Lottery Forecasting
- Thread starter Tarantula
- Start date
ntrile,
In the third paragraph of my original post, I stated:
"I'm not going to debate the ability to "forecast" lotteries ...."
However, I've decided to address your comments anyway ...
You say:
"I do not believe in any forecasting tool if you have odds extremely against you. "
In any lottery, the odds are stacked against you so bad that there isn't any reason to play. Statistically speaking you have a better chance of being struck by lightning than winning the lottery. Those who play the lottery are looking for the long shot.
I personally don't believe in the concept of "random", I think it's an ideal that can never be achieved. Even using computers - if you ask your computer for a random number (using whatever method you choose), it isn't really random - the routine which generated the random number requires a seed, the random number is pseudo random - it is based on the seed provided,
so if you start your computer each day and get a random number sequence using 1 as the seed, you will always get the same sequence of "random" numbers. Most programs, in order to avoid this problem will use the number of seconds past midnight or clock ticks as the seed when the random number generator is intialized.
With all that in mind, in trying to forecast the lottery - I am not trying to predict the actual 5 or 6 numbers that are drawn (depending on the lottery), I am trying to find a method to select a small group of numbers that when played using the proper system will make money.
For a simple example: Here I play a 5/44 lottery. If I could manage to pick 5 numbers, and 50% of the time have two numbers right. I could then play $1 every drawing, and be guaranteed to break even - and since I'm playing every drawing there's a small chance that I would win more, possibly even the jackpot. I would consider this situation a success, anything better would be a phenominal success - and I'll be travelling around to win different lotteries.
I view the lottery not as a single event (each drawing), I view the lottery as a series of related events that is similating randomness - something I don't believe can be achieved.
If you toss a coin 9 times in a row and it comes up "heads", when you go to toss the 10th time the chance it will come up heads is 1/2, however the chance that it will come up 10 times in a row is 1/1024 - so when you throw the coin the 10th time, I'm betting it comes up tails. This may not seem to fit, but it's much simpler than an example using large numbers.
By looking at the trends in the last 5 - 15 drawings, and some overall history facts, I believe it can be done, and I think I can do it using the computer as a tool (I wrote my first computer program some 8 or so years before there was a "Personal Computer"), so I have significant resources to draw on. Of course that also shows my age, so I'm hoping to accomplish this quickly 8).
If the computer picks a "random" set of numbers - this isn't really random now is it?
In the case of the "ball" machines, imagine how difficult it must be to create 49 (or however many) balls that have their numbers painted on the side in some fashion to be EXACTLY the same there have to be differences, no matter how minute they are.
You say:
Only 50% of the money wagered are reurned to players.
In all forms of gambling the odds are with "the house", they have to get something otherwise they wouldn't be "the house". Also I can't say about all lotteries, but I think at this point most of the lotteries give some of the money to beneficial programs. I know in PA it goes to senior citizens, and in MO it goes to public education - both things I don't mind losing to. I realize that alot of the money also goes into the "overhead" associated with running a lottery - imagine just the money the lottery spends on advertising - it's everywhere.
I've probably said way more than necessary.
Thanks.
In the third paragraph of my original post, I stated:
"I'm not going to debate the ability to "forecast" lotteries ...."
However, I've decided to address your comments anyway ...
You say:
"I do not believe in any forecasting tool if you have odds extremely against you. "
In any lottery, the odds are stacked against you so bad that there isn't any reason to play. Statistically speaking you have a better chance of being struck by lightning than winning the lottery. Those who play the lottery are looking for the long shot.
I personally don't believe in the concept of "random", I think it's an ideal that can never be achieved. Even using computers - if you ask your computer for a random number (using whatever method you choose), it isn't really random - the routine which generated the random number requires a seed, the random number is pseudo random - it is based on the seed provided,
so if you start your computer each day and get a random number sequence using 1 as the seed, you will always get the same sequence of "random" numbers. Most programs, in order to avoid this problem will use the number of seconds past midnight or clock ticks as the seed when the random number generator is intialized.
With all that in mind, in trying to forecast the lottery - I am not trying to predict the actual 5 or 6 numbers that are drawn (depending on the lottery), I am trying to find a method to select a small group of numbers that when played using the proper system will make money.
