Welcome Guest
You last visited December 8, 2016, 6:57 am
All times shown are
Eastern Time (GMT-5:00)

# Backtesting and Simulating Lottery Systems

Topic closed. 136 replies. Last post 6 years ago by RL-RANDOMLOGIC.

 Page 6 of 10

United States
Member #59354
March 13, 2008
3972 Posts
Offline
 Posted: April 12, 2011, 12:45 am - IP Logged

Jimmy

You've got to be kidding!

Are you saying that picking so many digits to play needs some advanced math to calculate what the odds

for matching 3,4,5 or more of them.  Looks like another side step to me.

RL

mid-Ohio
United States
Member #9
March 24, 2001
19828 Posts
Online
 Posted: April 13, 2011, 12:11 am - IP Logged

Don't you guys think it's time for you to put aside your personal animosities
and get back to the question at hand?

Personally, I'd appreciate some constructive comments regarding the role of
'substitution' as it applies to lottery strategies.

If the odds of this or that happening for everything is the general rule,
then we should just mail a check to our state lottery and go watch TV.

If, on the other hand, there is good reason to believe that, despite the odds, there are
ways to beat the lottery, then we should pursue them, as best we can.

I've been alive for a long, long time. I'm fully aware that people are always
doing what they believe is in their best self-interest.

Expecting strangers at a lottery forum to come together and do something
constructive for the benefit of all players is probably a big stretch.

However, I could be wrong. Maybe there are a few folks among the +100,000 members
here who would join in and do something that benefits more than themselves.

I've been using 'substitution' to guide my lottery plays for years.
While I've had some some wins, I think my techniques could be improved.
Not seeing the forest because of the trees is a factor in many situations.

Pick 3 is a pari-mutual game.

It sure would be worthwhile if several players would join together and develop
a strategy that can generate more wins than losses.

What do you think??

Expecting strangers at a lottery forum to come together and do something constructive for the benefit of all players is probably a big stretch.

However, I could be wrong. Maybe there are a few folks among the +100,000 members here who would join in and do something that benefits more than themselves.

Lottery systems aren't designed by groups or teams, but by individualist.  I doubt is anyone who has spent time designing a system for himself would have any interests in joining a team to work on someone else ideas.  About the best you are going to get is them sharing and discussing their ideas.  What others do with those ideas is up to them.  Too many people are looking for absolute answers about what they can expect using a system.  There are no more assurances when using a system than there are when you buy some QPs.  You simply pick the best strategy you can come up with and wait for the results.  If those results are better than you expected, then you might have yourself a winning system.

* you don't need to buy more tickets, just buy a winning ticket *

United States
Member #59354
March 13, 2008
3972 Posts
Offline
 Posted: April 13, 2011, 12:57 am - IP Logged

RJOh

I think you are correct, The best we can hope for is for someone to just post a filter or other idea that

may help or give someone another idea to test or add to what they are using, kind of a tips and tricks

post.

RL

United States
Member #105312
January 29, 2011
435 Posts
Offline
 Posted: April 13, 2011, 9:25 am - IP Logged

Expecting strangers at a lottery forum to come together and do something constructive for the benefit of all players is probably a big stretch.

However, I could be wrong. Maybe there are a few folks among the +100,000 members here who would join in and do something that benefits more than themselves.

Lottery systems aren't designed by groups or teams, but by individualist.  I doubt is anyone who has spent time designing a system for himself would have any interests in joining a team to work on someone else ideas.  About the best you are going to get is them sharing and discussing their ideas.  What others do with those ideas is up to them.  Too many people are looking for absolute answers about what they can expect using a system.  There are no more assurances when using a system than there are when you buy some QPs.  You simply pick the best strategy you can come up with and wait for the results.  If those results are better than you expected, then you might have yourself a winning system.

I've wondered about that a lot reading the posts on the systems and this forum.  A lot of people on the systems forum and some of the others are putting a lot of effort into trying to build systems capable of getting an edge on lottery operations.  But a lot of those attempts rely on nuances easily corrected by the lottery administrations if the systems worked and the lottery administrations became aware of what was being done to beat them and how it was being done.

