- Home
- Premium Memberships
- Lottery Results
- Forums
- Predictions
- Lottery Post Videos
- News
- Search Drawings
- Search Lottery Post
- Lottery Systems
- Lottery Charts
- Lottery Wheels
- Worldwide Jackpots
- Quick Picks
- On This Day in History
- Blogs
- Online Games
- Premium Features
- Contact Us
- Whitelist Lottery Post
- Rules
- Lottery Book Store
- Lottery Post Gift Shop
The time is now 3:37 am
You last visited
July 10, 2025, 2:34 am
All times shown are
Eastern Time (GMT-5:00)
New play methodologyPrev TopicNext Topic
-
Part 1.
I'm giving up on numbers - permanently. No more number analysis. Why? Law of averages made me do it.
Below a couple of definitions/descriptions of law of averages for better understanding (sources specified in parenthesis)
Law of averages (Merriam-Webster Dictionary)
The commonsense observation that probability influences everyday life so that over the long term the possible outcomes of a repeated event occur with specific frequencies
-----------------------------------------------------------------
Law of averages (Wikipedia)
The law of averages is the commonly held belief that a particular outcome or event will, over certain periods of time, occur at a frequency that is similar to its probability
-----------------------------------------------------------------
Law of averages (Math is Fun)
In the long run random events tend to average out at the expected value, but that does not help us predict the next value at all.
-----------------------------------------------------------------
Law of Large Numbers / Law of Averages (How statistics work)
The Law of Large Numbers shows us that if you take an unpredictable experiment and repeat it enough times, what you’ll end up with is an average.
A simple example: throw a die and you’ll get a random number (for a six-sided die, you’ll get 1,2,3,4,5,6). Throw the die 100,000 times and you’ll get an average of 3.5 — which is the expected value.In jackpot lotteries we always play groups. The smallest group is the size of a single ticket, the largest - the whole set in lottery. My lottery is BC49 with 6 regular & 1 bonus number drawn from the same pool of 49.
Each group has its statistical average hit (match) rate. It is easy to calculate the rate according to this formula:
C = (A * D) / B
A = numbers drawn in lottery (in my lottery 7)
B = all numbers in lottery (49)
C = average for a group (say 24 numbers)
D = group size (24 as desired)In this example C = 3.43 (rounded). That's the statistical average hit rate for a group of 24 numbers out of 49. That many hits can be expected when playing a group of 24 numbers. Actual average hit rates computed for a specific number of draws will jump around the statistical average within a certain (and usually limited) range (like 0.7 above or below the statistical average).
And that's where the problem begins. And it's a big one.Below are the statistical hit ratios for my lottery (group size & hit ratio):
7 1.00 15 2.14 23 3.29 31 4.43
8 1.14 16 2.29 24 3.43 32 4.57
9 1.29 17 2.43 25 3.57 33 4.71
10 1.43 18 2.57 26 3.71 34 4.86
11 1.57 19 2.71 27 3.86 35 5.00
12 1.51 20 2.86 28 4.00
13 1.86 21 3.00 29 4.14
14 2.00 22 3.14 30 4.29Sticking to the above example of 24 numbers group, with statistical hit ratio of 3.43.
What can I realistically (read: statistically) expect from playing this? Well, hitting 3 numbers quite frequently. I can't win more than I hit. So winning 3s is a realistic expectation as well. But not much more beyond that. Why? I just don't have enough numbers.
It is not my goal to end up winning 3s frequently. The goal is a bit more ambitious. But I'm not likely to get it with just 24 numbers (unless I could invent a utility that can stretch 3 number hits into 6 number winners but I won't be wasting any time for that). So I need to play bigger groups to increase my statistical chances. Where does this lead? To 49 number group, obviously. Play them all!
2 hypothetical scenarios:
1) Play a system that selects 15 numbers and distributes them into 10 tickets. (hit ratio: 2.14)
2) Buy 10 QP (or make your own) tickets that have all 49 numbers (hit ratio: 7.00)
Which one of those is more likely to win something? You can't win more than you hit.So much for playing small groups (and softwares or systems that advocate them). Plain waste of effort.
Good bye numbers, hello combinations. More in part 2.
-
Quote: Originally posted by Player649 on May 16, 2025
Part 1.
I'm giving up on numbers - permanently. No more number analysis. Why? Law of averages made me do it.
Below a couple of definitions/descriptions of law of averages for better understanding (sources specified in parenthesis)
Law of averages (Merriam-Webster Dictionary)
The commonsense observation that probability influences everyday life so that over the long term the possible outcomes of a repeated event occur with specific frequencies
-----------------------------------------------------------------
Law of averages (Wikipedia)
The law of averages is the commonly held belief that a particular outcome or event will, over certain periods of time, occur at a frequency that is similar to its probability
-----------------------------------------------------------------
Law of averages (Math is Fun)
In the long run random events tend to average out at the expected value, but that does not help us predict the next value at all.
-----------------------------------------------------------------
Law of Large Numbers / Law of Averages (How statistics work)
The Law of Large Numbers shows us that if you take an unpredictable experiment and repeat it enough times, what you’ll end up with is an average.
A simple example: throw a die and you’ll get a random number (for a six-sided die, you’ll get 1,2,3,4,5,6). Throw the die 100,000 times and you’ll get an average of 3.5 — which is the expected value.In jackpot lotteries we always play groups. The smallest group is the size of a single ticket, the largest - the whole set in lottery. My lottery is BC49 with 6 regular & 1 bonus number drawn from the same pool of 49.
Each group has its statistical average hit (match) rate. It is easy to calculate the rate according to this formula:
C = (A * D) / B
A = numbers drawn in lottery (in my lottery 7)
B = all numbers in lottery (49)
C = average for a group (say 24 numbers)
D = group size (24 as desired)In this example C = 3.43 (rounded). That's the statistical average hit rate for a group of 24 numbers out of 49. That many hits can be expected when playing a group of 24 numbers. Actual average hit rates computed for a specific number of draws will jump around the statistical average within a certain (and usually limited) range (like 0.7 above or below the statistical average).
And that's where the problem begins. And it's a big one.Below are the statistical hit ratios for my lottery (group size & hit ratio):
7 1.00 15 2.14 23 3.29 31 4.43
8 1.14 16 2.29 24 3.43 32 4.57
9 1.29 17 2.43 25 3.57 33 4.71
10 1.43 18 2.57 26 3.71 34 4.86
11 1.57 19 2.71 27 3.86 35 5.00
12 1.51 20 2.86 28 4.00
13 1.86 21 3.00 29 4.14
14 2.00 22 3.14 30 4.29Sticking to the above example of 24 numbers group, with statistical hit ratio of 3.43.
