United States
Member #5,599
July 13, 2004
1,207 Posts
Offline
The Lottery Post has alot of ideas, methods, and systems being kicked around for everyone to see and weigh for themselves. Aside from the current posts there is a large history to refer to for just about any lottery topic your your interested in. So, the question is this...
Out of all the things you've read or seen here, what is the one single thing that you feel has helped you the most in selecting your lottery numbers more accurately?
For me, it was the concept of paying attention to the last digit of the lottery numbers (pick 5). When looking at them you find there are usually 4 to 5 different last digit numbers in every draw. Not only that, only one or two last digits repeat from the previous draw. I feel this info in part has made for an effective filter.
All comments are welcome.
You are a slave to the choices you have made. jk
Even a blind squirrel will occasionally find an acorn.
There is no elevator to success, you will have to take the stairs.
Switching between Fairfax, VA and Belgium Belgium
Member #19,286
July 29, 2005
2,254 Posts
Offline
Hi,
i'm pretty new to this forum.
for 6/42 lottogame i have been spending the last couple of months testing for the development of my own software. (that i do not plan on distributing, due to the fact it seems it never will be finished... whatever, if it ever gets finished, it will be available at no cost, that is, if it is any good ofcourse...)
For a couple of weeks now i've been reading up on this forum. B/c pick 3 did not had my intrest, i never spend any time on analysing it.
Until i stumbled upon this thread; https://www.lotterypost.com/thread/111125
Lantern (and califdude) did give some "lessons" in filtering and they are pretty interesting. So interesting that i even started a systemsanalysis for another prog now...
Anyway, lantern gave some great tips for everyone. Thumbs up for that!
As for selecting my numbers until short while ago, for 6/42, i run some software i wrote that is based upon some parts of MDIEditor (www.saliu.com) that i felt were applicable on Belgian lotto, in particular usage of the median.
Further on i noticed this lottosync-prog. As from what i read about it so far, it is pretty much based on probability, in that way that it "creates a zillion future drawings", after that the math rushes in to track a path back to the last 3 drawings, for constructing a visible pattern. (where the "weight" is more important than the "timefactor", i guess even Heisenberg must show up in the math! Promising 'tho!!!) Anyway, at first glance this thing looked promising, but i still have a lot of reading up to do about this "subject". (there must be as many threads about lottosync as there are possible combinations in euromillions ;-)
Then there's VRA (thanx to NemeSys). Been testing this program for several weeks now, and it looks like i finally achieve some progress in selecting the correct parameters. Still alot of testing up a head...
In final, i noticed that the Canadian Super7, drawn on friday-evening, holds at average 2 numbers (of 7) that are also drawn in the Belgian lotto on either saturday or following wednesday. On friday the 15th of july, it even had 6 numbers that also came in Belgian lotto the very next day. Too bad my eye only "fell" on this similarity a few days after that. I could track the drawings back to september (or october...) last year, but i cannot seem to get my hands on an entire drawingsfile from it... (if anyone has a link, thanx alot if you would provide it! ;-)
So these days i don't put too much "work" in selecting the numbers, b/c the Super7 does that for me!!! I just ad a few numbers and wheel them.
Canada
Member #12,229
March 8, 2005
28 Posts
Offline
paurths: Canada Super 7 draw files from the beginning of the draw can be downloaded from the British Columbia Lottery Corp's site at http://www.bclc.com (sorry, the board apparently doesn't support copy/paste from Mozilla or I'd post the whole link).
Tx United States
Member #4,570
May 4, 2004
5,180 Posts
Offline
paurths Hi!
Welcome !
Are you making software for analysing the pick 3 ?
I am interested. What analysis ideas do you have ? Is this software just for analysis or also for "Guessing-Predicting" the most likely numbers-combos to be played.
What basis or factors are you using or will use for your analysis ?
To me, almost everything is filters and filter patterns, other than using birthdays-dates, quick picks, same numbers every draw, dreams and the ocult, which I don't use, I only believe in filter patterns, including Hot and Cold patterns and Hot-Warm-Cold.
Analysis is the key. What won't come next is almost as important as what will come next.
Good luck.
Switching between Fairfax, VA and Belgium Belgium
Member #19,286
July 29, 2005
2,254 Posts
Offline
paurths Hi!