For a simple example: Here I play a 5/44 lottery. If I could manage to pick 5 numbers, and 50% of the time have two numbers right. I could then play $1 every drawing, and be guaranteed to break even - and since I'm playing every drawing there's a small chance that I would win more, possibly even the jackpot. I would consider this situation a success, anything better would be a phenominal success - and I'll be travelling around to win different lotteries.
I view the lottery not as a single event (each drawing), I view the lottery as a series of related events that is similating randomness - something I don't believe can be achieved.
If you toss a coin 9 times in a row and it comes up "heads", when you go to toss the 10th time the chance it will come up heads is 1/2, however the chance that it will come up 10 times in a row is 1/1024 - so when you throw the coin the 10th time, I'm betting it comes up tails. This may not seem to fit, but it's much simpler than an example using large numbers.
By looking at the trends in the last 5 - 15 drawings, and some overall history facts, I believe it can be done, and I think I can do it using the computer as a tool (I wrote my first computer program some 8 or so years before there was a "Personal Computer"), so I have significant resources to draw on. Of course that also shows my age, so I'm hoping to accomplish this quickly 8).
If the computer picks a "random" set of numbers - this isn't really random now is it?
In the case of the "ball" machines, imagine how difficult it must be to create 49 (or however many) balls that have their numbers painted on the side in some fashion to be EXACTLY the same there have to be differences, no matter how minute they are.
You say:
Only 50% of the money wagered are reurned to players.
In all forms of gambling the odds are with "the house", they have to get something otherwise they wouldn't be "the house". Also I can't say about all lotteries, but I think at this point most of the lotteries give some of the money to beneficial programs. I know in PA it goes to senior citizens, and in MO it goes to public education - both things I don't mind losing to. I realize that alot of the money also goes into the "overhead" associated with running a lottery - imagine just the money the lottery spends on advertising - it's everywhere.
I've probably said way more than necessary.
Thanks.
You wrote:
In any lottery, the odds are stacked against you so bad that there isn't any reason to play.
I just want to say that you will probably spend all your life playing classical Lotto games and earn nothing. The point is to find the game where you have better chance to beat to house if equiped with proper tools.
In any lottery, the odds are stacked against you so bad that there isn't any reason to play.
I just want to say that you will probably spend all your life playing classical Lotto games and earn nothing. The point is to find the game where you have better chance to beat to house if equiped with proper tools.
Hi Tarantula,
Compliments on the efforts of forecasting the lotteries. Certainly you are not alone in this noble and almost (should I say totally) impossible task. You better don't discuss about the ability of forecasting as you stated in your first post. How wrong you are in betting "for sure" after nine heads your tail (when you know that the probability for head to come out is still 1/2) is well demonstrated in what Charles2 has posted about the number 53 in the Venice lottery in this forum.
Anyways, I wish all the possible success in your forecasting activities.
Rob
Compliments on the efforts of forecasting the lotteries. Certainly you are not alone in this noble and almost (should I say totally) impossible task. You better don't discuss about the ability of forecasting as you stated in your first post. How wrong you are in betting "for sure" after nine heads your tail (when you know that the probability for head to come out is still 1/2) is well demonstrated in what Charles2 has posted about the number 53 in the Venice lottery in this forum.
Anyways, I wish all the possible success in your forecasting activities.
Rob
ntrile,
You are probably correct, I will spend the remainder of my life playing various lotteries and in the end win nothing major, hell I probably won't even break even.
If it was "gambling" that I was looking for, then I'd get a few people here locally to play poker once a week - and a combination of luck and skill would determine my results (I think I could do well). But I have a thing for lotteries, and numbers - so here I am, within that I am doing exactly what you describe - I've picked the lottery with the best odds available to me locally, and I'm designing the tools I use myself on the computer - bottom line is that simple. Someone has to design the tools initially.
From your posts, it sounds like you don't like the lottery - is that correct? If so, what are you doing here?
What's your game of choice?
Rob50,
I like impossible just about as much as random. What seemed impossible 100 years ago, is not only possible today but an everyday thing. Today's science fiction, is tomorrow's technology.
I'll have to look up the number 53 thing and check it out.
