# A digital look at a jackpot lottery

Part 1 - Introduction.

This presentation is a response to requests to provide some guidelines and principles on how to improve winning in lotteries. While it is a lottery specific it may serve as a blueprint scheme for any other jackpot lottery. It describes a complete system with data analysis, predictions and backtracks. It is a also an attempt to evaluate a completely new approach towards jackpot lotteries.

The lottery in question is BC lotto 649. The reasons for this choice is that it's in my area and I intend to play it regularly (after giving all the systems and wheels a thorough testing).

My presentation has to be broken into smaller entities as it is simply too big for one shot. I will try to squeeze one subject in each post to maintain consistency. I will also try to proceed from one subject to another in logical steps as one would follow when actually playing. And there will be a lot of images to illustrate each presentation point.

Let's start from the data. These are not the real numbers from the lottery but random draws generated by the random numbers generator of Java in which the software was compiled. There are 250 of these draws. I consider them random enough for the purpose of this presentation. None of the draws was entered manually, the entire randomization process was totally automated. Below are 2 screenshots of the latest draws.

I present only 2 screens from the data file because they include the latest draws and the main samples in the presentation later on are based on these latest numbers, so if you would like to verify the correctness of the computations you have the base to do so.

The numbers in each draw are sorted in ascending order, the last number being the bonus; it is always the last to identify its status. However, because it does participate in winning prizes, in computations it is treated as any other number.

BC Lotto 649 has some peculiarities that differentiate it from other similar lotteries. It draws 6 numbers and a bonus with every draw but the bonus matters only for the wins of 5 and 2. Below are the prizes for each win level:

5 + bonus  =\$75,000
5               = \$750
4               = \$75
3               = \$10
2 + bonus  = \$5

I include the prizes here because they play an essential role in computing ROIs in backtracks. So when you see backtracks later you will know where the numbers come from.

These prizes are generally low but, as you will see later, 2 + bonus are quite often very important for ROIs. In general, the bonus number plays much more significant role than I expected at the beginning.

Now let me explain why it is a digital approach. For years I used to analyse relationships between numbers from each draw. This corresponds to analysing combos in Pick 3. There are 1000 such combos but only 10 digits combined into combos. Where is it easier to identify relationships - for 1000 or just 10 elements? No brainer answer and my app for Pick 3 confirmed that theory big time by identifying frequent and long lasting winning clusters.

Well, jackpot lotteries use the same 10 digits. Now, I guess, you realize what I'm up to. Yes, to analyse digits in jackpot lotteries, not numbers, a completely different approach. So the subject of the next post will be the analysis of the relationships between digits.

Part 2 - Digits in numbers

Every number in a jackpot lottery consists of 2 digits. I call them the first and the second digits. For example, number 5 consists of digits 0 (the first digit) and 5 (the second digit), number 38 - 3 as the first digit and 8 as the second digit. So far this should be easy to understand.

There are only 5 possible first digits - 0 to 4 but 10 possible second digits (0 to 9). Because of this difference they are not directly comparable. Therefore, any statistical analysis, frequency for example, must include 2 charts: one for the first digits and one for the second digits.

Like in Pick 3, numbers in jackpot lotteries have their sums. Below is the distribution of numbers in sums. I present these sums here because groups of sums may be used for predictions, based on each sum popularity.

In this system all data manipulation is based exclusively on digits, there are no numbers involved. All numbers are always split into their basic digits. Only sums involve numbers but they are selected based on their digits sums. That's why I call this whole methodology digital.

Now, having 2 groups of digits we can run some statistics on each. That's the subject for the next post.

Part 3 - Statistics

In this presentation I concentrate on statistical computations that are simple and easy to generate but still very effective in real play. Here I describe 3 of then.

First, digit frequency for each group of digits. Below are the results for the latest 20 draws.

In an instant you can see which first and second digits were the most frequent. For the first digit in just 20 draws the difference between the best (1) and the worst (3) is 14 which is quite large. Thus, if you played numbers 30 to 39 during that time you probably did not win much. For the second digit the difference is 10 (1 versus 2). This clearly demonstrates that in that period of 20 draws using certain digits in numbers could make a big difference in hitting drawn numbers. Without knowing these relationships you are playing blind, it's that simple.