Welcome !
Are you making software for analysing the pick 3 ?
I am interested. What analysis ideas do you have ? Is this software just for analysis or also for "Guessing-Predicting" the most likely numbers-combos to be played.
What basis or factors are you using or will use for your analysis ?
To me, almost everything is filters and filter patterns, other than using birthdays-dates, quick picks, same numbers every draw, dreams and the ocult, which I don't use, I only believe in filter patterns, including Hot and Cold patterns and Hot-Warm-Cold.
Analysis is the key. What won't come next is almost as important as what will come next.
Good luck.
Hi Lantern,
well, currently i'm working on a systemsanalysis for the software itself, meaning that i'm still "investigating" the needs the software must have to analyse the past drawings. I don't make software that only just analyses the past draws and make no predictions.
It has to make predictions, otherwise i can well so do it all with pen and paper. (like Califdude stated; sometimes its easier and faster to just use pen and paper, or at least, thats what i think he meant...)
The information i will use is not complete so far. I've been reading up in "Lottery systems" on this forum, and sometimes some interesting ideas pop up. I've tested some of them on the Belgian pick3, and it seems that most of the ideas work only for a period and then they become useless. After a random period they pop up again and can be used again. Still having problems in determening the "fingerprint" for these methods so my software could recognize whenever a pattern pops up again and take advantage of it before it fades away again.
I guess you know what i mean.
I've been using your filters lately (even in "real time" and hit a box the other day! thanx), from the thread i mentioned in my earlier post, and they do a pretty bunch of filtering. Also the sums-root-chart, from https://www.lotterypost.com/thread/111125/4 , is a powerfull tool.
The difficult thing here is, if not done manually, that "it automatically" does not "overfilter" to achieve an acceptable range of combinations to play. Hope i make myself clear. If this turns out in a small prog, the user will still have to decide which parameters to use and which not... sadly, but true... (i hate software where the software lets you decide on just about everything, b/c then you just can't see the woods through the trees anymore...)
I'm pretty sceptic about the hot-cold issue. I still have a lot of research to do for pick3, but for regular lotto (at least 6/42) it is my experience that hot-cold often is a guide that leads to nothing.
I (well, the code does that ofcourse) tend to calculate the median for all numbers, for every drawing and archive that. Also the position of "amount of times a number came out" is something i use. The positions are recalculated for every drawing. (in Belgium it never happened that all 6 numbers were in the top 12, which is an interesting thing to know, b/c lots of people use these top-numbers) As you probably know, the top-10 or so changes quite often and is also a bad guide. What interests me in this top-n, is the numbers that shift position. (upwards AND downwards)
Same for the median, it does show patterns when comparing them for each drawing. (numbers don't, b/c they stay the same. 33 will for ever be 33. Medians and positions not! They only represent! You can put different pilots at different times in a jet, but the jet will stay the same. Some pilots will fly the thing as it should, others won't... does this make sens?)
Further on, it is my thought that all combinations that came out must be excluded from coming out again. (they might come out again, but i feel there's no need whatsoever to take that in account) Also the percentage of all combinations of 5 within the combinations of 6 that already came out, that come out again is low. (Also, no pattern there)
Combinations of 4, within 6, that's another issue on which i have not yet decided what to think of them.
Couples of 2 and 3 however are interesting, as i noticed, in Belgian lotto, some of them (especially 34 +38 is extremely interesting) do show patterns about when they will hit. They might miss one or two drawings but then they do hit.
Another thing i'm looking at is for each and every number a "hot-period" that returns. For example, number 7 has a interval of 5 years, then it becomes extremely hot for a period during a little less than 3 months.
On "my" sql-server, how convenient to have access to the servers at my workplace as being a programmer and being able to set up a "test-environment", i have all full-wheel for 8, 9, ... up to 21. (which runs high in the billions) It took for over 2 months to let them all crosscheck with past drawings, and you would be amazed with the result. At least i was. Some full-wheels, and i'm talking about 14 or so, have a hit-history that would make one cry. (not talking about e.g. 14 consecutive numbers or so) 'tho i feel that there's no gain there, first b/c it takes tooooooooooooooo long to do the calculations (and those are some pretty powered Dell-servers!) , second b/c a lack of patterns.