I'm not saying that betting "Tails" is the correct choice, I'm saying that under those circumstances I'll do it every time - and I'll win 50% of the time just like anyone who makes either choice. So choosing Tails is just as good as choosing Heads - you're only disagreeing with my choice based on the logic behind it.
Thanks for the encouragement, win or lose - I'll be playing the games and having fun trying!
You are probably correct, I will spend the remainder of my life playing various lotteries and in the end win nothing major, hell I probably won't even break even.
If it was "gambling" that I was looking for, then I'd get a few people here locally to play poker once a week - and a combination of luck and skill would determine my results (I think I could do well). But I have a thing for lotteries, and numbers - so here I am, within that I am doing exactly what you describe - I've picked the lottery with the best odds available to me locally, and I'm designing the tools I use myself on the computer - bottom line is that simple. Someone has to design the tools initially.
From your posts, it sounds like you don't like the lottery - is that correct? If so, what are you doing here?
What's your game of choice?
Rob50,
I like impossible just about as much as random. What seemed impossible 100 years ago, is not only possible today but an everyday thing. Today's science fiction, is tomorrow's technology.
I'll have to look up the number 53 thing and check it out.
I'm not saying that betting "Tails" is the correct choice, I'm saying that under those circumstances I'll do it every time - and I'll win 50% of the time just like anyone who makes either choice. So choosing Tails is just as good as choosing Heads - you're only disagreeing with my choice based on the logic behind it.
Thanks for the encouragement, win or lose - I'll be playing the games and having fun trying!
Brad
Member
Tarantula said:
I think 4+B from 9 picks is a very good result. Let's say someone played a full wheel with 9 numbers of 84 lines at a cost of $56 ($2/3 lines) the win would have been $260 (plus 12 free lines for the next draw).
If your can forecast like this regularly, even once every 3 or 4 draws, you'll allow for a positive ROI and make a lot of new friends in the process
Cheers
Tarantula said:
Tarantula,
Thanks, I will start monitoring your forecasts.
My handle is not linked to anyone famous. It stems from prologed periods in the bush during my army days. These prolonged periods in the bush changed your personality and you you became waht is known as bush happy.
Cheers,
BushHappy
Brad,
Thanks for your insight, for now I'm going to call it beginner's luck - but it's a start.
That prediction was made using only the data from the last 14 drawings. One of the parts of my program is a simulator, it takes my analysis / weighting routines and starts at the beginning of the data applying it's forecast, then comparing it to the actual winners to see how it performed (not including the bonus ball - I ignore that for analysis purposes, and consider it just that - a bonus).
My analysis consider the last n drawings to be more significant than the first drawings (they are still taken into consideration however) - right now because I have a limited amount of data to work with n = 5, I've found that for most lotteries n = 7 - 15 is much better - in some cases I have it set as high as 42.
Since the simulation does not actually generate tickets used (way too much time and space involved), the results are a little more vague but using my simulator, and forecasting the 10 drawings prior to my forecast the results are:
0/12 - 2 times (18%)
1/12 - 3 times (27%)
2/12 - 3 times (27%)
3/12 - 1 time (9%)
4/12 - 1 time (9%)
followed by my posted prediction for another 4/12 (n = 12, for the target group - I set this to about 25% of the numbers typically) - all of the other parameters are borrowed from my Pick 5 and Pick 6 simulations, so I think there's much room for parameter adjustment. I still have alot of work to do on the analysis / weighting portion of the program as well.
Thanks for your insight, for now I'm going to call it beginner's luck - but it's a start.
That prediction was made using only the data from the last 14 drawings. One of the parts of my program is a simulator, it takes my analysis / weighting routines and starts at the beginning of the data applying it's forecast, then comparing it to the actual winners to see how it performed (not including the bonus ball - I ignore that for analysis purposes, and consider it just that - a bonus).
My analysis consider the last n drawings to be more significant than the first drawings (they are still taken into consideration however) - right now because I have a limited amount of data to work with n = 5, I've found that for most lotteries n = 7 - 15 is much better - in some cases I have it set as high as 42.
