Algorithms for number selection?

Reply #51

RJOh — Sun, 20 May 2012 23:32:48 GMT

There's nothing proprietary about what I do. I just do the work of researching past drawings and seeing what worked most of the time and hope it will work again eventually.

Players are capable of coming up with all kinds of great ideas but great ideas that never worked aren't likely to just start working. For example playing hot numbers may be a great idea but how often did playing hot numbers ever win in the past? If you want the answer then you have to define what makes numbers hot and do... [ More ]

Reply #50

LottoBoner — Sun, 20 May 2012 22:39:05 GMT

so what are your targeting techniques??

You can IM me if you want to protect your proprietary information...

Reply #49

RJOh — Sun, 20 May 2012 22:00:20 GMT

While predicting can be a good way to see what happens when the rubber hits the road, it can also null ones expectations. Even with all the winning numbers in a pool of 25, it's hard to get even three in the same line with 20 lines. It appears the targeted numbers has to be even more defined to get a decent win.

Reply #48

RJOh — Sat, 19 May 2012 22:33:25 GMT

I meant to say:

Winning a lottery isn't something that hasn't happened many times before - RJ

I shall press on with my quest making predictions as tests. Good luck to others with the same quest.

Reply #47

LottoBoner — Sat, 19 May 2012 22:31:27 GMT

If you don't play, you can't win

Reply #46

SergeM — Sat, 19 May 2012 18:56:41 GMT

I have plenty of analyzing codes, and have ideas for many more but this has to rest for a few months.

Reply #45

RJOh — Sat, 19 May 2012 15:58:46 GMT

You have to play something! - the roulette player

There are no professional roulette players! - the professional

You can't win at roulette, unless you steal from it! - Albern Einstein

But, if you can't win at roulette, how could you possibly win in the lottery?

Winning a lottery is something that hasn't happened many times before - RJ

Somebody is always going to win, why not study the past to see if there aren't some commonalities among the winning combinations that might help me... [ More ]

Reply #44

SergeM — Sat, 19 May 2012 10:35:32 GMT

I googled algorithm and found that there are listed ordered kinds of algorithms.

There is a page full of those on Wiki. I don't have time now to read all of that.

It might be smarter to start with probability and sequences, meanwhile analyse

the drawings.

My first question is, what kind of algorithms are there? Are there a few main

groups that can be covered? What are that main ways to crunch numbers,

leaving no gaps? It probably takes a motivated dr. prof. in mathematics to... [ More ]

Reply #43

LottoBoner — Fri, 18 May 2012 22:59:34 GMT

my original intention in this comment was to say spreads.

Reply #42

dr san — Fri, 18 May 2012 12:57:00 GMT

Hello, RL, rondo, if I share your lottery 39/5 in 4 parts or 4 sectors In 100% of the sector will draw a zero or a number and another sector will have at least Two numbers (of course there could be more) in 100% 100% this standard is to use to filter? Sure you do not know which sector but the two features mentioned above are at 100% of the draws so do not care how this filter is always losing, you can play with three groups, but because in all three groups have 4-5 at 100%. this would be a great... [ More ]

Reply #41

RJOh — Fri, 18 May 2012 12:41:56 GMT

You could be right, however not being familiar with Matrix Walk Charts and not using spread sheets I can't respond to your statements and questions properly.

I'm writing all the routines I'm using just for my own use so there's no need for bells and whistles if they can't contribute to the results I'm hoping for.

I'll be testing those results by posting them on the prediction board so anyone can check them there if they wish.

Reply #40

LottoBoner — Fri, 18 May 2012 07:49:08 GMT

One hypnotises I came up with was Pick 6 draws tend to bias to the outer individual position distributions over time

Analysis showed this to be true on past history on a large number of draws I tested.

So are you saying i would do better to observe distribution pattern movement better if i was put under hypnosis

Reply #39

LottoBoner — Fri, 18 May 2012 07:41:23 GMT

yes the points are key...and that is why i use fib integers represented as skips to filter my set.