I have some little thing now, in Access, that i test every drawing, and it has some icecold periods, but then it hits for about 4 drawings by giving me at least 3 or 4 of the six numbers. I developed this little prog in less than a day, after reading up on the "lottosync-threads" where i picked up some ideas, in combination with saliu's MDIEditor (drawing range). Too bad it is highly unrelyable. I ran it against the entire history of the lotto here and there is no gain, exept in the period when it filters out the correct numbers. But i feel it might be a good start.
VRA also got my attention, and i believe that that might just be the best tool so far. Well, the math behind it might just be that.
For the time being i create 300 12-wheels per drawing of Belgian lotto, and almost every time i have all 6, or 7 (with the powerball) in at least 1 combination of them. (sometimes even 3 combinations) The problem is, for the moment, but i tend to find a solution to it (don't we all?), is that the parameters used to create those hitting combinations differ all. I'm surfing the net to find out alot more about the subject, http://en.wikipedia.org is a good start, and in the end it will be necesary to study upon it like a maniac.
I probably forgot to mention some things as they must have slipped my mind as it is filled with integrals, molecular dynamics, deterministic and non-deterministic algorithems, and god knows what else... My head is spinning so hard due to all these terms that i sometimes think my girlfriend must think i'm becoming a lunatic or something.
Anyway, always open for all ideas, as i don't believe in "this is my little secret". Why should i care if i ever wrote some stuff and someone hits every week with it? Won't change the odds for my playing in any way.
As for the "connection" to the Canadian Super7. I downloaded the entire history, thanx mabman, and it is truly astonishing!!! I don't think i can ever write some software that beats this! I must ad that i also tested for the value preceeding the actual number, so in theory the amount of numbers used doubles. In practice it does not, b/c first of all Super7 has more than 42 balls and second there are the consecutive numbers that only provide one extra number.
I have included the "extra" numbers.
Out of 705 drawings (twice belgian for each canadian) it hits all seven numbers from belgan lotto 14 times!!!
47 times all 6 numbers come out, and 125 times 5 numbers hit.
The average of used balls to achieve this rate is 15.
Tx United States
Member #4,570
May 4, 2004
5,180 Posts
Offline
paurths
"I'm pretty sceptic about the hot-cold issue. I still have a lot of research to do for pick3, but for regular lotto (at least 6/42) it is my experience that hot-cold often is a guide that leads to nothing"
------
Lantern.
You didn't understand what I meant when I mentioned Hot and Cold and Hot, Cold, Warm.
Too me they are just patterns such as High-Low, Even-Odd, Consecutives and any others.
I would use Hot-Cold patterns as "FILTER PATTERNS", same as I would use any other filter patterns, or patterns.
I would not use Hot-Cold as most other people use them, to me they are just one more kind of filter patterns.
-----
As to the rest of the post, it is kind of long and don't have time to study it right now.
Will post on it later.
-------
Also later I will talk to you about filter patterns, but maybe in private and not openly here.
The U.S. state lotteries have eyes and ears every-where. Not that it will do them any good, maybe for the most part they are morons, they probably don't understand much of anything.
Netherlands
Member #3,476
January 24, 2004
212 Posts
Offline
Hi Paurths,
"For the time being i create 300 12-wheels per drawing of Belgian lotto, and almost every time i have all 6, or 7 (with the powerball) in at least 1 combination of them. (sometimes even 3 combinations) The problem is, for the moment, but i tend to find a solution to it (don't we all?), is that the parameters used to create those hitting combinations differ all. "
I guess you create 300 full 12 wheels. So you'll have 300*924 = 277,200 combinations.
The odds are 1 chance in 973 to have 6 right in a 6/42 game when using 12 numbers (including the bonus number in the calculation). So 300 times 12-wheels the odds are 1 chance to 3.24(=(973/300)) approx. to have all 6....
How do you select your (300 times) 12 numbers? Randomly or do you use a specific method for selecting those numbers?
If you have lets say 50% of the time all 6 or 7 in the 277,200 combinations then I advice you to use professional software (like Lotto Architect or Expert Lotto) to reduce the amount of combinations dramatically almost on a safely way...