Second, the digit trends for both first and second digits. Trends indicate how each digit performed in the last 20 draws in comparison with the previous 20 draws. Differences can be quite astonishing.

Third, the sums of digits, how did they perform in the last 20 draws. Results below.

The actual sum frequency is compared with the expected frequency for the selected number of draws. Both these value will change with the change of the number of draws.

Few sums were above their expected average with 1 clearly dominating - the sum of 4 (which includes the numbers 4,13,22,31,40). Which means that these numbers were more frequently drawn that others. Checking stats like these should definitely help to increase hitting the numbers drawn.

Fourth, the spread of the digits - how many different digits appeared in each draw. For example, for the first digit, 4 different digits dominated the field, like 9,12,18,32,44 49 (first digits: 0,1,3,4). For the second digit 6 different digit dominated as well. Again, it's extremely important to know these relationships as, for example, picking only 4 different digits for the second digit would end up in a total disaster in the latest 20 draws. Therefore, in this example, the only rational choice would be playing 4 different digit for the fist digit and either 5 or 6 for the second. This would generate 20 or 24 playable numbers.

This leads us to the next subject - wheeling. That's in the next post.

Part 4 - wheeling

Whenever you play more than 2 lines (or tickets) in a jackpot lottery, whether you realize it or not - you are wheeling. You can use your own distribution methods, mathematically designed or haphazard, does not matter - you are wheeling.

Generally used description for a wheel is, for example 3if5of20. I don't use it as it misses 1 very important element of the wheel - its size, how many lines (tickets) it contains. I want to know this number before I even open the wheel.

Therefore I use a different wheel definition, like 3W 4H 14N 13T. The translation is: you should win at least 3 numbers if you hit 4 numbers out of selected 14 distributed into 13 tickets. All required element are visible in the wheel definition.

Now, I used phrase "you should win", not "you will win" because I did not test all the wheels below if they guarantee the W win number. But, frankly speaking, this may not matter that much, as I will explain in the backtrack section. Therefore, I'm willing to give a chance to any wheel.

I accumulated over 300 wheels, many my own and a large number from CoverMaster and other sources. They are listed below, sorted first by N (numbers selected), then W (win), then H (hit) and finally T (the number of tickets). I limited N (the numbers selected) to the maximum of 30.

Generally wheels have 3 qualities:
1) Distribute numbers into tickets
2) May guarantee a minimum win (some may not have this ability)
3) May generate multiple winners

In this presentation I will put emphasis on the 3rd quality. I want to see which wheels and how often can generate multiple lower prize winners, sometimes to the extent that they not only offset the cost but actually generate a profit. It is possible and I will show how later. Why does it matter so much? Because I want to play at least possible cost.

The last wheel in the list 3W 6H 38N 21T is an odd fellow. This is my old timer, so old that I'm not even sure if I did not suck it out with the milk of my mother. I tested it and it does not guarantee win 3, or actually anything else but long time ago, during a simulation on real Canadian Lotto 649 numbers it was the only wheel that won 5 + bonus in one of its tickets. But it was only a simulation so no money came with it. Nevertheless, it got my saliva dripping and since then it's my sweetheart.

Below is an example of a wheel. It's one of my favourites and proved its valour in the real play. Watch for the overlap in the 2nd and 3rd pictures.

Now the list of the wheels that I could potentially use.

7 numbers
2W 2H 7N 3T
3W 3H 7N 4T
4W 4H 7N 5T
5W 5H 7N 6T
6W 6H 7N 7T

8 numbers
2W 2H 8N 3T
2W 3H 8N 2T
3W 3H 8N 4T
3W 4H 8N 3T
4W 4H 8N 7T
4W 4H 8N 8T
4W 5H 8N 3T
5W 5H 8N 9T
5W 6H 8N 4T
5W 6H 8N 12T
6W 6H 8N 28T

