- 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 7:46 pm
You last visited
April 17, 2024, 7:42 pm
All times shown are
Eastern Time (GMT-5:00)
Backtesting and Simulating Lottery SystemsPrev TopicNext Topic
-
Quote: Originally posted by RJOh on Apr 8, 2011
I was planning to back-test and/or simulate ONE component of the "Digit System," namely the first filter applied which rejects all sets from the (5,39) matrix that have less than 5 or more than 6 unique digits.
No need to run a simulation to get those figures, it's part of my regular routine to pick parameters for my predictions. This is part of the results for Ohio Rolling Cash5(5/39) with 2236 drawings so far. I use 95% for my parameters. I can create a text file with all the information but I can only load 800 lines to sort and analyze. I would have to rewrite the program to use virtual memory since my programs are in GWBasic, which I may do sometime in the future or learn a new Basic and rewrite all my programs so they will work in Window7. Years ago when I wrote this program as a utility, 800 lines were plenty and still is for the other games I play. They have more than 800 total records but their latest matrix have less than 800 drawings.
Different Digits
0: 0 10: 0
1: 0 11: 0
2: 0 12: 0
3: 14 13: 0
4: 222 14: 0
5: 763 15: 0
6: 892 16: 0
7: 319 17: 0
8: 26 18: 0
9: 0 19: 095% DD = 4-7
Total Digits
0: 0 10: 552
1: 0 11: 0
2: 0 12: 0
3: 0 13: 0
4: 0 14: 0
5: 0 15: 0
6: 22 16: 0
7: 140 17: 0
8: 557 18: 0
9: 965 19: 095% TD = 7-10
Very nice post RJOH. Similiar to Texas 5/37 in 2720 draws:
2 digits = 1
3 digits = 54
4 digits = 501
5 digits = 1177
6 digit = 806
7 digits = 175
8 digits = 6
total from 4 to 7 = 2659/2720 = 97.757%
Overall the breakdown for a 5/37 game are:
2 = 18
3 = 3044
4 = 44970
5 = 157632
6 = 169287
7 = 56428
8 = 4518
This is where brain freeze occurs. I've said this to RL privately. This accounts for the why and how people dismiss Digit Systems in general. One has to quit looking at numbers and force themselves to see digits.
With RL's assistance and advice, I have proved this to myself. I know it works. Since March 5th, it has produced 3 jackpot prizes in Texas. It is an exciting idea with a lot of promise if one takes the time to aquaint themselves with the basics and let the digits do the work.
It's like train meets baquette. Unless the conductor is hungry, that train won't stop rollin', it won't stop rollin' nohow!
-
Quote: Originally posted by garyo1954 on Apr 8, 2011
Very nice post RJOH. Similiar to Texas 5/37 in 2720 draws:
2 digits = 1
3 digits = 54
4 digits = 501
5 digits = 1177
6 digit = 806
7 digits = 175
8 digits = 6
total from 4 to 7 = 2659/2720 = 97.757%
Overall the breakdown for a 5/37 game are:
2 = 18
3 = 3044
4 = 44970
5 = 157632
6 = 169287
7 = 56428
8 = 4518
This is where brain freeze occurs. I've said this to RL privately. This accounts for the why and how people dismiss Digit Systems in general. One has to quit looking at numbers and force themselves to see digits.
With RL's assistance and advice, I have proved this to myself. I know it works. Since March 5th, it has produced 3 jackpot prizes in Texas. It is an exciting idea with a lot of promise if one takes the time to aquaint themselves with the basics and let the digits do the work.
It's like train meets baquette. Unless the conductor is hungry, that train won't stop rollin', it won't stop rollin' nohow!
This is where brain freeze occurs. I've said this to RL privately. This accounts for the why and how people dismiss Digit Systems in general. One has to quit looking at numbers and force themselves to see digits.
The problem I have with that is when you limit the number of digits to 5-6 you only cover 75% of the combinations which is fine if that's the only parameter you use but if you are limiting other parameters like sums, ranges, gaps between numbers and etc. the same amounts then you eliminate 50% or more of the potential winners. With the number of parameters I use, 95% works better for me because 5-10% of the combinations picked randomly are flagged for my final check and acceptances.