Since the simulation does not actually generate tickets used (way too much time and space involved), the results are a little more vague but using my simulator, and forecasting the 10 drawings prior to my forecast the results are:
0/12 - 2 times (18%)
1/12 - 3 times (27%)
2/12 - 3 times (27%)
3/12 - 1 time (9%)
4/12 - 1 time (9%)
followed by my posted prediction for another 4/12 (n = 12, for the target group - I set this to about 25% of the numbers typically) - all of the other parameters are borrowed from my Pick 5 and Pick 6 simulations, so I think there's much room for parameter adjustment. I still have alot of work to do on the analysis / weighting portion of the program as well.
ntrile,
Where do you get the House Advantage of 2.5%?
I am not trying to predict the lottery, I'm forecasting it much like your local weatherman attempts to forecast the weather.
A prediction would be that these 6 (or 5, or whatever) will be drawn. A forecast is generalizing what will happen.
From memory (correct me if I'm wrong) roulette has 0, 00, and 1 - 36 which from my terms translates it into a 1/38 lottery, and yes my systems can forecast for a 1/38 lottery (in fact I can forecast for any lottery with my system).
Now roulette has a few extra bets you can make red/black, numbers 1-12,13-24,25-36, and a few other variations playing on the lines between numbers. So if I were going to write a program to assist roulette players, I would factor in these additional bets, the payouts for each type of bets and come up with a forecast based on the past n "draws". Then I would port the code to a PDA platform so I could walk into a casino and record several roulette results until I had at least n+1 draws of data.
I understand your disbelief at anyone being able to forecast the lottery, I also have doubts at times. However, I have had a few successes - on Jan 22, 2005 I picked 5/5 in Virginia's cash 5 game (5/34), on Feb 5, 2005 I picked 5/5 in Pennsylvania's cash 5 game (5/39). Now this is in the infancy of my systems, I am still adjusting the parameters by which the forecast is made - each lottery has it's own individual set of parameters, in any pick 5 lottery it's relatively easy for me to pick 4 - I also have much work to do on the wheels systems that I am using which can also improve my results.
The bottom line is - no matter what anyone else says or does, I'm going to forecast lotteries, and post the forecasts and results on my website for the world to see. If I fail to acheive my desired results, you and the rest of the world can laugh at the fool who thought he could beat the lottery. If I succeed - I'll be laughing all the way to the bank, and so will those who use my forecasts.
Where do you get the House Advantage of 2.5%?
I am not trying to predict the lottery, I'm forecasting it much like your local weatherman attempts to forecast the weather.
A prediction would be that these 6 (or 5, or whatever) will be drawn. A forecast is generalizing what will happen.
From memory (correct me if I'm wrong) roulette has 0, 00, and 1 - 36 which from my terms translates it into a 1/38 lottery, and yes my systems can forecast for a 1/38 lottery (in fact I can forecast for any lottery with my system).
Now roulette has a few extra bets you can make red/black, numbers 1-12,13-24,25-36, and a few other variations playing on the lines between numbers. So if I were going to write a program to assist roulette players, I would factor in these additional bets, the payouts for each type of bets and come up with a forecast based on the past n "draws". Then I would port the code to a PDA platform so I could walk into a casino and record several roulette results until I had at least n+1 draws of data.
I understand your disbelief at anyone being able to forecast the lottery, I also have doubts at times. However, I have had a few successes - on Jan 22, 2005 I picked 5/5 in Virginia's cash 5 game (5/34), on Feb 5, 2005 I picked 5/5 in Pennsylvania's cash 5 game (5/39). Now this is in the infancy of my systems, I am still adjusting the parameters by which the forecast is made - each lottery has it's own individual set of parameters, in any pick 5 lottery it's relatively easy for me to pick 4 - I also have much work to do on the wheels systems that I am using which can also improve my results.
The bottom line is - no matter what anyone else says or does, I'm going to forecast lotteries, and post the forecasts and results on my website for the world to see. If I fail to acheive my desired results, you and the rest of the world can laugh at the fool who thought he could beat the lottery. If I succeed - I'll be laughing all the way to the bank, and so will those who use my forecasts.
South Africa Lottery 16-Feb-05
Tarantula.
Your first forecast for the South Africa Lottery drawn on 16-Feb-05 was not very good.
Only 1 correct out of the top ranked 12 numbers
Only 2 correct out of the top 24 ranked numbers
Only 3 correct out of the top 33 ranked numbers
Only 4 correct out of the top 39 ranked numbers
Only 5 correct out of the top 41 ranked numbers
Only 6 correct out of the top 48 ranked numbers
Will the forecasting improve?