9 numbers
2W 2H 9N 3T
2W 3H 9N 2T
2W 4H 9N 2T
3W 3H 9N 7T
3W 3H 9N 8T
3W 3H 9N 9T
3W 4H 9N 3T
3W 5H 9N 2T
4W 4H 9N 9T
4W 4H 9N 12T
4W 5H 9N 3T
4W 6H 9N 3T
5W 5H 9N 30T
5W 5H 9N 33T
5W 6H 9N 8T
5W 6H 9N 9T
6W 6H 9N 56T

10 numbers
2W 2H 10N 5T
2W 3H 10N 2T
2W 4H 10N 2T
2W 5H 10N 2T
3W 3H 10N 10T
3W 4H 10N 4T
3W 5H 10N 2T
3W 5H 10N 7T
3W 6H 10N 2T
4W 4H 10N 20T
4W 4H 10N 23T
4W 4H 10N 30T
4W 5H 10N 7T
4W 6H 10N 3T
5W 5H 10N 50T
5W 5H 10N 56T
5W 6H 10N 15T
5W 6H 10N 16T

11 numbers
2W 2H 11N 6T
2W 3H 11N 2T
2W 4H 11N 2T
2W 5H 11N 2T
2W 6H 11N 2T
3W 3H 11N 12T
3W 3H 11N 13T
3W 3H 11N 15T
3W 4H 11N 5T
3W 4H 11N 6T
3W 5H 11N 2T
4W 4H 11N 34T
4W 4H 11N 38T
4W 5H 11N 10T
4W 5H 11N 12T
4W 6H 11N 5T
5W 6H 11M 25T
5W 6H 11N 26T
5W 6H 11N 27T
5W 6H 11N 29T

12 numbers
2W 2H 12N 6T
2W 3H 12N 2T
2W 4H 12N 2T
2W 5H 12N 2T
2W 6H 12N 2T
3W 3H 12N 15T
3W 4H 12N 8T
3W 5H 12N 2T
3W 5H 12N 3T
3W 6H 12N 2T
4W 4H 12N 53T
4W 5H 12N 15T
4W 5H 12N 17T
4W 6H 12N 6T
4W 6H 12N 8T
5W 6H 12N 42T
5W 6H 12N 45T
5W 6H 12N 46T

13 numbers
2W 2H 13N 7T
2W 3H 13N 4T
2W 4H 13N 2T
2W 5H 13N 2T
2W 6H 13N 2T
3W 3H 13N 21T
3W 3H 13N 23T
3W 3H 13N 24T
3W 4H 13N 10T
3W 5H 13N 5T
3W 6H 13N 2T
4W 5H 13N 25T
4W 5H 13N 26T
4W 6H 13N 11T
4W 6H 13N 12T

14 numbers
2W 2H 14N 9T
2W 3H 14N 4T
2W 4H 14N 3T
2W 5H 14N 2T
2W 6H 14N 2T
3W 3H 14N 26T
3W 3H 14N 27T
3W 4H 14N 12T
3W 4H 14N 13T
3W 4H 14N 14T
3W 4H 14N 16T
3W 5H 14N 5T
3W 5H 14N 7T
3W 6H 14N 4T
4W 5H 14N 34T
4W 5H 14N 37T
4W 5H 14N 40T
4W 6H 14N 15T
4W 6H 14N 16T
4W 6H 14N 18T

15 numbers
2W 2H 15N 10T
2W 3H 15N 4T
2W 4H 15N 3T
2W 5H 15N 3T
2W 6H 15N 2T
3W 3H 15N 33T
3W 3H 15N 34T
3W 4H 15N 15T
3W 4H 15N 16T
3W 4H 15N 17T
3W 5H 15N 8T
3W 6H 15N 4T
4W 5H 15N 48T
4W 5H 15N 50T
4W 6H 15N 21T
4W 6H 15N 23T
4W 6H 15N 27T

16 numbers
2W 2H 16N 10T
2W 3H 16N 6T
2W 4H 16N 3T
2W 5H 16N 3T
2W 6H 16N 3T
3W 3H 16N 42T
3W 3H 16N 43T
3W 4H 16N 19T
3W 4H 16N 21T
3W 5H 16N 9T
3W 5H 16N 10T
3W 6H 16N 5T
4W 6H 16N 31T
4W 6H 16N 33T
4W 6H 16N 38T