* you don't need to buy every combination, just the winning ones *
-
Where is a good introduction to the whole "digit" theory?
How you do anything is how you do everything.
-
Quote: Originally posted by RJOh on Apr 8, 2011
I was planning to back-test and/or simulate ONE component of the "Digit System," namely the first filter applied which rejects all sets from the (5,39) matrix that have less than 5 or more than 6 unique digits.
No need to run a simulation to get those figures, it's part of my regular routine to pick parameters for my predictions. This is part of the results for Ohio Rolling Cash5(5/39) with 2236 drawings so far. I use 95% for my parameters. I can create a text file with all the information but I can only load 800 lines to sort and analyze. I would have to rewrite the program to use virtual memory since my programs are in GWBasic, which I may do sometime in the future or learn a new Basic and rewrite all my programs so they will work in Window7. Years ago when I wrote this program as a utility, 800 lines were plenty and still is for the other games I play. They have more than 800 total records but their latest matrix have less than 800 drawings.
Different Digits
0: 0 10: 0
1: 0 11: 0
2: 0 12: 0
3: 14 13: 0
4: 222 14: 0
5: 763 15: 0
6: 892 16: 0
7: 319 17: 0
8: 26 18: 0
9: 0 19: 095% DD = 4-7
Total Digits
0: 0 10: 552
1: 0 11: 0
2: 0 12: 0
3: 0 13: 0
4: 0 14: 0
5: 0 15: 0
6: 22 16: 0
7: 140 17: 0
8: 557 18: 0
9: 965 19: 095% TD = 7-10
I'm sorry RJ, but when you say things like, "No need to run a simulation to get those figures, it's part of my regular routine to pick parameters for my predictions," it tells me we are not on the same page. You're talking in specifics about how you pick parameters for your predictions with a limited history and I'm talking about Monte Carlo Methods for back-testing or simulating systems over arbitrarily long periods of time.
It doesn't appear that you've spent much time with the references in my opening post. They're there for the groundwork required before moving on to talking about system specifics and modeling. If "there" is where I want to lead you, and "here" is where you are, I'm going to fall back on the old joke that people play on travelers asking them for directions - "You can't get there from here; you have to start from somewhere else."
-
"You can't get there from here; you have to start from somewhere else."
I always try to get there from where I am rather than from somewhere else, it's easier.
* you don't need to buy every combination, just the winning ones *
-
Quote: Originally posted by jimmy4164 on Apr 8, 2011
"There are those who come up with theories and systems and check them out and admit that they don't work, and then there are those who come up with theories and systems and if the results prove them wrong they ignore the results." Coin Toss
I think you're absolutely right Coin Toss. For these reasons I'm questioning whether it makes any sense for me to devote any more time to this. I was planning to back-test and/or simulate ONE component of the "Digit System," namely the first filter applied which rejects all sets from the (5,39) matrix that have less than 5 or more than 6 unique digits. The results showing that selecting from this reduced set of possibilities provides no statistically significant advantage to the player would probably be rejected for at least 3 reasons: one, the general distrust of the use of RNGs in the analysis, crucial to Monte Carlo Methods; two, the leap of faith required (for many) to see that the result was useful in isolation from the other filters applied in practise; and, of course, the missing "human element."
Another way to state your observation is to say, "Anyone who would accept the results of a Monte Carlo simulation didn't need to see it in the first place, and all those who need to see the results, would not accept them!"
What do you think of that dude putting up $30K to challenge betting systems?
JIMBO
Ive been waiting on your backtest since you started you fooled by random post. You now want to do a backtest
that involves using 5 or 6 digits. Wow, I could have posted these results before the first draw. When you come up
with a method that can match my ability making selections then test away. Lets say that I did play everyday and
I played 6 digits. 40% of the draws fall into this range so it would cost me $229,473 per day to play all the combos
or $83,757,645 dollars per year. Hmmmm. and 365 * .40 = 146 so 146 * $140,000 = 20,732,000 and $83,757,645
minus 20,732,000 equals $63,025,645 loss over the year just using six digits. Wow, gee wizz, amazing, wohoo.