What can I realistically (read: statistically) expect from playing this? Well, hitting 3 numbers quite frequently. I can't win more than I hit. So winning 3s is a realistic expectation as well. But not much more beyond that. Why? I just don't have enough numbers.
It is not my goal to end up winning 3s frequently. The goal is a bit more ambitious. But I'm not likely to get it with just 24 numbers (unless I could invent a utility that can stretch 3 number hits into 6 number winners but I won't be wasting any time for that). So I need to play bigger groups to increase my statistical chances. Where does this lead? To 49 number group, obviously. Play them all!
2 hypothetical scenarios:
1) Play a system that selects 15 numbers and distributes them into 10 tickets. (hit ratio: 2.14)
2) Buy 10 QP (or make your own) tickets that have all 49 numbers (hit ratio: 7.00)
Which one of those is more likely to win something? You can't win more than you hit.So much for playing small groups (and softwares or systems that advocate them). Plain waste of effort.
Good bye numbers, hello combinations. More in part 2.
Part 2.
Hello combinations. All 13.9+M of you. Oops, how many? Rather a big oops, no kidding.
Cannot play that, that's for sure. Gotta shrink it somehow. But how?
The first thing that comes to mind is sums of numbers in combinations. Each combination has a sum of all its numbers which can easily be calculated. The sums are not unique to combinations (many combinations will have the same sum) but this will split 13.9+ M set into smaller chunks. How can we do that? Well, can my new friend Python help, possibly?
Yes, it could. Below is a partial list of the most popular sums among all the combinations (with at least 100K occurrences):
150 - 165772
149 - 165732
151 - 165732
148 - 165490
152 - 165490
147 - 165176
153 - 165176
146 - 164654
154 - 164654
145 - 164062
155 - 164062
144 - 163273
156 - 163273
143 - 162410
157 - 162410
142 - 161354
158 - 161354
141 - 160236
159 - 160236
140 - 158923
160 - 158923
139 - 157554
161 - 157554
138 - 156004
162 - 156004
137 - 154397
163 - 154397
136 - 152617
164 - 152617
135 - 150794
165 - 150794
134 - 148800
166 - 148800
133 - 146771
167 - 146771
132 - 144587
168 - 144587
131 - 142370
169 - 142370
130 - 140008
170 - 140008
129 - 137629
171 - 137629
128 - 135109
172 - 135109
127 - 132581
173 - 132581
126 - 129930
174 - 129930
125 - 127274
175 - 127274
124 - 124507
176 - 124507
123 - 121751
177 - 121751
122 - 118889
178 - 118889
121 - 116048
179 - 116048
120 - 113119
180 - 113119
119 - 110215
181 - 110215
118 - 107235
182 - 107235
117 - 104295
183 - 104295
116 - 101285
184 - 101285Anyone of them is still way too big to be of practical use. Need to shrink them further. But how?
Every combination has a feature called its digital composition. Split each number into its FD (front digit) & BD (back digit). How many unique FDs are there in a combination, how many unique BDs? What are the results for all combinations? Python, go and get it!
This is what Python got (I call these digit spreads):
FD spread
FD: 0 - 0
FD: 1 - 924
FD: 2 - 337392 2.41%
FD: 3 - 4207500 32.23%
FD: 4 - 7458000 53.33%
FD: 5 - 1980000 14.16%BD spread
BD: 0 - 0
BD: 1 - 0
BD: 2 - 8316
BD: 3 - 462000 3.30%
BD: 4 - 3696000 26.43%
BD: 5 - 6930000 49.56%
BD: 6 - 2887500 20.65%
BD: 7 - 0
BD: 8 - 0
BD: 9 - 0About 86% of all combinations will have 4 different FDs and about 79% 5 different BDs.
Here is what I'm going to use. I must have all 5 FDs and 6 unique BDs in every line. The former because I want to have all numbers, the latter because I intentionally want to extend this range from 5 to 6. Each line in a sum group must meet this condition. Python, can you do it?
I don't think there is much Python cannot do. It just breezed through this, formidable after all, task.
Here are the results for the most popular sum 150:
List sum150 full - 165772 combinations
List sum150 reduced - 6480 combinations (96.09% reduction)Hey, that's a pretty decent reduction. Other big sums had similar around 95% reductions as well.
Now I can do something with this. What - in a moment.Back in October last year I created a utility for evaluating wheel potential (or capacity): how many actual winners a wheel can generate with all possible hit configurations. I ran several of my favorite wheels through it and got some idea what these wheels can and cannot achieve. Neat. All the wheels were balanced because that was my preference. But how about imbalanced wheels? Are they better or worse? What's the difference?
The results shattered my wheeling world. There was no difference! I even tried to really mess up the balance to a degree to make the wheel almost unplayable - and there was not a iota of difference. Wheels with the same number of numbers and lines (tickets) have exactly the same winning potential! The internal structure means nothing! For decades I was under impression that the structure (number distribution) is a deciding factor - and it is not. Don't bother your a$$ about this anymore, its meaningless.
But, after the shock subsided, I came to a conclusion that this may not be a bad thing after all. Because it does not matter how we build a wheel, manually, through software, or even randomly - all wheels with the same number of numbers and lines will have the same winning potential.
Did I say software and random? What do you think about this, Python? That's the areas of your expertise. You have homogeneous sums with lines of the same parameters, why don't you put them into wheels of desired size, say 15, 20, 25 & 30 lines? It would be a manual long term pain for me to do but it's a random snap for you, Python.
Python agreed and that's what it produced. A few wheel examples for the sum 127.