17 numbers
2W 2H 17N 12T
2W 3H 17N 7T
2W 4H 17N 3T
2W 5H 17N 3T
2W 6H 17N 3T
3W 3H 17N 50T
3W 3H 17N 55T
3W 4H 17N 23T
3W 4H 17N 25T
3W 5H 17N 11T
3W 5H 17N 13T
3W 5H 17N 15T
3W 6H 17N 6T
4W 6H 17N 40T
4W 6H 17N 43T

18 numbers
2W 2H 18N 14T
2W 3H 18N 9T
2W 4H 18N 3T
2W 5H 18N 3T
2W 6H 18N 3T
3W 3H 18N 48T
3W 3H 18N 52T
3W 4H 18N 26T
3W 4H 18N 27T
3W 4H 18N 29T
3W 5H 18N 12T
3W 6H 18N 9T
3W 6H 18N 10T
4W 6H 18N 42T
4W 6H 18N 54T
4W 6H 18N 56T

19 numbers
2W 2H 19N 16T
2W 3H 19N 9T
2W 4H 19N 5T
2W 5H 19N 3T
2W 6H 19N 3T
3W 4H 19N 33T
3W 4H 19N 34T
3W 5H 19N 15T
3W 5H 19N 16T
3W 5H 19N 22T
3W 6H 19N 11T
3W 6H 19N 12T

20 numbers
2W 2H 20N 18T
2W 3H 20N 10T
2W 4H 20N 5T
2W 5H 20N 4T
2W 6H 20N 3T
3W 4H 20N 39T
3W 4H 20N 43T
3W 5H 20N 19T
3W 5H 20N 20T
3W 5H 20N 24T
3W 6H 20N 12T
3W 6H 20N 13T
3W 6H 20N 14T

21 numbers
2W 2H 21N 18T
2W 3H 21N 12T
2W 4H 21N 5T
2W 5H 21N 4T
2W 6H 21N 4T
3W 4H 21N 48T
3W 4H 21N 49T
3W 5H 21N 21T
3W 5H 21N 25T
3W 5H 21N 28T
3W 6H 21N 18T

22 numbers
2W 2H 22N 19T
2W 2H 22N 21T
2W 3H 22N 13T
2W 3H 22N 11T
2W 4H 22N 7T
2W 5H 22N 4T
2W 6H 22N 4T
3W 4H 22N 56T
3W 5H 22N 24T
3W 5H 22N 29T
3W 5H 22N 33T
3W 6H 22N 20T
3W 6H 22N 22T

23 numbers
2W 2H 23N 22T
2W 2H 23N 23T
2W 3H 23N 13T
2W 3H 23N 15T
2W 4H 23N 8T
2W 5H 23N 4T
2W 6H 23N 4T
3W 5H 23N 26T
3W 5H 23N 35T
3W 5H 23N 39T
3W 6H 23N 23T
3W 6H 23N 26T

24 numbers
2W 2H 24N 24T
2W 2H 24N 25T
2W 3H 24N 13T
2W 3H 24N 14T
2W 4H 24N 8T
2W 4H 24N 10T
2W 5H 24N 4T
2W 6H 24N 4T
3W 5H 24N 30T
3W 5H 24N 40T
3W 5H 24N 44T
3W 6H 24N 21T
3W 6H 24N 26T
3W 6H 24N 28T

25 numbers
2W 2H 25N 23T
2W 2H 25N 27T
2W 3H 25N 13T
2W 3H 25N 18T
2W 4H 25N 9T
2W 4H 25N 10T
2W 5H 25N 6T
2W 6H 25N 4T
3W 5H 25N 36T
3W 5H 25N 47T
3W 5H 25N 50T
3W 6H 25N 29T
3W 6H 25N 35T

26 numbers
2W 2H 26N 27T
2W 3H 26N 13T
2W 3H 26N 18T
2W 4H 26N 10T
2W 4H 26N 11T
2W 5H 26N 6T
2W 6H 26N 5T
3W 5H 26N 40T
3W 5H 26N 55T
3W 6H 26N 33T
3W 6H 26N 36T