Anyone who thinks in these terms has totaly missed it. My claims are based on my ability to predict and then play
only the days I can predict several of these. I can put as many or as few digits into play as I wish to use. You must
have read over the wild card post. Many days I play 8-9-10 digits but place limits on them I can set a digit to hit once
or I can set it to hit 5 or 6 times per set. I gues you misses watching the videos I posted showing the effectiveness
of placing limits on the digits in play. Just because I may select 4 digits and set 2 or 3 more giving them the option
to hit or miss which is beyound your ability to conceive my system. How stupid do you think I am. I hope it is not
nearly as stupid as I think your post are suggesting a backtest to prove something that has occured in the past.
Run all the test you want they mean nothing. I doubt all your claims as to what you have done in the past but I don't
hark on them as if it's the only way to feel important. I think it is high time to prove a few of your claims as to your
past experences, as your Jr high math does not attest to any of these in my opinion. You are either the side-step
queen of the universe or you just don't understand what is written. Selecting and using digits is not some magic
win every time you play system and I play on the days that I feel the data I have shows strong for the digits I play.
Once I make a setup I run the program without any filters and some days I end up with 2000+ lines and other days I
end up with as few as 20. I only play the days when both of these come togeather. My selections change every
day based on how I interpet the data. You missguided assumptions are useless and make you look foolish when
you claim something you don't understand. I told the pen and paper players who play maybe $5.00 in tickets to use
digit counts and totals that would put there play in the deepest end of the pool. Every set has the exact same
odds. Every number has the exact same odds. Digits don't, simple as that. Digits are way more likely to hit
within a much smaller window of time then the munbers, analyzing digits gives better results in my opinion then
any other method I have ever used hands down. Does this mean that using digits is a sure way to win, NO NO
NO NO. Do you understand what NO means. I know that my digit selections end in many more lower prizes over
using numbers. When you analyse numbers you can compare them to other numbers or how many days out or
any methods that you like but I analyse digits without regard for the numbers formed by these same digits.
Playing a lottery is a game and only a game. My attemps at playing have paid off for me without counting the
jackpot win. I hit a 5 of 5 sometimes a couple times a month. Do you see me out buying these tickets, no I
have a budget of around $500.00 a year for the lottery. Had I thought or known these days would have hit then
you bet I would have got the tickets. Sometimes I spend less then $500 per year and sometimes more based on my
winnings. You claim 11 x is unthinkable but I say that it runs closer to 17 to 23 and this is really not that hard to do.
Sorry you can't maybe you should try digits. I tell everyone to play on paper first because if you can't hit on paper
then you can't hit playing. Some do very well and some do not but I expected this from the start. The digit system
was posted so that those that arn't doing to well picking numbers would have something else to try. If it works for
them then use it if not then don't. I still say that if you want to tell people not to play then I would agree but for
those that want to take a chance and do this by picking there own numbers then try using digits. I got several PM's
from people who said they won some good amounts but did I ask them to prove it, NO. Self picks is all about trying
to predict which will hit before the draw and some of us feel better about or plays vs a quickpick. If you want to
Help people out then start a post about what the odds are in terms people can think about. Like how long would a
line of pennies be if you placed one for each powerball ticket side to side, over 2311 miles long. If you had a ticket
for each penny from just one mile it would you would have 47,520 pennies and the weight of them would be over 261
pounds. Most people can't lift 261 pounds and can't begin to afford to buy this many tickets which amounts to only
00024338146089045839884069297462512% of the total sets. I can say this for you, you sure know how to take the
safe side of things. I do have one question for you. How many wins counting all prizes would a person need to win
before it would change the overall stats enough to concern lottery officals.
RL
-
Quote: Originally posted by RL-RANDOMLOGIC on Apr 8, 2011
JIMBO
Ive been waiting on your backtest since you started you fooled by random post. You now want to do a backtest
that involves using 5 or 6 digits. Wow, I could have posted these results before the first draw. When you come up
with a method that can match my ability making selections then test away. Lets say that I did play everyday and
I played 6 digits. 40% of the draws fall into this range so it would cost me $229,473 per day to play all the combos
or $83,757,645 dollars per year. Hmmmm. and 365 * .40 = 146 so 146 * $140,000 = 20,732,000 and $83,757,645
minus 20,732,000 equals $63,025,645 loss over the year just using six digits. Wow, gee wizz, amazing, wohoo.