A person would be almost forced to conclude they don't believe what they are working on will ever work, or the satisfaction of speculating that it will on public forums and providing all the details for systems so fixes by the lottery administrations can be rapidly employed is greater than the hope of actually reaping benefits.

Kentucky
United States
Member #32652
February 14, 2006
7313 Posts
Offline
 Posted: April 13, 2011, 11:39 am - IP Logged

I've wondered about that a lot reading the posts on the systems and this forum.  A lot of people on the systems forum and some of the others are putting a lot of effort into trying to build systems capable of getting an edge on lottery operations.  But a lot of those attempts rely on nuances easily corrected by the lottery administrations if the systems worked and the lottery administrations became aware of what was being done to beat them and how it was being done.

A person would be almost forced to conclude they don't believe what they are working on will ever work, or the satisfaction of speculating that it will on public forums and providing all the details for systems so fixes by the lottery administrations can be rapidly employed is greater than the hope of actually reaping benefits.

"But a lot of those attempts rely on nuances easily corrected by the lottery administrations if the systems worked and the lottery administrations became aware of what was being done to beat them and how it was being done."

What could any state lottery do to prevent one player from beating them that has no affect on the rest of the players?

mid-Ohio
United States
Member #9
March 24, 2001
19828 Posts
Online
 Posted: April 13, 2011, 12:39 pm - IP Logged

I've wondered about that a lot reading the posts on the systems and this forum.  A lot of people on the systems forum and some of the others are putting a lot of effort into trying to build systems capable of getting an edge on lottery operations.  But a lot of those attempts rely on nuances easily corrected by the lottery administrations if the systems worked and the lottery administrations became aware of what was being done to beat them and how it was being done.

A person would be almost forced to conclude they don't believe what they are working on will ever work, or the satisfaction of speculating that it will on public forums and providing all the details for systems so fixes by the lottery administrations can be rapidly employed is greater than the hope of actually reaping benefits.

...a lot of those attempts rely on nuances easily corrected by the lottery administrations if the systems worked...

Many of those nuances are just the natures of the game and there's nothing lottery administration can or would do to change them and still have an honest game.

For example in Ohio 649 game, 50% of the winning combinations have 2 or more adjacent numbers while only one third of PowerBall and MegaMillions have them, so avoiding those types of combinations might give a PB or MM players an edge.

When Tennessee switched their pick3 game to a computer RNG, the game ran almost a month before the administration noticed no doubles were coming up.  It was the complaints of players that finally alerted them to the fact that their RNG wasn't set right.  During that time,those betting box had a 50% edge since no doubles or triples could be picked.

* you don't need to buy more tickets, just buy a winning ticket *

United States
Member #59354
March 13, 2008
3972 Posts
Offline
 Posted: April 13, 2011, 5:01 pm - IP Logged

RJOh

I have a quick question, when you pick your numbers does your software generate and filter at the same time

or do you generate the sets first and then apply the filters after the fact.  I added a RNG option to what I am

using and it filters as it goes.  If I use more then just a few filters then the run can take a very long time  to

pick even 10 lines.  #2. Do you set ranges for each number/position in the generating stage. I included the

code for the RNG that I am using

QB64 CODE

'_____________________________

Pickt = 5 'numbers per set
Nset = 39 'numbers in matrix

'Pick 5 or 6 random numbers
RANDOMPICK:
RANDOMNUM\$ = ""
DO WHILE LEN(RANDOMNUM\$) < (PICKT * 3)
GENRAN = RAND(1, NSET)
IF GENRAN < 10 THEN GENRAN\$ = "0" + LTRIM\$(STR\$(GENRAN)) ELSE GENRAN\$ = LTRIM\$(STR\$(GENRAN))
IF INSTR(RANDOMNUM\$, GENRAN\$) < 1 THEN RANDOMNUM\$ = RANDOMNUM\$ + GENRAN\$ + " "
LOOP