27 numbers
2W 2H 27N 29T
2W 2H 27N 31T
2W 3H 27N 14T
2W 3H 27N 21T
2W 4H 27N 10T
2W 4H 27N 12T
2W 5H 27N 6T
2W 6H 27N 5T
3W 6H 27N 30T
3W 6H 27N 37T
3W 6H 27N 41T

28 numbers
2W 2H 28N 30T
2W 2H 28N 33T
2W 3H 28N 16T
2W 3H 28N 22T
2W 4H 28N 14T
2W 5H 28N 8T
2W 6H 28N 5T
3W 6H 28N 36T
3W 6H 28N 42T
3W 6H 28N 45T

29 numbers
2W 2H 29N 31T
2W 2H 29N 34T
2W 3H 29N 17T
2W 3H 29N 25T
2W 4H 29N 13T
2W 4H 29N 15T
2W 5H 29N 9T
2W 6H 29N 5T
3W 6H 29N 48T
3W 6H 29N 51T

30 numbers
2W 2H 30N 31T
2W 2H 30N 35T
2W 3H 30N 22T
2W 3H 30N 28T
2W 4H 30N 16T
2W 4H 30N 18T
2W 5H 30N 9T
2W 5H 30N 11T
2W 6H 30N 5T
3W 6H 30N 54T
3W 6H 30N 57T

38 numbers
3W 6H 38N 21T

So now we have statistics and we have wheels. Time to move forward and talk about predictions - in the next post.

Part 5 - Predictions

Now we are ready to make some predictions. How do we make them in this system? By combining a certain number of first and second digits into playable numbers. In this example I will use 3 first digits and 5 second digits. This will amount to 15 numbers. These numbers will have to be wheeled for appropriate distribution.

Next, I need to selected an appropriate wheel. For 15 numbers I have these choices:

2W 2H 15N 10T
2W 3H 15N 4T
2W 4H 15N 3T
2W 5H 15N 3T
2W 6H 15N 2T
3W 3H 15N 33T
3W 3H 15N 34T
3W 4H 15N 15T
3W 4H 15N 16T
3W 4H 15N 17T
3W 5H 15N 8T
3W 6H 15N 4T
4W 5H 15N 48T
4W 5H 15N 50T
4W 6H 15N 21T
4W 6H 15N 23T
4W 6H 15N 27T

I will pick this wheel: 2W 3H 15N 4T. Just 4 tickets for every draw, a bargain. Below is how the numbers are distributed in the wheel:

Now I would like to know how many numbers my predictions hit in the last 20 draws, regardless of the wheel. Why is this important? Because hits indicate what is a possible maximum win you can have in each draw. If you hit only 3 numbers don't count on winning 4 or 5. So this is an important metric to observe. Examples below. Observe the number of hits. The last picture is a summary.

Don't confuse hits with wins. These are totally different metrics and I will demonstrate the difference in backtracks. Hits are generated by your number selections, wins - by the wheel. That's the reason why they are not the same. So you need to monitor them separately.

What can be done to increase the number of hits? There are 2 ways: improve the quality of your selections and/or increase the number of selected numbers. The common misconception is that more numbers necessitate much larger wheel. Not necessarily the case. Compare these 2 groups:

2W 2H 15N 10T
2W 3H 15N 4T
2W 4H 15N 3T
2W 5H 15N 3T
2W 6H 15N 2T
3W 3H 15N 33T
3W 3H 15N 34T
3W 4H 15N 15T
3W 4H 15N 16T
3W 4H 15N 17T
3W 5H 15N 8T
3W 6H 15N 4T
4W 5H 15N 48T
4W 5H 15N 50T
4W 6H 15N 21T
4W 6H 15N 23T
4W 6H 15N 27T

2W 2H 28N 30T
2W 2H 28N 33T
2W 3H 28N 16T
2W 3H 28N 22T
2W 4H 28N 14T
2W 5H 28N 8T
2W 6H 28N 5T
3W 6H 28N 36T
3W 6H 28N 42T
3W 6H 28N 45T

Almost a double increase in numbers but nowhere close to that in tickets. True that guaranteed wins in each group are different but this will matter only when you decide what strategy to adopt. I see 3 such strategies:

1) Play only for lower win levels (2+, 3, 4)
2) Play only for higher win levels (5, 5+, 6)
3) Play for both at the same time.