Do you only use the previous 13 draws for your forecasts?
The complete history can be downloaded from the South African National Lottery Web site.
Cheers,
BushHappy
Tarantula.
Your first forecast for the South Africa Lottery drawn on 16-Feb-05 was not very good.
Only 1 correct out of the top ranked 12 numbers
Only 2 correct out of the top 24 ranked numbers
Only 3 correct out of the top 33 ranked numbers
Only 4 correct out of the top 39 ranked numbers
Only 5 correct out of the top 41 ranked numbers
Only 6 correct out of the top 48 ranked numbers
Will the forecasting improve?
Do you only use the previous 13 draws for your forecasts?
The complete history can be downloaded from the South African National Lottery Web site.
Cheers,
BushHappy
BushHappy,
Not very good is an understatement - it sucked!
For the forecast, I use all the data that I have - the more the better (although recent drawings carry more weight than older drawings). I downloaded the data from their web site but had problems importing it due to the non-us date format, so I just manually added a minimum number of drawings.
In order to optimize the parameters that define the forecast, I need 30 - 60 drawings worth of data - currently for the SA 6/49 I'm using the same parameters from one of the other 6/49 lottos that I am tracking and as a result are probably nowhere near optimum.
I'm sure the forecasts will improve, once I get more data and optimize the parameters of the forecast.
Not very good is an understatement - it sucked!
For the forecast, I use all the data that I have - the more the better (although recent drawings carry more weight than older drawings). I downloaded the data from their web site but had problems importing it due to the non-us date format, so I just manually added a minimum number of drawings.
In order to optimize the parameters that define the forecast, I need 30 - 60 drawings worth of data - currently for the SA 6/49 I'm using the same parameters from one of the other 6/49 lottos that I am tracking and as a result are probably nowhere near optimum.
I'm sure the forecasts will improve, once I get more data and optimize the parameters of the forecast.
Tarantula,
I assume by main page, you mean http://weblotto.org/
I went there and this site does not open fpr me; just a blank screen.
I assume by main page, you mean http://weblotto.org/
I went there and this site does not open fpr me; just a blank screen.
Success!!!
I realize that everyone here is for the bigger lotteries, however I have to tell you about tonight. Sunday night I decided to play my favorite lottery 5/44 here, and I did fair picking 3 / 5 from a 6$ investment made $29, I skipped playing monday morning, but decided to play Monday evening - I played $14 worth of tickets and managed to pick 5/11 in my forecast, I had 4/5 in my system for just under $800 in winners - It's now official - I'm playing on their money!
There are many adjustments I still have to make on my forecasting system, and many wheels I wish to add, but as of today I've picked 5/5 in three different state lotteries. Hopefully with these strategies I am developing I'll be able to get us a 6/6, or 7/7 in the big lotteries (I have already managed 5/6 in a few lotteries as well, but no 6/6 yet).
Keep checking my web site to see my progress, the lottery that I'm currently actively playing is the MO Show Me 5 Paydown, I'll probably be attacking the PowerBall next! I'll continue to forecast any lotteries that I have and are requested.
Thanks again for everyone's comments and questions.
May your luck run as good as mine has!!!
I realize that everyone here is for the bigger lotteries, however I have to tell you about tonight. Sunday night I decided to play my favorite lottery 5/44 here, and I did fair picking 3 / 5 from a 6$ investment made $29, I skipped playing monday morning, but decided to play Monday evening - I played $14 worth of tickets and managed to pick 5/11 in my forecast, I had 4/5 in my system for just under $800 in winners - It's now official - I'm playing on their money!
There are many adjustments I still have to make on my forecasting system, and many wheels I wish to add, but as of today I've picked 5/5 in three different state lotteries. Hopefully with these strategies I am developing I'll be able to get us a 6/6, or 7/7 in the big lotteries (I have already managed 5/6 in a few lotteries as well, but no 6/6 yet).
Keep checking my web site to see my progress, the lottery that I'm currently actively playing is the MO Show Me 5 Paydown, I'll probably be attacking the PowerBall next! I'll continue to forecast any lotteries that I have and are requested.
Thanks again for everyone's comments and questions.
May your luck run as good as mine has!!!