15 ticket wheel
7,8,16,23,32,41
2,5,14,26,39,41
1,4,12,25,36,49
6,9,15,22,31,44
6,8,10,21,39,43
2,11,16,20,35,43
4,5,10,27,39,42
4,5,12,27,33,46
3,6,17,24,32,45
2,4,19,25,30,47
1,6,13,25,34,48
4,5,12,28,37,41
5,8,11,24,36,43
1,2,18,29,37,40
5,7,10,24,38,4320 ticket wheel
2,7,15,20,39,44
6,9,11,27,30,44
4,7,13,25,38,40
3,7,14,28,35,40
4,10,17,22,31,43
2,7,10,23,36,49
3,7,15,28,34,40
2,5,19,23,37,41
1,4,15,22,37,48
2,3,15,26,34,47
3,9,15,28,30,42
1,3,18,22,37,46
3,7,12,26,39,40
2,5,16,24,39,41
7,8,12,29,30,41
1,4,17,28,32,45
3,5,14,21,36,48
2,9,14,28,33,41
1,6,13,28,37,42
2,6,17,20,39,4325 ticket wheel
3,6,15,28,31,44
5,9,11,26,32,44
3,9,10,25,34,46
2,5,14,21,39,46
3,9,12,21,37,45
3,7,12,21,38,46
2,6,17,29,30,43
2,13,14,20,31,47
4,9,10,23,36,45
1,7,15,20,38,46
4,8,17,20,32,46
1,3,18,24,35,46
5,6,12,23,34,47
1,8,19,27,32,40
3,6,12,27,34,45
2,3,11,28,37,46
2,9,17,20,34,45
5,6,11,23,34,48
4,5,11,22,38,47
2,9,16,25,34,41
1,6,15,22,39,44
1,10,17,22,33,44
2,4,17,23,36,45
5,13,16,21,30,42
3,5,17,22,31,49and finally
30 ticket wheel
2,4,19,27,30,45
3,5,14,27,38,40
6,9,10,27,33,42
3,9,11,27,32,45
5,9,12,27,30,44
5,8,17,24,33,40
1,4,18,25,37,42
4,7,10,23,35,48
5,7,13,22,36,44
5,7,14,22,31,48
1,6,10,25,38,47
5,7,19,23,31,42
2,3,17,26,38,41
1,4,16,28,33,45
3,9,15,26,30,44
1,9,14,25,36,42
1,8,17,26,32,43
2,3,11,29,35,47
3,9,14,21,38,42
4,5,10,26,33,49
7,8,13,25,30,44
4,7,12,20,36,48
1,8,16,29,30,43
5,8,11,29,34,40
2,8,14,27,36,40
6,8,14,25,31,43
3,14,17,20,31,42
6,7,11,20,35,48
4,5,13,20,39,46
4,9,15,20,37,42All these wheels must meet 2 additional conditions (in addition to the conditions specified above regarding the sum, FDs & BDs in each line): every wheel must have full set of 49 numbers, regardless of its size, and no duplicate lines are allowed. Do you realize what this means? It eliminates the effects of randomness! It does not matter how the numbers are drawn - fully random, quasi random, this random, that random - all numbers are always hit. Remember, you cannot win more than you hit. To win big you must hit big. Now I always have a chance to win big because I always have all the numbers. This chance does not exists when playing small groups.
None of the wheels is balanced. It does not matter. So, how do I know which one to play? All wheels have a feature called the current performance in a given period of time. In other words, wheels can be tested to compare which ones are better than others. Something that Python can do with a smile (and make me smile, too). What a buddy!
Why bother to test when they all have the same potential? Because their time frame performance will not be the same, and there are usually big differences. My overall strategy is to always play the best of the best. And to do that I need to know what is that best.
More about it - in Part 3.
-
I play the Lotto 649, as well as Atlantic 49. As well the Lotto Max.They are really difficult game's to figure out. Also, I cannot play too many combination's at one time. Just 1 or 2 combination's per- draw!
-
Quote: Originally posted by Player649 on May 17, 2025
Part 2.
Hello combinations. All 13.9+M of you. Oops, how many? Rather a big oops, no kidding.
Cannot play that, that's for sure. Gotta shrink it somehow. But how?
The first thing that comes to mind is sums of numbers in combinations. Each combination has a sum of all its numbers which can easily be calculated. The sums are not unique to combinations (many combinations will have the same sum) but this will split 13.9+ M set into smaller chunks. How can we do that? Well, can my new friend Python help, possibly?
Yes, it could. Below is a partial list of the most popular sums among all the combinations (with at least 100K occurrences):
150 - 165772
149 - 165732
151 - 165732
148 - 165490
152 - 165490
147 - 165176
153 - 165176
146 - 164654
154 - 164654
145 - 164062
155 - 164062
144 - 163273
156 - 163273
143 - 162410
157 - 162410
142 - 161354
158 - 161354
141 - 160236
159 - 160236
140 - 158923
160 - 158923
139 - 157554
161 - 157554
138 - 156004
162 - 156004
137 - 154397
163 - 154397
136 - 152617
164 - 152617
135 - 150794
165 - 150794
134 - 148800
166 - 148800
133 - 146771
167 - 146771
132 - 144587
168 - 144587
131 - 142370
169 - 142370
130 - 140008
170 - 140008
129 - 137629
171 - 137629
128 - 135109
172 - 135109
127 - 132581
173 - 132581
126 - 129930
174 - 129930
125 - 127274
175 - 127274
124 - 124507
176 - 124507
123 - 121751
177 - 121751
122 - 118889
178 - 118889
121 - 116048
179 - 116048
120 - 113119
180 - 113119
119 - 110215
181 - 110215
118 - 107235
182 - 107235
117 - 104295
183 - 104295
116 - 101285
184 - 101285Anyone of them is still way too big to be of practical use. Need to shrink them further. But how?
Every combination has a feature called its digital composition. Split each number into its FD (front digit) & BD (back digit). How many unique FDs are there in a combination, how many unique BDs? What are the results for all combinations? Python, go and get it!
This is what Python got (I call these digit spreads):
FD spread
FD: 0 - 0
FD: 1 - 924
FD: 2 - 337392 2.41%
FD: 3 - 4207500 32.23%
FD: 4 - 7458000 53.33%
FD: 5 - 1980000 14.16%BD spread
BD: 0 - 0
BD: 1 - 0
BD: 2 - 8316
BD: 3 - 462000 3.30%
BD: 4 - 3696000 26.43%
BD: 5 - 6930000 49.56%
BD: 6 - 2887500 20.65%
BD: 7 - 0
BD: 8 - 0
BD: 9 - 0About 86% of all combinations will have 4 different FDs and about 79% 5 different BDs.
Here is what I'm going to use. I must have all 5 FDs and 6 unique BDs in every line. The former because I want to have all numbers, the latter because I intentionally want to extend this range from 5 to 6. Each line in a sum group must meet this condition. Python, can you do it?
I don't think there is much Python cannot do. It just breezed through this, formidable after all, task.
Here are the results for the most popular sum 150:
List sum150 full - 165772 combinations
List sum150 reduced - 6480 combinations (96.09% reduction)Hey, that's a pretty decent reduction. Other big sums had similar around 95% reductions as well.
Now I can do something with this. What - in a moment.Back in October last year I created a utility for evaluating wheel potential (or capacity): how many actual winners a wheel can generate with all possible hit configurations. I ran several of my favorite wheels through it and got some idea what these wheels can and cannot achieve. Neat. All the wheels were balanced because that was my preference. But how about imbalanced wheels? Are they better or worse? What's the difference?