'sort numbers low to high
L1 = 1
RANPAT\$ = ""
DO WHILE L1 <= NSET
IF L1 < 10 THEN L1\$ = "0" + LTRIM\$(STR\$(L1)) ELSE L1\$ = LTRIM\$(STR\$(L1))
IF INSTR(RANDOMNUM\$, L1\$) > 0 THEN RANPAT\$ = RANPAT\$ + L1\$ + " "
L1 = L1 + 1
LOOP

n1\$ = mid\$(RANPAT\$, 1, 2)
n2\$ = mid\$(RANPAT\$, 4, 2)
n3\$ = mid\$(RANPAT\$, 7, 2)
n4\$ = mid\$(RANPAT\$, 10, 2)
n5\$ = mid\$(RANPAT\$, 13, 2)
IF Pickt = 6 then n6\$ = mid\$(RANPAT\$, 16, 2)

'filters

print RANPAT\$
IF INKEY\$ <> "" THEN SYSTEM 'PRESS ANY KEY TO EXIT
GOTO RANDOMPICK

' Simple RNG fUNCTION
FUNCTION RAND (Low, Hi)
RANDOMIZE TIMER
RANDOMIZE RND * 10
IF Low > Hi THEN
TEMP = Low
Low = Hi
Hi = TEMP
END IF
RAND = INT((Hi - Low + 1) * RND + Low)
END FUNCTION

'___________________________________

RL

mid-Ohio
United States
Member #9
March 24, 2001
19828 Posts
Online
 Posted: April 13, 2011, 8:24 pm - IP Logged

RJOh

I have a quick question, when you pick your numbers does your software generate and filter at the same time

or do you generate the sets first and then apply the filters after the fact.  I added a RNG option to what I am

using and it filters as it goes.  If I use more then just a few filters then the run can take a very long time  to

pick even 10 lines.  #2. Do you set ranges for each number/position in the generating stage. I included the

code for the RNG that I am using

QB64 CODE

'_____________________________

Pickt = 5 'numbers per set
Nset = 39 'numbers in matrix

'Pick 5 or 6 random numbers
RANDOMPICK:
RANDOMNUM\$ = ""
DO WHILE LEN(RANDOMNUM\$) < (PICKT * 3)
GENRAN = RAND(1, NSET)
IF GENRAN < 10 THEN GENRAN\$ = "0" + LTRIM\$(STR\$(GENRAN)) ELSE GENRAN\$ = LTRIM\$(STR\$(GENRAN))
IF INSTR(RANDOMNUM\$, GENRAN\$) < 1 THEN RANDOMNUM\$ = RANDOMNUM\$ + GENRAN\$ + " "
LOOP

'sort numbers low to high
L1 = 1
RANPAT\$ = ""
DO WHILE L1 <= NSET
IF L1 < 10 THEN L1\$ = "0" + LTRIM\$(STR\$(L1)) ELSE L1\$ = LTRIM\$(STR\$(L1))
IF INSTR(RANDOMNUM\$, L1\$) > 0 THEN RANPAT\$ = RANPAT\$ + L1\$ + " "
L1 = L1 + 1
LOOP

n1\$ = mid\$(RANPAT\$, 1, 2)
n2\$ = mid\$(RANPAT\$, 4, 2)
n3\$ = mid\$(RANPAT\$, 7, 2)
n4\$ = mid\$(RANPAT\$, 10, 2)
n5\$ = mid\$(RANPAT\$, 13, 2)
IF Pickt = 6 then n6\$ = mid\$(RANPAT\$, 16, 2)

'filters

print RANPAT\$
IF INKEY\$ <> "" THEN SYSTEM 'PRESS ANY KEY TO EXIT
GOTO RANDOMPICK

' Simple RNG fUNCTION
FUNCTION RAND (Low, Hi)
RANDOMIZE TIMER
RANDOMIZE RND * 10
IF Low > Hi THEN
TEMP = Low
Low = Hi
Hi = TEMP
END IF
RAND = INT((Hi - Low + 1) * RND + Low)
END FUNCTION

'___________________________________

RL

...when you pick your numbers does your software generate and filter at the same time or do you generate the sets first and then apply the filters after the fact.

It kinda does both. I usually pick combinations within parameters that cover 95% of previous winners in their categories so when combinations are generated that don't fit within all those parameters, they are rejected until one that fits is found and then I have the choice of accepting or rejecting it.