I don't fool myself: whatever approach I adopt, whatever predictive methodology I apply, my chances of winning a jackpot are very, very, very tiny (repeat about 13 million times). So I don't target a jackpot. If it comes that would be only by luck. But, I believe, 5 and 5+ are achievable goals. So in order to make it happen I have to hit 5+ frequently. Obviously easier said than done. But I want to investigate which prediction methodologies in combination with which wheels produce the best results - at any given time. Because it's highly unlikely that I would use 1 wheel only all the time - no matter how good it is in theory. We will further explore this topic in backtracks.

I'm not going to rely on 1 prediction methodology, either. Pick 3 simulations taught me that it's a silly approach. Nothing wins forever, panta rei applies to the full extent. So I want to have choices. At this time I'm planning to have 13 prediction methods with high probability of more to come in the future. The future will also tell if this multitude is more of a benefit or a headache.

Finally we arrive at the essence of the whole exercise - backtrack verifications. That will be in the next post.

Part 6 - backtracks

This is the final stage of the system - the verification of the results, otherwise called backtracks. This stage is also the most important because it allows to see what prediction methods combined with what wheels produced the best results in a selected number of draws, starting from the latest one.

For this presentation I will backtrack only up to 20 draws. In the final version of the software I will make it variable, from 10 to 50; I don't see a benefit in going beyond that.

Backtracks are fully automated, the only manual entry is the selection of digits and a wheel. I will continue with the example (the digits and the wheel) used for predictions.

Backtracks measure prediction/wheel performance in the following categories:
1) Number and type of hits.
2) Number and type of wins.
3) Amounts spent
4) Amounts won
5) ROI (return on investment)

Below are the results for the selected example:

Now let's take a look at the total results. Hit table first. 2 meaningful results: hit 3 times 4 numbers (draws #1, #8 and #12) and 4 times 3 numers (draws #4, #11, #14 and #16) - 7 potentially winning hits in 20 draws.

Now the winners table.

Out of 7 pottential winners we got 4 wins. Let's analyse 4 numbers hits. For the draw #1 it won 1 X 4 and 1 x 3, draw 8 was a complete dud, draw #12 (1 x 3) another money maker. So hitting 4 numbers with this wheel does not guarantee too much - the wheel is just too small. Nevertheless, in this example, it did generate a neat profit although too small to overly get excited about.

The next example: the same 10 latest draws, 3 first digits, 5 second digits but different wheel: 2W 2H 15N 10T. Hit are going to be the same but not the wins:

No profit but for the past 20 draws I spent only \$10. I can play like that, no ifs or buts.

Next small baby-wheel, other params the same: 3W 6H 15N 4T.

This time we nailed it. Any doubts that different wheels, even with similar parameters, may generate totally different results? Always remember: your prediction generates hits but it is the wheel that makes winners.

As I mentioned earlier my goal is to hunt for 5 and 5+ hits. One more example: predictions from 10 latest draws, 5 first digits, 6 second digits, wheel 2W 5H 30N 9T.

2 x 5+1 and 3 x 5 were hit. None won but at least I had a documented chance. That's what I want to see: chances. Without them how can I win?

A word of caution. I noticed this in Pick 3 and it seems to apply for jackpot lotteries as well. In Pick 3, if you select combos based on other combos you HAVE TO make sure that digits in your selections comply with the latest digit trends. If they don't you may be significantly reducing your chances of winning.

Now look at these stats. Numbers selected from the latest 10 draws using high frequency for both 1st and 2nd digits, with the wheel 2W 3H 15N 4T. Look at winners and ROI.

Compare with this: low frequency digits for 1st and 2nd digits, the rest the same as above.

What created such a huge difference in outcome? The quality of the digits, what else? My advice: if you select numbers from numbers, before you play, analyse the quality of digits in your selected numbers. Where do they stand in respect of their frequency? Exactly the same story as in Pick 3.