The results shattered my wheeling world. There was no difference! I even tried to really mess up the balance to a degree to make the wheel almost unplayable - and there was not a iota of difference. Wheels with the same number of numbers and lines (tickets) have exactly the same winning potential! The internal structure means nothing! For decades I was under impression that the structure (number distribution) is a deciding factor - and it is not. Don't bother your a$$ about this anymore, its meaningless.
But, after the shock subsided, I came to a conclusion that this may not be a bad thing after all. Because it does not matter how we build a wheel, manually, through software, or even randomly - all wheels with the same number of numbers and lines will have the same winning potential.
Did I say software and random? What do you think about this, Python? That's the areas of your expertise. You have homogeneous sums with lines of the same parameters, why don't you put them into wheels of desired size, say 15, 20, 25 & 30 lines? It would be a manual long term pain for me to do but it's a random snap for you, Python.
Python agreed and that's what it produced. A few wheel examples for the sum 127.
15 ticket wheel
7,8,16,23,32,41
2,5,14,26,39,41
1,4,12,25,36,49
6,9,15,22,31,44
6,8,10,21,39,43
2,11,16,20,35,43
4,5,10,27,39,42
4,5,12,27,33,46
3,6,17,24,32,45
2,4,19,25,30,47
1,6,13,25,34,48
4,5,12,28,37,41
5,8,11,24,36,43
1,2,18,29,37,40
5,7,10,24,38,4320 ticket wheel
2,7,15,20,39,44
6,9,11,27,30,44
4,7,13,25,38,40
3,7,14,28,35,40
4,10,17,22,31,43
2,7,10,23,36,49
3,7,15,28,34,40
2,5,19,23,37,41
1,4,15,22,37,48
2,3,15,26,34,47
3,9,15,28,30,42
1,3,18,22,37,46
3,7,12,26,39,40
2,5,16,24,39,41
7,8,12,29,30,41
1,4,17,28,32,45
3,5,14,21,36,48
2,9,14,28,33,41
1,6,13,28,37,42
2,6,17,20,39,4325 ticket wheel
3,6,15,28,31,44
5,9,11,26,32,44
3,9,10,25,34,46
2,5,14,21,39,46
3,9,12,21,37,45
3,7,12,21,38,46
2,6,17,29,30,43
2,13,14,20,31,47
4,9,10,23,36,45
1,7,15,20,38,46
4,8,17,20,32,46
1,3,18,24,35,46
5,6,12,23,34,47
1,8,19,27,32,40
3,6,12,27,34,45
2,3,11,28,37,46
2,9,17,20,34,45
5,6,11,23,34,48
4,5,11,22,38,47
2,9,16,25,34,41
1,6,15,22,39,44
1,10,17,22,33,44
2,4,17,23,36,45
5,13,16,21,30,42
3,5,17,22,31,49and finally
30 ticket wheel
2,4,19,27,30,45
3,5,14,27,38,40
6,9,10,27,33,42
3,9,11,27,32,45
5,9,12,27,30,44
5,8,17,24,33,40
1,4,18,25,37,42
4,7,10,23,35,48
5,7,13,22,36,44
5,7,14,22,31,48
1,6,10,25,38,47
5,7,19,23,31,42
2,3,17,26,38,41
1,4,16,28,33,45
3,9,15,26,30,44
1,9,14,25,36,42
1,8,17,26,32,43
2,3,11,29,35,47
3,9,14,21,38,42
4,5,10,26,33,49
7,8,13,25,30,44
4,7,12,20,36,48
1,8,16,29,30,43
5,8,11,29,34,40
2,8,14,27,36,40
6,8,14,25,31,43
3,14,17,20,31,42
6,7,11,20,35,48
4,5,13,20,39,46
4,9,15,20,37,42All these wheels must meet 2 additional conditions (in addition to the conditions specified above regarding the sum, FDs & BDs in each line): every wheel must have full set of 49 numbers, regardless of its size, and no duplicate lines are allowed. Do you realize what this means? It eliminates the effects of randomness! It does not matter how the numbers are drawn - fully random, quasi random, this random, that random - all numbers are always hit. Remember, you cannot win more than you hit. To win big you must hit big. Now I always have a chance to win big because I always have all the numbers. This chance does not exists when playing small groups.
None of the wheels is balanced. It does not matter. So, how do I know which one to play? All wheels have a feature called the current performance in a given period of time. In other words, wheels can be tested to compare which ones are better than others. Something that Python can do with a smile (and make me smile, too). What a buddy!
Why bother to test when they all have the same potential? Because their time frame performance will not be the same, and there are usually big differences. My overall strategy is to always play the best of the best. And to do that I need to know what is that best.
More about it - in Part 3.
Part 3.
Now we come to the subject of testing wheels. How do we measure their performance? How do we compare them to select which one is better?
There are 2 such measuring yardsticks:
1) ROI (Return-On-Investment): the comparison between what the wheel generated in cash versus what it cost, in percent
2) P/L (Profit/Loss) - the difference between what the wheel generated in cash versus what it cost, in dollars
These can be measured in a given period of time (10, 20, 50, 100 draws, or whatever the fancy)The most relevant is ROI. It provides the best comparison for wheel performance. I decided to use 2 such measurement:
ROI 4 - for all winners of 3 & 4
ROI Full - for all possible winners, including biggies (5, 5+, 6)System is ready but gotta test it before any thought of using it.
Python, for 5 sums: 127, 138, 140, 150, 152 go and generate wheels for 15, 20, 25, 30 tickets, 20 wheels for each ticket size, for the total of 400. This should be sufficient to obtain statistically reliable results. Python obliged without a grunt.
The results clearly indicated that the best ROIs were obtained by smaller wheels: 15 tickets. 20 tickets ROIs were close behind while 25 & 30 clearly lagged behind. The decision was made to keep only 15 ticket wheels and, temporarily at least, ditch all the others.
5 sums are clearly not enough. I must cover more. So I decided to cover all the sums with popularity of at least 150K. That's 31 of them: 135 to 165. All these sums got 20 15 ticket wheels each, for the total of 620. I cannot handle that many myself but Python can - in a blitz.
They all got to be tested, no exceptions. I used real lottery draws until March 29/25. Here are the results. The qualification parameter was ROI 4 = 50% or more.