Anyone who thinks in these terms has totaly missed it. My claims are based on my ability to predict and then play
only the days I can predict several of these. I can put as many or as few digits into play as I wish to use. You must
have read over the wild card post. Many days I play 8-9-10 digits but place limits on them I can set a digit to hit once
or I can set it to hit 5 or 6 times per set. I gues you misses watching the videos I posted showing the effectiveness
of placing limits on the digits in play. Just because I may select 4 digits and set 2 or 3 more giving them the option
to hit or miss which is beyound your ability to conceive my system. How stupid do you think I am. I hope it is not
nearly as stupid as I think your post are suggesting a backtest to prove something that has occured in the past.
Run all the test you want they mean nothing. I doubt all your claims as to what you have done in the past but I don't
hark on them as if it's the only way to feel important. I think it is high time to prove a few of your claims as to your
past experences, as your Jr high math does not attest to any of these in my opinion. You are either the side-step
queen of the universe or you just don't understand what is written. Selecting and using digits is not some magic
win every time you play system and I play on the days that I feel the data I have shows strong for the digits I play.
Once I make a setup I run the program without any filters and some days I end up with 2000+ lines and other days I
end up with as few as 20. I only play the days when both of these come togeather. My selections change every
day based on how I interpet the data. You missguided assumptions are useless and make you look foolish when
you claim something you don't understand. I told the pen and paper players who play maybe $5.00 in tickets to use
digit counts and totals that would put there play in the deepest end of the pool. Every set has the exact same
odds. Every number has the exact same odds. Digits don't, simple as that. Digits are way more likely to hit
within a much smaller window of time then the munbers, analyzing digits gives better results in my opinion then
any other method I have ever used hands down. Does this mean that using digits is a sure way to win, NO NO
NO NO. Do you understand what NO means. I know that my digit selections end in many more lower prizes over
using numbers. When you analyse numbers you can compare them to other numbers or how many days out or
any methods that you like but I analyse digits without regard for the numbers formed by these same digits.
Playing a lottery is a game and only a game. My attemps at playing have paid off for me without counting the
jackpot win. I hit a 5 of 5 sometimes a couple times a month. Do you see me out buying these tickets, no I
have a budget of around $500.00 a year for the lottery. Had I thought or known these days would have hit then
you bet I would have got the tickets. Sometimes I spend less then $500 per year and sometimes more based on my
winnings. You claim 11 x is unthinkable but I say that it runs closer to 17 to 23 and this is really not that hard to do.
Sorry you can't maybe you should try digits. I tell everyone to play on paper first because if you can't hit on paper
then you can't hit playing. Some do very well and some do not but I expected this from the start. The digit system
was posted so that those that arn't doing to well picking numbers would have something else to try. If it works for
them then use it if not then don't. I still say that if you want to tell people not to play then I would agree but for
those that want to take a chance and do this by picking there own numbers then try using digits. I got several PM's
from people who said they won some good amounts but did I ask them to prove it, NO. Self picks is all about trying
to predict which will hit before the draw and some of us feel better about or plays vs a quickpick. If you want to
Help people out then start a post about what the odds are in terms people can think about. Like how long would a
line of pennies be if you placed one for each powerball ticket side to side, over 2311 miles long. If you had a ticket
for each penny from just one mile it would you would have 47,520 pennies and the weight of them would be over 261
pounds. Most people can't lift 261 pounds and can't begin to afford to buy this many tickets which amounts to only
00024338146089045839884069297462512% of the total sets. I can say this for you, you sure know how to take the
safe side of things. I do have one question for you. How many wins counting all prizes would a person need to win
before it would change the overall stats enough to concern lottery officals.
RL
RL-RANDOMLOGIC says, "Ive been waiting on your backtest since you started you fooled by random post."
Are you referring to these back-tests...
https://www.lotterypost.com/thread/220106
...or these:
https://www.lotterypost.com/thread/218174/1736549
When you say, "When you come up with a method that can match my ability making selections then test away," it's clear to me you have spent next to no time reviewing the references in my opening post to this Thread.