The program will continue picking combinations as long as numbers are left in the number pool.  For a 649 game I usually start out with 49 numbers (1-49).  After 6 combinations have been accepted, I know there are only 13 numbers left in the pool so if the program is having a hard time coming up new combinations, I can wait or add more numbers to the pool.  I also could have started out with more numbers in the pool, for example I could have started our with 245 numbers(1-49)x5.  I usually start out with just one of each numbers since that's the easiest way to be sure I come close to using all the possible numbers at least once.

* you don't need to buy more tickets, just buy a winning ticket *

United States
Member #59354
March 13, 2008
3972 Posts
Offline
 Posted: April 13, 2011, 9:31 pm - IP Logged

...when you pick your numbers does your software generate and filter at the same time or do you generate the sets first and then apply the filters after the fact.

It kinda does both. I usually pick combinations within parameters that cover 95% of previous winners in their categories so when combinations are generated that don't fit within all those parameters, they are rejected until one that fits is found and then I have the choice of accepting or rejecting it.

The program will continue picking combinations as long as numbers are left in the number pool.  For a 649 game I usually start out with 49 numbers (1-49).  After 6 combinations have been accepted, I know there are only 13 numbers left in the pool so if the program is having a hard time coming up new combinations, I can wait or add more numbers to the pool.  I also could have started out with more numbers in the pool, for example I could have started our with 245 numbers(1-49)x5.  I usually start out with just one of each numbers since that's the easiest way to be sure I come close to using all the possible numbers at least once.

RJOh

Thanks for the reply, This makes much more sense then what I was trying.  The approach I used above

does not take into account number coverage and treats each set in a first come first serve.  If the set

passes all the filters it is accepted which could produce sets with very few total numbers.  I was working

on a option to limit paris, trays and quads matches but need to work out a way to control how many of

each to allow without putting the process into a endless loop.  This seems like a easy thing to do but

has proven much more difficult to code.  I think I might be able to count number totals of accepted sets

and then use this as a self adjusting filter at the very last stage.  Thanks again for the reply.

RL

mid-Ohio
United States
Member #9
March 24, 2001
19828 Posts
Online
 Posted: April 14, 2011, 11:25 am - IP Logged

RJOh

Thanks for the reply, This makes much more sense then what I was trying.  The approach I used above

does not take into account number coverage and treats each set in a first come first serve.  If the set

passes all the filters it is accepted which could produce sets with very few total numbers.  I was working

on a option to limit paris, trays and quads matches but need to work out a way to control how many of

each to allow without putting the process into a endless loop.  This seems like a easy thing to do but

has proven much more difficult to code.  I think I might be able to count number totals of accepted sets

and then use this as a self adjusting filter at the very last stage.  Thanks again for the reply.

RL

I like having the number pool because it allows me to simulate two different styles of play.   (A)It's like having all the numbers in a sack and pulling out 5 or 6 at a time until they're all gone thus allowing me to nearly play all the numbers or (B) pulling out 5 or 6 at a time, recording them and throwing them back into the sack and picking again much as a regular drawing where the previous picks have no effect on the latest pick.  It all depends on my style of playing and I don't have to rewrite my program if I change my style, I just put in different parameters.

In the past I have called my program a system but in truth it's the parameters that I enter that is the system.   I can use any size number pool or a group of select numbers to pick combinations of particular sums, ranges or configurations.  I can even color code numbers with 3 colors red(hot), green(normal) and blue(cold) and mix them any way I want by color, number of digits, digit sums or even decade covered along with other variables I've noticed when analizing past winners.  Almost any system can be simulated by the parameters I enter in my program, all I need are the parameters that makes a system a winner.

* you don't need to buy more tickets, just buy a winning ticket *

United States
Member #59354
March 13, 2008
3972 Posts
Offline
 Posted: April 14, 2011, 8:38 pm - IP Logged

I like having the number pool because it allows me to simulate two different styles of play.   (A)It's like having all the numbers in a sack and pulling out 5 or 6 at a time until they're all gone thus allowing me to nearly play all the numbers or (B) pulling out 5 or 6 at a time, recording them and throwing them back into the sack and picking again much as a regular drawing where the previous picks have no effect on the latest pick.  It all depends on my style of playing and I don't have to rewrite my program if I change my style, I just put in different parameters.