Backtracking 100 draws
S136-T15-wheel(10)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 4
Win 3: 39
Win 2+: 15
Total wins: $765
Total cost: $1500
ROI 4: 51%
ROI Full: 51%
------------------------------------
S144-T15-wheel(16)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 5
Win 3: 38
Win 2+: 20
Total wins: $855
Total cost: $1500
ROI 4: 56%
ROI Full: 56%
------------------------------------
S146-T15-wheel(15)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 5
Win 3: 29
Win 2+: 22
Total wins: $775
Total cost: $1500
ROI 4: 51%
ROI Full: 51%
------------------------------------
S147-T15-wheel(12)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 8
Win 3: 32
Win 2+: 17
Total wins: $1005
Total cost: $1500
ROI 4: 67%
ROI Full: 67%
------------------------------------
S154-T15-wheel(1)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 5
Win 3: 31
Win 2+: 24
Total wins: $805
Total cost: $1500
ROI 4: 53%
ROI Full: 53%
------------------------------------
S156-T15-wheel(2)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 5
Win 3: 28
Win 2+: 22
Total wins: $765
Total cost: $1500
ROI 4: 51%
ROI Full: 51%
------------------------------------
S156-T15-wheel(18)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 4
Win 3: 39
Win 2+: 20
Total wins: $790
Total cost: $1500
ROI 4: 52%
ROI Full: 52%
------------------------------------
S157-T15-wheel(18)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 5
Win 3: 34
Win 2+: 31
Total wins: $870
Total cost: $1500
ROI 4: 57%
ROI Full: 57%
------------------------------------
S164-T15-wheel(4)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 6
Win 3: 30
Win 2+: 21
Total wins: $855
Total cost: $1500
ROI 4: 56%
ROI Full: 56%
------------------------------------
S165-T15-wheel(10)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 6
Win 3: 28
Win 2+: 27
Total wins: $865
Total cost: $1500
ROI 4: 57%
ROI Full: 57%
------------------------------------Another list of top performers (for a shorter time frame):
Backtracking 50 draws
S135-T15-wheel(13)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 3
Win 3: 18
Win 2+: 11
Total wins: $460
Total cost: $750
ROI 4: 61%
ROI Full: 61%
------------------------------------
S137-T15-wheel(4)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 2
Win 3: 20
Win 2+: 9
Total wins: $395
Total cost: $750
ROI 4: 52%
ROI Full: 52%
------------------------------------
S139-T15-wheel(8)
Win 6: 0
Win 5+: 0
Win 5: 1
Win 4: 3
Win 3: 14
Win 2+: 5
Total wins: $1140
Total cost: $750
ROI 4: 52%
ROI Full: 152%
------------------------------------
S142-T15-wheel(4)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 2
Win 3: 20
Win 2+: 15
Total wins: $425
Total cost: $750
ROI 4: 56%
ROI Full: 56%
------------------------------------
S144-T15-wheel(16)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 3
Win 3: 14
Win 2+: 7
Total wins: $400
Total cost: $750
ROI 4: 53%
ROI Full: 53%
------------------------------------
S145-T15-wheel(1)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 3
Win 3: 17
Win 2+: 6
Total wins: $425
Total cost: $750
ROI 4: 56%
ROI Full: 56%
------------------------------------
S145-T15-wheel(20)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 3
Win 3: 11
Win 2+: 14
Total wins: $405
Total cost: $750
ROI 4: 54%
ROI Full: 54%
------------------------------------
S146-T15-wheel(12)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 2
Win 3: 20
Win 2+: 10
Total wins: $400
Total cost: $750
ROI 4: 53%
ROI Full: 53%
------------------------------------
S147-T15-wheel(12)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 6
Win 3: 14
Win 2+: 10
Total wins: $640
Total cost: $750
ROI 4: 85%
ROI Full: 85%
------------------------------------
S149-T15-wheel(9)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 2
Win 3: 17
Win 2+: 14
Total wins: $390
Total cost: $750
ROI 4: 52%
ROI Full: 52%
------------------------------------
S151-T15-wheel(1)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 2
Win 3: 17
Win 2+: 13
Total wins: $385
Total cost: $750
ROI 4: 51%
ROI Full: 51%
------------------------------------
S158-T15-wheel(14)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 3
Win 3: 12
Win 2+: 11
Total wins: $400
Total cost: $750
ROI 4: 53%
ROI Full: 53%
------------------------------------
S160-T15-wheel(3)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 3
Win 3: 14
Win 2+: 10
Total wins: $415
Total cost: $750
ROI 4: 55%
ROI Full: 55%
------------------------------------
S160-T15-wheel(5)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 3
Win 3: 12
Win 2+: 14
Total wins: $415
Total cost: $750
ROI 4: 55%
ROI Full: 55%
------------------------------------
S161-T15-wheel(2)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 3
Win 3: 18
Win 2+: 10
Total wins: $455
Total cost: $750
ROI 4: 60%
ROI Full: 60%
------------------------------------
S162-T15-wheel(5)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 3
Win 3: 13
Win 2+: 13
Total wins: $420
Total cost: $750
ROI 4: 56%
ROI Full: 56%
------------------------------------
S163-T15-wheel(3)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 2
Win 3: 16
Win 2+: 18
Total wins: $400
Total cost: $750
ROI 4: 53%
ROI Full: 53%
------------------------------------
S163-T15-wheel(14)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 3
Win 3: 13
Win 2+: 15
Total wins: $430
Total cost: $750
ROI 4: 57%
ROI Full: 57%
------------------------------------
S164-T15-wheel(4)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 3
Win 3: 17
Win 2+: 12
Total wins: $455
Total cost: $750
ROI 4: 60%
ROI Full: 60%
------------------------------------
S165-T15-wheel(10)
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 3
Win 3: 14
Win 2+: 19
Total wins: $460
Total cost: $750
ROI 4: 61%
ROI Full: 61%
------------------------------------
S165-T15-wheel(15)
Win 6: 0
Win 5+: 0
Win 5: 1
Win 4: 4
Win 3: 19
Win 2+: 10
Total wins: $1290
Total cost: $750
ROI 4: 72%
ROI Full: 172%
------------------------------------And this test is the most exciting of all: ROI Full above 80%. All these wheels had a 5 numbers win.
Backtracking 100 draws
S136-T15-wheel(7)
ROI 4: 36%
ROI Full: 86%
------------------------------------
S138-T15-wheel(7)
ROI 4: 35%
ROI Full: 85%
------------------------------------
S139-T15-wheel(8)
ROI 4: 38%
ROI Full: 88%
------------------------------------
S142-T15-wheel(15)
ROI 4: 41%
ROI Full: 91%
------------------------------------
S143-T15-wheel(2)
ROI 4: 33%
ROI Full: 83%
------------------------------------
S143-T15-wheel(7)
ROI 4: 34%
ROI Full: 84%
------------------------------------
S148-T15-wheel(14)
ROI 4: 33%
ROI Full: 83%
------------------------------------
S149-T15-wheel(20)
ROI 4: 34%
ROI Full: 84%
------------------------------------
S152-T15-wheel(4)
ROI 4: 43%
ROI Full: 93%
------------------------------------
S153-T15-wheel(9)
ROI 4: 33%
ROI Full: 83%
------------------------------------
S165-T15-wheel(15)
ROI 4: 49%
ROI Full: 99%
------------------------------------Traditional playing involves 2 steps:
1. Number selection
2. Number distribution (into tickets)Distribution is the one that generates cash, not selections. As this methodology plainly demontrates the selection step can safely be discarded. Sounds radical and definitely does not fit in the traditional logic box but that's also a statistical reality, fully supported by objective results. That't the reason I decided to break from numbers.