And really, do you still believe I would be interested in devoting any time to trying to design a selection system to beat chance in the lottery? Where have you been? What are you drinking tonight?
--Jimmy4164
-
Quote: Originally posted by mayhem on Apr 8, 2011
Where is a good introduction to the whole "digit" theory?
Here you go mayhem. It's a great thread, I think 46 pages. Maybe 20 if you can ignore Jimmy4164's posts.
-
Jimbo
I decided to take a shot at my own question.
Show me 5 FY2010
total sales = 26,700,000
based on 63.4% of ticket sales go to prizes
payouts = $16,927,800
* .1 = $1,692,780.00
* .01 = $169,278.00
*.001 = $16,927.80
*.0001 = $1,692.78
*.00001 = $169.28
RL
-
Quote: Originally posted by ameriken on Apr 8, 2011
Here you go mayhem. It's a great thread, I think 46 pages. Maybe 20 if you can ignore Jimmy4164's posts.
Awesome.
P.S. I enjoy Jimmy's posts. Where else would I get such awesome signatures?
How you do anything is how you do everything.
-
Quote: Originally posted by mayhem on Apr 9, 2011
Awesome.
P.S. I enjoy Jimmy's posts. Where else would I get such awesome signatures?
LOL, I think you'll find an abundance of Jimmy sigs there! Jimmy should be honored.
But seriously, it's a great thread, read the whole thing because as the thread progresses RL-Randomlogic gives a little more and little bit more and a little bit more info clarifying the system, plus there are many other great contributors there as well.
-
ameriken
Thanks, I think or at least I hope the readers understood that I was sharing what I felt was a better
way to play. When I first started building a system I used numbers and the data I could get from them.
I won a little but just a little and was not able to track my progress. I gave up many times and then
I would get another idea and start again. The one thing that I felt that stopped me from moving ahead
was that no matter what the data said I would somehow make it line up with what I wanted it to say.
This led me to play over and over almost the same thing every day. When I first started using digits
I was still unable to move away from this and I found myself picking digits that matched the numbers
I wanted to hit. My solution was to remove all the tools that used numbers so I had no option to view
the numbers. I also used binary strings of zeros and ones to repersent the digits. This let me focus
on the data without regard to what it was for. It was at this time I started to gain a little ground. I
can't put my finger on any one thing that makes me choose one over another it just sort of makes
sense to me. Some times it just a garbled mess but when something seems to stand out I play it.
Learning to wait until enough data is found to make a run is hard but over the years I have gotten a
little more patient. I have tried to write out a set of rules but each bit of data has an effect on the rest.
I make my choices very quickly and make a run, It is when I have to use many of the filters to reduce
sets that I do bad so on these days I just don't play. Several of the guys have hit 5 of 5's but chose
not to play which is something I have done many times. I tell myself that If I can hit it once I can do it
again. I don't have the funds to play more but often think of trying a hundred bucks for a week and then
the reality sets in and I know I could go a week without a good hit. In the past everytime I try and play
more I seem to do it on the days I make the most mistakes so I just stick to the plan and wait it out.
It's easy to get the big head when you win some good amounts so I stop playing for a few days after
a good hit. I still play on paper and it has shown that I made a good choice by stoping when I did.
Anyway glad you liked the read.
RL
-
Quote: Originally posted by RL-RANDOMLOGIC on Apr 9, 2011
Jimbo
I decided to take a shot at my own question.
Show me 5 FY2010
total sales = 26,700,000
based on 63.4% of ticket sales go to prizes
payouts = $16,927,800
* .1 = $1,692,780.00
* .01 = $169,278.00
*.001 = $16,927.80
*.0001 = $1,692.78
*.00001 = $169.28
RL
You're forgetting, you gave your software away to 40 disciples.
Aren't they up to speed yet?
BTW, based on an earlier remark of yours, you seem to be saying that after throwing out all the sets that don't have 5 or 6 unique digits, without further filtering, you would be forced to bet on all the thousands remaining.
