I am also thinking of looking at the Daily draw(7/28) on the uk lottery. I can do that very rapidly with your program as it would take some time to create alpha, beta files on it.
 
Patrick123

Patrick123

Member
Pleasure Springbok. I Selected the HTML format page, Selected all, copy & pasted into UltraEdit, there I selected one funny character, copy & pasted into search-replace and replaced with a space. Saved it as a csv file then opened it under Excel. Under Data conversion, I did a text-to-columns conversion settings - delimited by space and other character '-'. Date as mdy format. Then Cut the three columns & pasted them under each other, finally sorting by date and trimming off and junk (page headings).
I agree there are many possibilities in this draw, what I'm aiming to achieve is to see what other balls are starting to 'talk' to the highly probable numbers, thus building up the list of numbers that could come in in a single draw.
Regards
Patrick
 
Patrick123

Patrick123

Member
I've now added a percent filter on the predictions screen. Looking at the post-draw analysis, you can see on the LD Pairs Inclusive, the doubles N+x>=H range is about predominantly 22-28% and on the trips, 33-41%. By setting the appropriate range, we've eliminated most of the no-shows, on the pair & trip analysis grids (the filter only works on these two).
Regards
Patrick
 
T

turtle0747

Member
Fla 5/36

Hello

I too play FLA 5/36. I have downloaded your history file. Can you tell me what program you run it against.

Thanks
Turtle0747
 
S

Springbok

Member
Hi Patrick, so it was 43 that came in. I have not finished my conversations with the 8 number pairs. Will report back when I have finished. I have being playing around with trips on your program. WOW! they come at you like the hordes of Ghengis Khan. So many at 100%. I am not fully into the trip parameters on your program. Fact is trips are reasonably easy to hit in the next draw. If you take say bravo file and kick out two no hopers and ask granny to stick a pin in one of the remaining 14, she will have a 50% of hitting it in the very next draw. No rocket science there. Later I will post some of my weird and wonderful techniques with trips. I will experiment with them on your program to see how they work on it. One of the secrets with trips is to ignore the last one, two or three draws and then perform an analysis. This also works with any set of numbers.
Now going into the pairs territory, we seem to have lots of contenders. This should swing the other way in due course.I would say a trend will last no more than 3 draws. Chaotic systems pulsate and it seems to be pulsating this way at the moment but it will rebalance itself. You can often detect which way it swings with the Lotwon632 Pot filter. This filter reports on the pools of numbers where the winners come from. The median for real numbers(6/49) is round about 14 to 15. If the report shows 3 hits above the median then you can be pretty sure any real number with a skip of 15+ will not turn up next draw. This fact may be significant when I chat to these number pairs later.
 
You must log in or register to reply here.

Sidebar

Top