The testing is done, the results are in, what's possible and likely are both well defined within their ranges, now it's time to play. Just need to update the data file because it runs only till end of March.
This methodology can be applied equally well to all jackpot lotteries, big or small, even monsters like MM or PB. What it can deliver in each of them? Well, this needs to be tested to define ranges of probabilities which will be different for different lotteries. Playing lotteries is all about statistical probabilities. And, as you can see, a lot of meaningful statistics can be run about lotteries. The ones I used are just a small sample.
-
How is this going to make it possible for a dollar wager to return more than a dollar in expected wins? Most of your tests are in the 50% range, which is the game's paytables (the average of all combinations). Past drawings do not guarantee future wins.
-
Quote: Originally posted by Tucker Black on May 18, 2025
How is this going to make it possible for a dollar wager to return more than a dollar in expected wins? Most of your tests are in the 50% range, which is the game's paytables (the average of all combinations). Past drawings do not guarantee future wins.
How is this going to make it possible for a dollar wager to return more than a dollar in expected wins?
You can turn a profit in jackpot lottery in a short term. A couple of examples from some wheels for the same time period (ending March 29). In longer terms, however, this is not very likely.
Backtracking 25 draws
S138-T15-wheel(5) - 252%
Win 6: 0
Win 5+: 0
Win 5: 1
Win 4: 1
Win 3: 11
Win 2+: 2
Total wins: $945
Total cost: $375
ROI Full: 252%
------------------------------------
S139-T15-wheel(8) - 221%
Win 6: 0
Win 5+: 0
Win 5: 1
Win 4: 0
Win 3: 6
Win 2+: 4
Total wins: $830
Total cost: $375
ROI Full: 221%
------------------------------------
S140-T15-wheel(11) - 229%
Win 6: 0
Win 5+: 0
Win 5: 1
Win 4: 0
Win 3: 8
Win 2+: 6
Total wins: $860
Total cost: $375
ROI Full: 229%
------------------------------------
S146-T15-wheel(15) - 258%
Win 6: 0
Win 5+: 0
Win 5: 1
Win 4: 2
Win 3: 4
Win 2+: 6
Total wins: $970
Total cost: $375
ROI Full: 258%
------------------------------------
S147-T15-wheel(2) - 227%
Win 6: 0
Win 5+: 0
Win 5: 1
Win 4: 0
Win 3: 8
Win 2+: 5
Total wins: $855
Total cost: $375
ROI Full: 227%
------------------------------------
S148-T15-wheel(14) - 222%
Win 6: 0
Win 5+: 0
Win 5: 1
Win 4: 0
Win 3: 5
Win 2+: 7
Total wins: $835
Total cost: $375
ROI Full: 222%
------------------------------------
S149-T15-wheel(20) - 241%
Win 6: 0
Win 5+: 0
Win 5: 1
Win 4: 1
Win 3: 7
Win 2+: 2
Total wins: $905
Total cost: $375
ROI Full: 241%
------------------------------------
S153-T15-wheel(9) - 221%
Win 6: 0
Win 5+: 0
Win 5: 1
Win 4: 0
Win 3: 5
Win 2+: 6
Total wins: $830
Total cost: $375
ROI Full: 221%
------------------------------------
S153-T15-wheel(11) - 105%
Win 6: 0
Win 5+: 0
Win 5: 0
Win 4: 4
Win 3: 8
Win 2+: 3
Total wins: $395
Total cost: $375
ROI Full: 105%
------------------------------------
S160-T15-wheel(15) - 229%
Win 6: 0
Win 5+: 0
Win 5: 1
Win 4: 0
Win 3: 7
Win 2+: 8
Total wins: $860
Total cost: $375
ROI Full: 229%
------------------------------------Most of your tests are in the 50% range, which is the game's paytables (the average of all combinations).
The ROI 4 that was measured for 50% returns includes only wins for 2+, 3 & 4 numbers, and does not include higher wins (5, 5+, 6) and thus is not the average for all combinations. The lottery average for 2+, 3 & 4 winners in only 31%, not 50% as you stated, 20%+ difference. You are comparing apples to oranges.
Take 100 draws at 15 tickets per draw with 50% ROI. The cost is 15 x 100 = $1500; 50% ROI reduces this to $750.
Now 100 draws at 15 tickets per draw with 30% ROI. The cost is 15 x 100 = $1500; 30% ROI reduces this to $1050.
That's $300 yearly cost reduction difference.
Double this because I'm going to play $30 per game and the difference becomes quite significant.
And this does not include possible 5 number wins
Past drawings do not guarantee future wins.I'm sure you know that there no guarantees in lotteries. However, there are probabilities and they do play a role in predictability. This system is designed to maximize chances of winning something big by using all available numbers in the lottery (49). Each wheel has all the numbers. This eliminates effects of randomness because all drawn numbers are always matched. Thus there is no need to project any past numbers into the future because statistics related to past numbers are meaningless.
-
You could bet on all the numbers but they will be on different lines. You could buy all the combinations but that would be expensive, time consuming, and a guaranteed loss. Sorry but I'm just not getting this, that your system is going to improve your odds of winning.
-
So this is what you are saying:
You’re shifting away from traditional number-picking methods and instead using a statistical and combinatorial approach to playing the lottery. Here’s what that means in practice:
-
Instead of picking numbers based on patterns, birthdays, or "hot/cold" numbers, you focus on the math behind how groups of numbers perform on average.
-
You realize that playing only a small set of numbers (like 15 or 24) limits your chances of hitting the jackpot, because the expected number of matches is low based on probability.
-
To improve your chances, you analyze all possible combinations and use filters—like the sum of the numbers or the spread of digits—to dramatically reduce the number of combinations you play, making it more practical and cost-effective.
-
Your approach is less about luck or superstition and more about maximizing coverage and efficiency using mathematical reasoning.
Examples to illustrate your approach:
-
If you play 24 numbers out of 49, math shows you’ll average about 3 or 4 matches per draw—not enough for a jackpot.
-
By filtering combinations to only those with a specific sum (for example, all combinations that add up to 150), you reduce millions of possibilities down to a much smaller group, making it manageable.