So, are you then telling us that if one were to run a simulation that randomly generated (5,39) sets, rejecting all but 10 containing 5 or 6 unique digits, and playing them against the daily draw, that the result would NOT be any better than simply betting on the first 10 random sets produced? (QPs)
If your answer is NO, then I must assume this filter is no value.
If your answer is YES, I must ask why you would not support a simple backtest of this one component of your method.
-
Quote: Originally posted by RL-RANDOMLOGIC on Apr 9, 2011
Jimbo
I decided to take a shot at my own question.
Show me 5 FY2010
total sales = 26,700,000
based on 63.4% of ticket sales go to prizes
payouts = $16,927,800
* .1 = $1,692,780.00
* .01 = $169,278.00
*.001 = $16,927.80
*.0001 = $1,692.78
*.00001 = $169.28
RL
CORRECTION!
I reversed my YES and NO below. Sorry.
----------Corrected Post----------------
You're forgetting, you gave your software away to 40 disciples.
Aren't they up to speed yet?
BTW, based on an earlier remark of yours, you seem to be saying that after throwing out all the sets that don't have 5 or 6 unique digits, without further filtering, you would be forced to bet on all the thousands remaining.
So, are you then telling us that if one were to run a simulation that randomly generated (5,39) sets, rejecting all but 10 containing 5 or 6 unique digits, and playing them against the daily draw, that the result would NOT be any better than simply betting on the first 10 random sets produced? (QPs)
If your answer is YES, then I must assume this filter is of no value.
If your answer is NO, I must ask why you would not support a simple backtest of this one component of your method.
-
Quote: Originally posted by jimmy4164 on Apr 9, 2011
You're forgetting, you gave your software away to 40 disciples.
Aren't they up to speed yet?
BTW, based on an earlier remark of yours, you seem to be saying that after throwing out all the sets that don't have 5 or 6 unique digits, without further filtering, you would be forced to bet on all the thousands remaining.
So, are you then telling us that if one were to run a simulation that randomly generated (5,39) sets, rejecting all but 10 containing 5 or 6 unique digits, and playing them against the daily draw, that the result would NOT be any better than simply betting on the first 10 random sets produced? (QPs)
If your answer is NO, then I must assume this filter is no value.
If your answer is YES, I must ask why you would not support a simple backtest of this one component of your method.
Jimmy
You really do have a reading impairment don't you. Have you downloaded the video and watched it? I look
for data that I see as showing a good chance for a hit. When I find conflicting data I can allow one or both
to hit or miss. Lets say that I have 5 strong digits but see 2 others that could show in the next draw. I can
set them so that only one of them can hit within a set or both. I can set them to hit a maximun of once, twice
or any amount for that matter. I don't say that 5 digits will hit at the start. I choose the digits one at a time
and If I end up with 5 then I end up with 5. I may add a couple wild cards so to speak but I place limits on them
If I end up with seven then I will play 7 but I have another analysis that I can do to see what the odds for 7 digit
set is. If I see that 7 should not hit then I might take the two digits with the lowest chance of hitting and set them
to hit or miss but if they hit then they can only show once per set. This then gives me a range of 5 to 7 or 5 to 6
digit in play. I can control each digits count for every set that is produced. Most all my filters are digit related and
I calculate them thinking of the digits that are in play. If I select the correct digits then the software will produce
the exact winning set every time it is ran. Unless you can determind how I select the digits then you cannot do a
backtest unless you want to just make a guess. Since the software is mechanical in what it does and uses a numerical
set generator that goes through every set in the matrix each run then there is nothing to test unless you are looking
for errors in the code. How and what I select is based on so many bits of data that are calculated differently for each
draw and I may use different data for the same selection from draw to draw. Unless you are willing to spend the rest
of your life trying to code how I make decisions then your wasting time asking it. If it could be coded then I would have
done it. If you think you could do this with a simple program like you used in the P-3 analysis you nuts. I could code a
program to make those simple calculations in less then a hour and probably in less then 10 minuets. One of the guys
hit a 4 of 5 a week or so back. I don't keep in touch with all of them but they do contact me from time to time and let me
know how they are doing or maybe bragg a little. I take them at their word because they have nothing to gain. Many
use the program along with other software. DM has the ability to import sets from another program which can then be
filtered using digits or the other filters.
RL