Reply #38

LottoBoner — Fri, 18 May 2012 07:32:58 GMT

I may be closer than you RJoh...

I have a question for you...on your Matrix Walk Chart for Rolling Cash 5/39...do the numbers go from left to right or right to left?

I am sure you are aware of all the same parameters that I am aware of because the Matrix is equivalent.

And i strongly doubt your spreadsheets are any less sophisticated than mine. Although i dont know how Basic appears on screen but i guess it has all the same bells and whistles as excel...

so which way does it go

Reply #37

RJOh — Fri, 18 May 2012 03:20:55 GMT

I may be close to developing something like an algorithm but it's not as simple as A x B x C and it can involve up to 75%or more of the numbers in the regular lottery pool and it has to be recalculated after each drawing.

The real advantage is in knowing the numbers and the range assigned to each position of the combinations. The winning combination will usually be in a group generated using these parameters which can be as large as one to two thousands combinations but picking a few combina... [ More ]

Reply #36

dr san — Wed, 16 May 2012 02:17:30 GMT

Hello rl perfect, then to see the pairs of points The pairs are (the 1st and 5th) Limit of the 1st 01-14 and 5th in the limit of (29-39) it is possible 300 pairs Unclear whether the pair hit the first number to be within the limits 01-14 and 5th number in the limit from 29 to 39 = 02 37 the pair example, if points are the pairs of the suit is between the pair 02xxx37 = xxx (central suit), it is very clear that filter, the three frequencies, larger, medium and cold, etc. .. let a former emplo lot... [ More ]

Reply #35

RL-RANDOMLOGIC — Wed, 16 May 2012 01:59:47 GMT

dr san

That's all well and good if a person can lock in on the first and last number, I find it difficult enough to

hit the first and last group and would hate to try and pin it down to one number each. If I choose 5

groups to play then my total sets is equal to 4X4X4X4X4=1024. Since I have 4 numbers in each group

except group 10 then I have 16 possible combinations of first and last number pairs. I can sometimes

remove one number from some of the groups which reduces the tot... [ More ]

Reply #34

dr san — Tue, 15 May 2012 12:31:36 GMT

Hello, rl randomlogic, your lottery is 39/5, I give this idea to make the suit or the central trio Example = 02 12 16 25 37 12,16,25 suit is central, then taking the number 01 and number 39, but because the number can only be one of the 39 th and last position, So we have 37 numbers in central suit total = 7770 suit in 100% of any lottery, so then number of points is placed after, but these 7770 can create filters to decrease to 5% of this value would be in three bands the tip, the central strip... [ More ]

Reply #33

RL-RANDOMLOGIC — Tue, 15 May 2012 06:42:01 GMT

Developer

I am working on very much the same thing. For my 5-39 I use 10 groups, 9 have 4 and the 10th has 3.

The group from which the 1st and 5th numbers will come seems to be the easiest to hit and is very much

a product of what you suggest above. It may not make us winners overnight but I think it's a step in the

right direction. I try and select the groups for the outside numbers and many times block a couple on the

inside as this works well for me most days. the highest... [ More ]

Reply #32

LottoBoner — Mon, 14 May 2012 20:22:47 GMT

Usually opening kickoff at the super bowl...

Reply #31

Developer — Mon, 14 May 2012 20:12:56 GMT

I have done much study on lotteries over the years and have found some interesting algorithms on my way.

One hypnotises I came up with was Pick 6 draws tend to bias to the outer individual position distributions over time

Analysis showed this to be true on past history on a large number of draws I tested.