-
Adding another filter, such as requiring each combination to have a wide spread of digits (not just numbers in the 20s or 30s), cuts down the number of combinations even further.
-
Instead of picking numbers based on birthdays (which most people do, leading to more shared jackpots), you use statistical filters to select combinations that are less likely to be picked by others and more likely to cover the range of winning numbers
-
You avoid common mistakes like clustering numbers, only picking low numbers, or relying on personal "lucky" numbers, focusing instead on balanced and well-distributed combinations
your strategy is about playing smarter, not harder, by using probability and combinatorial math rather than intuition or tradition.
-
-
700 light yrs West of Milky Way Galaxy's Center
United States
Member #200,634
September 1, 2019
4,903 Posts
OfflineQuote: Originally posted by Player649 on May 16, 2025
Part 1.
I'm giving up on numbers - permanently. No more number analysis. Why? Law of averages made me do it.
Below a couple of definitions/descriptions of law of averages for better understanding (sources specified in parenthesis)
Law of averages (Merriam-Webster Dictionary)
The commonsense observation that probability influences everyday life so that over the long term the possible outcomes of a repeated event occur with specific frequencies
-----------------------------------------------------------------
Law of averages (Wikipedia)
The law of averages is the commonly held belief that a particular outcome or event will, over certain periods of time, occur at a frequency that is similar to its probability
-----------------------------------------------------------------
Law of averages (Math is Fun)
In the long run random events tend to average out at the expected value, but that does not help us predict the next value at all.
-----------------------------------------------------------------
Law of Large Numbers / Law of Averages (How statistics work)
The Law of Large Numbers shows us that if you take an unpredictable experiment and repeat it enough times, what you’ll end up with is an average.
A simple example: throw a die and you’ll get a random number (for a six-sided die, you’ll get 1,2,3,4,5,6). Throw the die 100,000 times and you’ll get an average of 3.5 — which is the expected value.In jackpot lotteries we always play groups. The smallest group is the size of a single ticket, the largest - the whole set in lottery. My lottery is BC49 with 6 regular & 1 bonus number drawn from the same pool of 49.
Each group has its statistical average hit (match) rate. It is easy to calculate the rate according to this formula:
C = (A * D) / B
A = numbers drawn in lottery (in my lottery 7)
B = all numbers in lottery (49)
C = average for a group (say 24 numbers)
D = group size (24 as desired)In this example C = 3.43 (rounded). That's the statistical average hit rate for a group of 24 numbers out of 49. That many hits can be expected when playing a group of 24 numbers. Actual average hit rates computed for a specific number of draws will jump around the statistical average within a certain (and usually limited) range (like 0.7 above or below the statistical average).
And that's where the problem begins. And it's a big one.Below are the statistical hit ratios for my lottery (group size & hit ratio):
7 1.00 15 2.14 23 3.29 31 4.43
8 1.14 16 2.29 24 3.43 32 4.57
9 1.29 17 2.43 25 3.57 33 4.71
10 1.43 18 2.57 26 3.71 34 4.86
11 1.57 19 2.71 27 3.86 35 5.00
12 1.51 20 2.86 28 4.00
13 1.86 21 3.00 29 4.14
14 2.00 22 3.14 30 4.29Sticking to the above example of 24 numbers group, with statistical hit ratio of 3.43.
What can I realistically (read: statistically) expect from playing this? Well, hitting 3 numbers quite frequently. I can't win more than I hit. So winning 3s is a realistic expectation as well. But not much more beyond that. Why? I just don't have enough numbers.
It is not my goal to end up winning 3s frequently. The goal is a bit more ambitious. But I'm not likely to get it with just 24 numbers (unless I could invent a utility that can stretch 3 number hits into 6 number winners but I won't be wasting any time for that). So I need to play bigger groups to increase my statistical chances. Where does this lead? To 49 number group, obviously. Play them all!
2 hypothetical scenarios:
1) Play a system that selects 15 numbers and distributes them into 10 tickets. (hit ratio: 2.14)
2) Buy 10 QP (or make your own) tickets that have all 49 numbers (hit ratio: 7.00)
Which one of those is more likely to win something? You can't win more than you hit.So much for playing small groups (and softwares or systems that advocate them). Plain waste of effort.
Good bye numbers, hello combinations. More in part 2.
"I'm giving up on numbers - permanently. No more number analysis. Why?"
Bcuzzz, Ya
didn't make it back to shore like Ole Stat$ told ya to...naah?You've run outta Air
Gluuk , Gluuk
-Stat$talker
-
Quote: Originally posted by Tucker Black on May 20, 2025
You could bet on all the numbers but they will be on different lines. You could buy all the combinations but that would be expensive, time consuming, and a guaranteed loss. Sorry but I'm just not getting this, that your system is going to improve your odds of winning.
Playing all the numbers does improve chances for winning, particularly bigger prizes.
I used to play 24 number groups. My biggest problem was that the groups of 24 were missing numbers well too often for my comfort. In order to spread the risk I even played 5 different groups at the same time. The statistical hit rate for 24 numbers is 3.43. I only played systems that exceeded this average. My budget limit was $30 per play which forced me to allocate only 6 lines per group. These tiny wheels did generate frequent 3 or 2+ winners nevertheless (and if I didn't see it multiple times I would not believe it) - but not beyond that. I just did not have sufficient number of numbers.
The idea behind this new methodology is twofold:
1) To maximize chances for winning any prize - by using all the numbers in each wheel.
2) To win at the least possible cost, whichever time frame.
Yes, distributing 49 numbers into just 15 tickets is definitely an obstacle. But if I could win (and quite frequently for that matter) with 24 numbers in just 6 lines wheel there is little doubt some results can be achieved for the 49/15 configuration as well.
Each wheel has its custom number distribution. All wheels have the same winning potential. But it is the distribution that defines how much a wheel wins in a period of time. I don't believe that here is one "golden" distribution formula that would cover majority of all possible distribution configurations. So I opted for a different option - the brute force of numbers.
One statistical run showed that in 100 draws a given sum is seldom represented more that 3 - 4 times. That's a very low percentage and cannot be reliably used for single best sum prediction. That's why there are 31 most popular sum groups each with 20 custom wheels, for the total of 620. I want to start with this to see if it is sufficient or needs enlargement.
And then come backtracks, the backbone of the methodology. Each wheel is tested/retested in 3 different time frames: 100, 50 & 25 draws. The currently used criterion for selection is set at 50% minimum ROI in all time frames. 25 & 50 draws have no problems in returning 2 - 3 dozen of possible candidates. With 100 there can be a struggle, as only very few will qualify. The law of averages catching up.