In the past I have called my program a system but in truth it's the parameters that I enter that is the system.   I can use any size number pool or a group of select numbers to pick combinations of particular sums, ranges or configurations.  I can even color code numbers with 3 colors red(hot), green(normal) and blue(cold) and mix them any way I want by color, number of digits, digit sums or even decade covered along with other variables I've noticed when analizing past winners.  Almost any system can be simulated by the parameters I enter in my program, all I need are the parameters that makes a system a winner.

RJOh

I am still working on it and would like to know if I can pick your mind  from time to time.  I sometimes

have to wait a couple weeks on the data to show using my program and want to try and come up with

a method that would allow me to play more often.

Thanks

RL

mid-Ohio
United States
Member #9
March 24, 2001
19828 Posts
Online
 Posted: April 14, 2011, 9:21 pm - IP Logged

RJOh

I am still working on it and would like to know if I can pick your mind  from time to time.  I sometimes

have to wait a couple weeks on the data to show using my program and want to try and come up with

a method that would allow me to play more often.

Thanks

RL

I know what you mean, I'm have the same problems with more than 2200 records of Rolling Cash5, I trying to come up with ways to complied that data faster since RC5 is a daily game and most of my programs are limited to less than 600 records so I usually have to result to other methods of getting that data complied which I don't have enough time before another drawing has occured.

I'll be glad to share any knowledge I have with you but remember I've done all my programming in GWBasic and I noticed your program was in QB.  The only connection I've ever had with QB is using its complier to complie my GWBasic programs.

* you don't need to buy more tickets, just buy a winning ticket *

Fort Worth, TX
United States
Member #106060
February 11, 2011
188 Posts
Offline
 Posted: April 15, 2011, 12:16 am - IP Logged

Why does this data take so long to compile guys?

How you do anything is how you do everything.

United States
Member #59354
March 13, 2008
3972 Posts
Offline
 Posted: April 15, 2011, 1:20 am - IP Logged

Why does this data take so long to compile guys?

mayhem

For me it's a matter of finding supporting data.  I know my choices are just guesses in the grand scheme

of things but a guess can have a lot of thought behind it.  I can set everything and make a run in less then

5 minuets if I just wanted to take the top 2 or 3 values for each filter or stage  but I wait for certain cycles and

then hold off as long as I dare and then make a very refined run.  My main system goes through every set

in the matrix each run and may find as few as 10 lines total.  There are certain mixes of key values that

when combined reduce a rate that is almost unbelievable.  I have to wait until so many of these values line

up so to speak before I attempt to play.  It is easy for me to reduce to less then 20 lines any day if I want to

use many of the filter options but I find that when using many settings I make many mistakes.  I like to think

of playing the lottery like looking for a lost treasure.  Each night I get out all my information and make a plan

where I will search the next day.  I don't worrie about someone finding it before I do because a new chest

will take it's place the next draw.  Playing the Lottery should be fun because it is a game and where else

can you hunt a major treasure from your favorite easy chair.  Most days the wife will even run and pick up

the tickets.

RL

mid-Ohio
United States
Member #9
March 24, 2001
19828 Posts
Online
 Posted: April 15, 2011, 1:32 am - IP Logged

Why does this data take so long to compile guys?

You obviously have spent little or no time compiling lottery data.

The time depends on the amount of data you have and the amount of information you want the computer to generate.  For example I have 2243 drawings in my RC5 file and it take about .3 sec. to compare one drawing with the other 2242 but if I want to compare all 2243 with the other 2243, that's 750 secs. or about 10 minutes.  That's very fast compared to doing the same thing with paper and pencil but a super computer using a program designed by a super programmer using machine language could do the whole job in less than 10 secs.  It takes about 25 secs for my program to check all 2243 drawing combinations for length, range, sums and etc., but once I have all that information it has to be sorted several different ways to make sense of it which could take another 5-10 minutes each time sorted. Five and ten minutes here and there add up after a while and I may have spent almost an hour and that's more time than most folks are willing to spend generating data to play a game.

Keep in mind that my programs are written using a high level computer language and them complied which doesn't make the speediest computer programs.

* you don't need to buy more tickets, just buy a winning ticket *

 Page 6 of 10