I could provide multitude of similar examples - digits can be analysed for 5 to 50 draws; there are 9 prediction methods; add to that all 300+ wheels, possibilities are almost endless.

One final thought: the power of wheels lies in not what they guarantee but in their ability to generate multiple winners in the same draw.

You can use this presentations as a blueprint for other jackpot lotteries. Is the digital approach to jackpot lotteries better than other statistical methods? Based on what I've seen so far I believe so, to the extent that I'm going to play it for real. The software is not finished yet and I need to do a dry run on real data to gain more practical experience how to handle this beast. Hopefully, the money will be worth the effort.

And that's all folks. Good luck in your playing.

Nice job of explaining what you are doing, start to finish. So many others will post their systems or methods and just assume we know what it is they are talking about. You make it very clear, step by step. Thank you for that.

Best of luck to you!

Thx Mr-B 216,

I hope that this clarity will help others to not only understand what's happening but also develop their own systems with similar features for the lotteries they can play.

Likewise, I have difficulties to understand majority of other systems presented on LP. Sometimes I have an impression that even inventors have only a vague idea how the whole system works. So, until I have a clear understanding of how the system operates and, even more important, what results it can produce, I wouldn't be inclined to look at it. And, unfortunately, the results are very seldom presented, like if the inventors were afraid of presenting them from fear of failure. So what? My system produces total crap at times, too, I've seen it already. The point is to accept it as a part of the system operation - and avoid it in the future. Nothing wins all the time - and nothing looses, either. Find out what works in your system, when and how. That's why I do so many simulations - to see what wins, when and why. That's the only way to evaluate merits of your system - and that's why I put so much emphasis on creating and analysing backtracks. In my estimate, developing backtrack routines took about 40% of the total development time (while they constitute only about 10% of the features) but this was the time truly worth spending. There is absolutely no substitute for the information backtracks can provide. I would even go as far as to say - don't play lotteries without them.

My next post demonstrates what I mean by the importance of backtracks. These results came up when I developed a new prediction method and applied backtracks to it. My eyeballs almost popped out - never expected that, and I'm not a newbie with lotteries and seen my share of successes and failures.

Part 7 - High score wheel

I'm still testing the whole system and barely scratched the surface of its capability. It's still being developed and refined but what I've seen so far is becoming more and interesting. I already showed that smaller wheels can make money. Well, not only. Look at the examples below. That's a large wheel, 40 tickets a game. I use the same sample data as before for comparison.

9 latest draws, 8 winners, unquestionably a winning cluster. If my math is correct the winnings amounted to \$495. Total paid = 9 x \$40 = \$360.

In the last 20 draws the wheel generated 83% ROI which means I played only 17 cents a ticket. That's what I call playing jackpot lottery cheap. Looks like it's possible.

Years ago I was proud that my similar wheel (also 40 tickets per draw, and I played in a pool) generated 52% ROI. Well, I'm still learning.

So the winning clusters do exist in jackpot lotteries as well. Now I know, I don't need to guess. Like in Pick 3 where winning clusters are quite common, certain conditions have to be met for a cluster to form. But what comes as a surprise to me is that this cluster lasted for so long - 9 draws??? That's rare even in Pick 3.

As I test the system I keep bumping into these surprising results over and over again. One can only guess - what else is in there? A cluster of jackpots? (just kidding).

Part 8 - The hidden power of wheels.

A very interesting example of a wheel that can generate multiple winners:
6W 6H 8N 28T (you should win jackpot if you hit 6 out of 8 numbers distributed into 28 tickets), 10 latest draws analysed:

Now look at the draw #18. Hit only 3 and bonus, not much by any standards. But look at the wins: 25 out of 28 tickets were winners. That's the hidden power of a wheel - it's ability to generate multiple winners even with a low hit.

And the very next draw, #19. Hit even less (2 + bonus only) but more than half of the tickets were winners.

Does it really matter what this wheel guarantees? It can be a big winner even if you hit well below the guarantee.