What's the purpose of such qualification process? 50% in all 3 time frames tells that the wheel has stable performance in long run and no major fluctuations. Stability is what I'm looking for (at least at this stage but this may change depending on future results).
-
Quote: Originally posted by ethanl on May 20, 2025
So this is what you are saying:
You’re shifting away from traditional number-picking methods and instead using a statistical and combinatorial approach to playing the lottery. Here’s what that means in practice:
-
Instead of picking numbers based on patterns, birthdays, or "hot/cold" numbers, you focus on the math behind how groups of numbers perform on average.
-
You realize that playing only a small set of numbers (like 15 or 24) limits your chances of hitting the jackpot, because the expected number of matches is low based on probability.
-
To improve your chances, you analyze all possible combinations and use filters—like the sum of the numbers or the spread of digits—to dramatically reduce the number of combinations you play, making it more practical and cost-effective.
-
Your approach is less about luck or superstition and more about maximizing coverage and efficiency using mathematical reasoning.
Examples to illustrate your approach:
-
If you play 24 numbers out of 49, math shows you’ll average about 3 or 4 matches per draw—not enough for a jackpot.
-
By filtering combinations to only those with a specific sum (for example, all combinations that add up to 150), you reduce millions of possibilities down to a much smaller group, making it manageable.
-
Adding another filter, such as requiring each combination to have a wide spread of digits (not just numbers in the 20s or 30s), cuts down the number of combinations even further.
-
Instead of picking numbers based on birthdays (which most people do, leading to more shared jackpots), you use statistical filters to select combinations that are less likely to be picked by others and more likely to cover the range of winning numbers
-
You avoid common mistakes like clustering numbers, only picking low numbers, or relying on personal "lucky" numbers, focusing instead on balanced and well-distributed combinations
your strategy is about playing smarter, not harder, by using probability and combinatorial math rather than intuition or tradition.
Yes, you got it all right, credit to you.
Although backtracks look already encouraging the whole methodology needs to be tested if it pans out, as hoped for, in reality.
-
-
Quote: Originally posted by Player649 on May 21, 2025
Yes, you got it all right, credit to you.
Although backtracks look already encouraging the whole methodology needs to be tested if it pans out, as hoped for, in reality.
Give it up, give your fingers a rest. There is no positive outcome here, it's all luck. You're just spinning your wheels.
It's always possible that I'm wrong~
But that's never at the top of my list of possibilities.
-
Anything is possible.
-
700 light yrs West of Milky Way Galaxy's Center
United States
Member #200,634
September 1, 2019
4,903 Posts
OfflineQuote: Originally posted by Player649 on May 21, 2025
Playing all the numbers does improve chances for winning, particularly bigger prizes.
I used to play 24 number groups. My biggest problem was that the groups of 24 were missing numbers well too often for my comfort. In order to spread the risk I even played 5 different groups at the same time. The statistical hit rate for 24 numbers is 3.43. I only played systems that exceeded this average. My budget limit was $30 per play which forced me to allocate only 6 lines per group. These tiny wheels did generate frequent 3 or 2+ winners nevertheless (and if I didn't see it multiple times I would not believe it) - but not beyond that. I just did not have sufficient number of numbers.
The idea behind this new methodology is twofold:
1) To maximize chances for winning any prize - by using all the numbers in each wheel.
2) To win at the least possible cost, whichever time frame.
Yes, distributing 49 numbers into just 15 tickets is definitely an obstacle. But if I could win (and quite frequently for that matter) with 24 numbers in just 6 lines wheel there is little doubt some results can be achieved for the 49/15 configuration as well.
Each wheel has its custom number distribution. All wheels have the same winning potential. But it is the distribution that defines how much a wheel wins in a period of time. I don't believe that here is one "golden" distribution formula that would cover majority of all possible distribution configurations. So I opted for a different option - the brute force of numbers.
One statistical run showed that in 100 draws a given sum is seldom represented more that 3 - 4 times. That's a very low percentage and cannot be reliably used for single best sum prediction. That's why there are 31 most popular sum groups each with 20 custom wheels, for the total of 620. I want to start with this to see if it is sufficient or needs enlargement.
And then come backtracks, the backbone of the methodology. Each wheel is tested/retested in 3 different time frames: 100, 50 & 25 draws. The currently used criterion for selection is set at 50% minimum ROI in all time frames. 25 & 50 draws have no problems in returning 2 - 3 dozen of possible candidates. With 100 there can be a struggle, as only very few will qualify. The law of averages catching up.
What's the purpose of such qualification process? 50% in all 3 time frames tells that the wheel has stable performance in long run and no major fluctuations. Stability is what I'm looking for (at least at this stage but this may change depending on future results).
"I used to play 24 number groups. My biggest problem was that the groups of 24 were missing numbers well too often for my comfort."
Playa649,..Ya sholda been uzin
Probability Math insteada all dem Non- Viable DUDZZ 
Just don't undastand whyyy you refuze to meazure up Stat$tistically 
..Hurry, make it back to shore & stand on Solid Groundz..
YO
"New play methodology" will neevah defeat the 1st concrete Law of ALL Methodz..Garbage in = Garbage out
-Stat$talker
-
"1st Concrete Law of All Methods". Let's analyze this logically:
1. What would be the "1st Concrete Law of All Methods"?
In probability and statistics applied to lotteries, the fundamental law is:
✅ "Numbers in lotteries are drawn randomly and independently, and no method can predict or influence the future outcome."This means:
There are no patterns that can be exploited to guarantee victories.
Statistics (such as averages, frequencies, lags) are only a posteriori analyzes and do not alter the probability of future draws.
2. Why doesn't the "Average of Numbers" defeat this law?
If your methodology is based on calculating the average of the numbers drawn (eg: adding the numbers and dividing by the amount), this does not alter the randomness.Example: If in a 60-number lottery (like the Mega-Sena), the historical average of the draws was ~30, that does not mean that the next numbers will tend to be close to that average.
Each combination has the same probability, regardless of past statistics.
➡️ The law of randomness always prevails: No calculation (averages, frequencies, sequences) changes the fact that all numbers have equal chances in each draw.
3. Why do people believe in methods like this?
Cognitive bias: The human brain looks for patterns even where there are none (such as understanding that "lagged numbers" or "averages" influence results).Gambler Fallacy: Believing that past events affect future events in random systems (e.g.: "The number 5 hasn't come up in 10 draws, so it's 'late'").
Conclusion
Your observation is correct: No methodology based on statistics (such as averages) can "mock" the randomness of lotteries. The "1st Law" is inviolable because the draws are independent and unpredictable.If you want to optimize games, focus on:
Bankroll management (don’t spend more than you can afford to lose).
Reduce donkey bets (avoid obvious combinations that can split prizes).