I analysed all possible outcomes which showed why this would be true, it was simply due to no lower numbers than 1 or higher numbers than 49 (in this instance) meaning the field of pla... [ More ]

Reply #30

RL-RANDOMLOGIC — Mon, 14 May 2012 19:29:50 GMT

RJ

Most of the time I play it's a rush to get tickets before they shut down for the draw. The way I count progress

has nothing to do with tickets but my setup. The predictions board has no real value to me and I can't see where

posting tickets I don't intend to buy does anything to improve my play. I think it's more of a bragging contest than

anything else. Many post more tickets in a week than I would play in a year, I think many would be shocked at how

little I play the lottery.... [ More ]

Reply #29

SergeM — Mon, 14 May 2012 15:13:01 GMT

There is a simple approach. When you flip a coin, what is the next outcome

Reply #28

RJOh — Mon, 14 May 2012 14:56:19 GMT

RL

You don't have to believe in prediction to use the prediction board, it can be an excellent independent testing facility for evaluating theories and strategies for winning lottery games. Using the prediction board is as close as you'll ever come to applying those theories and strategies in a real lottery without actually buying tickets.

RJ

Reply #27

RL-RANDOMLOGIC — Mon, 14 May 2012 12:31:35 GMT

LB

I agree, there are the simple and then there are some that are very complex. I just added a couple

suggestion algorithms for the lower level filters and was suprised that it's hitting 80% and up for

my 5-39. The problem is that I still have to find the 20% which are not correct so I don't know if

there is any avantage to using it. I don't much believe in prediction but I am open minded to the

possibility that more can be done. I started a new algorithm for my group settings and... [ More ]

Reply #26

RJOh — Sun, 13 May 2012 23:45:30 GMT

Hello dr san,

I consider the validity of ones posts by the ratings of their predictions. Since you don't make predictions, your posts lack any significance to me. Good luck to you.

Reply #25

dr san — Sun, 13 May 2012 23:03:58 GMT

Hello, RJOH a lotto 49/6 divided by 20 positions to see the suit 1,2,3 1,2,4 1,2,5 1,2,6 ..... = 4,5,6 are 20 possible positions One could vera likelihood of repetition in each of the 20 positions Example suit 05,12,45 = the best position to bring this suit? Then they could see which of 20 possible lottery 49/6, you could play as fixed For a calculation of probabilities of the positions is not on the forum The frequency of repetition of certain suits for use in the same positions as fixed

Reply #24

RJOh — Sun, 13 May 2012 20:15:43 GMT

After observing some of the types of wins that have occured lately with MM and PB where one person or group played the same combination several times with different bonus numbers and winners having several other lines that won seconds I not sure that's true. Apparently there are strategies that some players have developed and maybe even programmed but they are proprietary.

Reply #23

RJOh — Sun, 13 May 2012 19:52:38 GMT

You don't have to use a computer to come up with variables or use them, it just that using a computer might make the process faster. Besides if you come up with something that shows promise, the potential rewards could be an incentive to learn what ever more you think you need to know to use a computer to do the work.

I'm experimenting with ways to come up with variables that are reliable 75% of the time using the local game (6/49) but the targeted number groups are too large to get a winner... [ More ]

Reply #22

LottoBoner — Sun, 13 May 2012 18:46:32 GMT

writing an algorithm that a computer can run is not necessarily easy even if you know the variables...

i observe numerous variables...but i dont know how to write a algorithm...

not in VBA, not in C+, not in Java,

and variables move in trends also...so there is no set of variables that are reliable 75% if time...it is up to the player to time the variables appearance.

the more important question is how often does a specific set of variables produce a jackpot in say 3 months or 6 mont... [ More ]

Reply #21

SergeM — Sun, 13 May 2012 18:19:20 GMT

Apparently, there is no program that does that.

Reply #20

RJOh — Sun, 13 May 2012 01:46:32 GMT

Writing an algorithm will be easy once someone has done the work of coming up with the variables that are related to the winning numbers in the drawings at least 75% of the time.

Also those variables much target a group of numbers small enough that most of the possible combinations could fit in the budgets of a small pool of players or a single player who doesn't mind spending a little money. Even if they don't win a jackpot, most of the cost sould be covered by the smaller winnings such tha... [ More ]

Reply #19

nickbrownsfan — Thu, 10 May 2012 01:38:38 GMT

that is pretty cool and I also think that numbers are connected in ways that we just somehow dont really see. I was once working on on the distance of numbers and releating them to decades its something most people will blow off as a waste of time but I do see releationships between numbers in this way.