How many wheels can provide results like these with even low hits? At this moment I still don't know but I suspect many as I've already seen several.
Same wheel again, this time data from the 30 latest draws; look at draw #14:

Only 4 hits but 22 tickets were winners with the total - whoa! And you paid for this only \$28. For the 20 draws backtrack the wheel had 140% ROI. What is significant for this win is that it does not include any 2 + bonus wins which are BC49 lottery specific; all the wins are 4 and 3 number category - which can be scored in any other jackpot lottery.

BTW, these results are based on real data from the lottery, that's not the sample I used before. I'm testing the software now using the real numbers. Big job in itself as there are 9 prediction methods each with multiple wheels and many additional options. I'm learning myself as I go and some findings are simply stunning; I've never thought I will see anything like that.

I was aware before that some wheels may generate multiple winners if the hit is big (6, 5+, 5). What surprised me here is that even lower hits (4, 3, 2+) may also win substantial amounts.

I had a favourite wheel that on 2 occasions won 20 out of its 40 tickets (\$200 for the cost of \$40). I thought it outstanding at that time. I don't think this way any more, following what Stat\$Talker (I believe) says - numbers do not lie.

Now readers, judge for yourself - is it worth to study wheels? .....My apology for asking a stupid question. Now there is a proof that even if my predictions hit low I can still end up a pretty good winner - thanks to the wheel.

Aquarius

good job, lotteries are all about repetitions and delays, short medium and long, with their cycles
frequencies and jumps, etc. it's like a passing cloud, but we have a problem !! is that in lotteries mathematically you can only reach 80%
the remaining 20% will always be random, the best
wheels are positional ones, in ascending order
to match the draws in ascending order, the last digits are good filters but the positional in each column vertically, the use of the carteziano would be good, to position the filter within the matrix in its coordinates,

Part 9 - The inglorious finale.

First of all, my apologies to all LP members that got interested in this thread for the mess I created.

From the beginning I had good intentions to describe my new concept for playing jackpot lotteries. I tried to be clear, systematic and informative describing my newest software. Little I knew what I'm getting into, otherwise I would not have started this thread.

The idea behind the new system was to improve my ROI for a jackpot lottery. My original score was 52% and I thought of it as at least good. But I wanted to look for ways to improve it - maybe to 70-80% range with hopefully at least occasional break even.

I finished the software in mid October and then entered real draws from BC49 lottery till the end of October for thorough testing before actually playing. Things looked good, some ROIs went up and break evens became quite frequent. The testing process is slow as there are virtually hundreds of options (number of draws analysed in combination with multiple wheels). So far I was able to test only 5-10% of what's available.

And then came this, all of a sudden and without warning.

Draw #9 (in red circle), yellow line. That's not illusion, mirage or blurred vision. That's 225 grand coming from 3 wins 5 + bonus in one wheel. I give only one example but 2 different systems and 3 different wheels scored like this, apart from a dozen or so wheels that had "only" one 5 + bonus win for the "meagre" 75 grand each.

I hoped that 5 + bonus wins were a reasonable expectation but what stunned me was that they came so soon and in batches. This changes the value of the software dramatically. It's a different ball game now.

I had to re-evaluate my goals. It's not breaking even anymore, it's a much bigger game. And the software is not shareable anymore and had to be taken down.
After 10 years on LP I came to conclusion it's time to sign off. Not much is happening on LP these days and new and innovative ideas are non-existent. Besides, I got this tingling feeling in my butt that there is some cash waiting for me in the future and I want to dedicate some time to check if this is true.

Wish all the LP members success in their lottery endeavors. Good luck.

And the final note to Todd. I intentionally broke the LP rules: I have 2 accounts on LP (aquariuslottery and Pick3Max) and used the latter for this post. I guess Todd will now do what Todd is supposed to do - suspend both accounts permanently. Which is exactly what I would like him to do. He can take the whole thread down as well if he so wishes.

Sayonara

Wait what!!

How can I get a copy??

Cheers
MM

As one who was following this thread, I am happy to see your excitement with your findings. I have had the same experience with some of my own software. I also fully  understand you wanting to leave LP and spend your time tweaking your software. Like you, I am not interested in sharing my software at any price. Keep at it and go get a jackpot!