Keep up your research its something i would be really interested in.

Reply #18

Artist77 — Wed, 09 May 2012 22:01:56 GMT

I tried a new strategy today with my PB pick. I did an individual calculation for each number...I picked the first number based on the first number drawn last time and compared it to the last 10 (up one or down one based on an average, etc)....then picked the rest of the numbers the same way. One of the numbers seemed to be almost always connected with multiples of another number. Gave me an interesting set of numbers I would never have picked.

Reply #17

nickbrownsfan — Tue, 08 May 2012 04:11:22 GMT

Reply #16

RJOh — Mon, 07 May 2012 14:57:14 GMT

No lottery jackpot winner has ever regretted the time and effort he/she spent trying to win it.

Reply #15

meskhov — Mon, 07 May 2012 03:24:22 GMT

RL:

Yes, you are so right and this is exatly what has been getting me frustrated! After the draw comes out it seems so simple but before the draw just too many numbers. You might have an idea of studying the igits due instead. Might be worth looking into as it can't hurt as what I'm doing now doesn't help.

Will keep grinding away at it.

Thanks for your help as it makes me feel good that someone else sees my logic in trying.

M

Reply #14

LottoBoner — Sun, 06 May 2012 07:14:59 GMT

Ok mister unknowledgable....i would like to see you do the analysis...and dont forget the f--king fibbies...or i will have to fast furious fully function fabricate fully a function fiddly diddly fideism... U...f---E-R

Reply #13

RL-RANDOMLOGIC — Sun, 06 May 2012 03:36:36 GMT

meskhoy

I worked on something simialr to this but finally gave it up. Taking the numbers from a couple draws

and using hindsight to build the next set is easy once the draw is over with but doing it before the

draw is a bit harder. It's amazing just how many numbers can be gotten from this sort of analysis.

Once the draw has come and gone it's much easier to come up with the correct 5 but that dose not

help us win. Ever notice how often one or more numbers come from the date.... [ More ]

Reply #12

SergeM — Fri, 04 May 2012 11:56:18 GMT

I think that all you find is that generally there is no such thing as full repeats or totally different, especially if you divide your numbers into groups.

Example of changes to nearest:

0 -1 +7 0 -4 0

0 +-1 -1 -4 -5 0

0 -3 +-1 -6 +1 +1

+15 +6 -1 -3 +3 0

-4 -6 -8 +-10 -1 -3

+-5 +3 +1 +-1 +1 -16

+-3 +2 0 -2 -5 +3

0 -1 -2 -3 -3 -5

0 = repeat, ... . For repeats there is an easy average probability, but as you can see an average isn't a ru... [ More ]

Reply #11

RJOh — Tue, 01 May 2012 02:07:40 GMT

It's going to be interesting to see the algorithm you develop using that data.

Good luck.

Reply #10

BlueDuck — Mon, 30 Apr 2012 23:39:23 GMT

Made from last 30MM drawings. 1,6,.... have 4hits last 30 games. 4,5,10 have 3hits last 30 games. Maybe 4 and 13 will get hits on 5-1-2012 to straighten out the curve??

0 1 2 3 4 5 6 7

1 1

2 2

3 3

4 4

5 5

6 6

7 7

8 8

9 9

10 10

11 11

12 12

13 13

14 14

15 15

16 16

17 17

18 18

19 19

20 20

21 21

22 22

23 23

24 24

25 25

26 26

27 27

28 28

29 29

30 30

31 31

32 32

33 33

34 34

35 35

36 36

Reply #9

RJOh — Sat, 28 Apr 2012 18:45:15 GMT

We all hit things at different angles and just miss but some day we will collectively pool all of it into a winner.

I'm surprised it hasn't already happened but I guess when it comes to winning a lottery jackpot, luck trumps good ideas every time and there isn't an algorithm for that.

Reply #8

nickbrownsfan — Sat, 28 Apr 2012 05:59:04 GMT

Thats why I love this place so much. If combined with what Im working on for Splinter trying to find the actual decades of numbers to be drawn the pool would be much easier to manage and perhaps much more focused.

We all hit things at different angles and just miss but some day we will collectively pool all of it into a winner. keep the up the great work that is an excellent approach.

Reply #7

meskhov — Fri, 27 Apr 2012 22:42:28 GMT

RJOH:

Yes, I see what you mean.

I'm going to try a few things in the upcoming games and see if it helps with the number selections.

Will let you know if anything develops.

We'll keep trying.

Thanks Again for all your assistance.

M

Reply #6

RJOh — Fri, 27 Apr 2012 20:57:37 GMT

It all depends on what you call simple. When Massachusetts had a run down of their state game when the jackpot hadn't been won after reaching more than $4M and would be shared among the second prizes winners, the couple from Michigan thought buying $150K to $300K worth of tickets to win a few second prizes a simple solution to winning part of the jackpot. But if I ever came up with an algorithm, it wouldn't be simple if it required me to spend more than $25-$50 to win, even win big.

Reply #5

meskhov — Fri, 27 Apr 2012 20:38:40 GMT

RJOH:

Thanks for the reply. I do believe there is a simple solution to this but I can't seem to nail it down to a reasonable set of numbers to wheel. As from my example almost every draw is similar to this but I just can't seem to narrow it down enough to even capture a 4/5 from the grouping. I was hoping for some simple equations but I guess that won't happen. I was hoping that I just am not seeing something simple in order to derive a better simple equation to narrow a group of winning numb... [ More ]

Reply #4

RJOh — Fri, 27 Apr 2012 19:42:40 GMT

That's a fairly easy one to test. I would be surprised if it hasn't already been tested for most games. A similar one for pick3 and pick4 games have been written about many times. Since most of those threads didn't get much attention, I assume it wasn't very successful.

Reply #3

LottoBoner — Fri, 27 Apr 2012 19:24:45 GMT

according to KISS, a simple algorithm would be as follows...

If your last draw results were (N1) (N2) (N3) (N4) (N5)... then the next draw could be

(N1) (N1+1) (N3+1) (N4+3) (N5-3)

of course you would have to try out all the differnt possible algorithms and back test to see which occurs the most.

well not all, but i would start with the core fibs...+0, +/-1, +/-2, +/-3 almost every draw has an adjacent number so you just have to guess when you think 2 or more will occur...

Reply #2

RJOh — Fri, 27 Apr 2012 19:07:48 GMT

I've been using a similar system to pick numbers for MM the last few weeks only I search all the drawings back to 06/24/l2 when the present matrix started and look at the numbers in the last drawing and what followed them every time they hit in past drawings in the same positions.

For example in the last drawing the lowest number was 3 and it has been the lowest number 54 of the 61 times it hit in the past 713 drawings. Thirty seven of those time when the previous number was higher, the lowes... [ More ]

Reply #1

meskhov — Thu, 26 Apr 2012 21:50:07 GMT

Ok, here is an example:

Draw 4/23/2012 3-7-9-20-23

Draw 4/22/2012 19-21-25-30-31

Draw 4/21/2012 6-7-10-33-35

Now pretending we did not know the outcome of April 23rd here is what I usually do to find the probable numbers.

Taking 10 minus 7 = 3

Taking 31 minus 25 = 6 + 1 = 7

taking 30 minus 21 = 9

Taking 33 minus (6+7) = 20

Taking 33 minus 10 = 23

Now the tricky part, it isn't always this easy as a lot of times these additions and subtractions are different and are f... [ More ]

Algorithms for number selection?

meskhov — Thu, 26 Apr 2012 21:30:53 GMT

Does anyone have or know of any simple algorithms to try on past drawings to determine the result of the next draw? I have been working on our pick 5 here in Mass and always seem to miss the big win. I have discovered that from the last 2 draws I can almost always find the next winning draw line. This is done by just adding or subrtracting from the last 2 lines to make up the proables for the next draw. Even knowing this I can't manage to even lock down 4/5 never mind 5/5.

I could post an ex... [ More ]