Welcome Guest
Log In | Register )
You last visited December 7, 2016, 10:55 pm
All times shown are
Eastern Time (GMT-5:00)

Gail Howard software

Topic closed. 39 replies. Last post 12 years ago by BobP.

Page 2 of 3
PrintE-mailLink
Avatar
New Member
Jacksonville, Florida
United States
Member #5685
July 16, 2004
9 Posts
Offline
Posted: July 18, 2004, 8:37 pm - IP Logged

Just another observation that comes to mind.

Take a 6/48 game.  It's just a hypothetical game for consideration.  (It's twice 24 numbers.)

If some software takes 24 numbers and "predicts" 2, 3 or 4 winners, it means:

  ... The other 24 numbers, not chosen by the software, had 4, 3, or 2 winners.

To have "prediction" value in 24 numbers the software would have to consistently pick 4 winners, leaving just 2 in the other 24.  For a 6/48 it would need to do that more than half of the time.  Getting 3/3 doesn't cut it.

It would need to show that kind of performance over a long series of draws.

Joe Roberts

CDEX

    Avatar
    Melbourne
    Australia
    Member #5731
    July 18, 2004
    36 Posts
    Offline
    Posted: July 19, 2004, 1:39 am - IP Logged

    Apagogeas,

    I have to say I find your success rates well worthy of follow up, but I am intrigued more about your choice of 24 balls and how on earth can one play such a number group successfully or at least within some sort of budget. Using a 4 from 6 wheel requires some 360 games and a 5 from 6 wheel probably 4500 games to play. This is way beyond my financial capabilities and I suspect many others. So can your program perform some sort of filtering process to remove unlikely gameplays but at the same time hopefully not destroy the chance of winning a major prize? 

    regards

    relowe

    and as you may have observed I am new in here. This is a great forum with good posting.  I hope I can maintain the standard.

      BobP's avatar - bobp avatar.png
      Dump Water Florida
      United States
      Member #380
      June 5, 2002
      3104 Posts
      Offline
      Posted: July 19, 2004, 3:24 am - IP Logged

      Just popping back in to point out you can wheel 100% 3if6 in 24 numbers in 21 lines without splitting the numbers into two independent wheels to do it. 

      What happens after that is more a matter of where you drop the numbers then which wheel you play. 

      Good lottery software will allow the user to select the pointer number positions to control where the numbers fall onto the wheel.

      Whether you are playing a 3if6 or a 5if6 wheel, the numbers you think will make the jackpot winning combination should be positioned to form that line(s) while playing the wheel for it's overall guarantee. 

      Sometimes I think the pointers should be followed below with wheel lines that use the numbers from pointer positions 01-06, 07-12, 13-18, 19-24,  so players know to take their jackpot shots while playing their wheels.  BobP

       

        lottoarchitect's avatar - waveform

        Greece
        Member #2815
        November 18, 2003
        502 Posts
        Offline
        Posted: July 19, 2004, 8:48 am - IP Logged

        CDEX,

        I made another run of my system in Greek Lotto (6/49 - even worse than your suggested test) and picked 24 numbers for each of the last 100 draws in my database. The same procedure applied as described in my previous posts (forecast). The classic probability for N=24 numbers in 6/49 games is:

        0/6=1,2%    1/6=9,1%    2/6=24,9%  3/6=33,3%    4/6=22,8%    5/6=7,5%    6/6=0,9%

        I used my HCD system and my results are:

        Draws reported: 100 Match 0: 0, Match 1: 2, Match 2: 23, Match 3: 33, Match 4: 30, Match 5: 9, Match 6: 3

        Here you can still see how much better are the results compared to classic probability. This is a totally different lottery than Florida Lotto and the same procedure used to predict all 100 draws (large scale as you say it).

        I'll not analyse the data more, you can see the facts. You just can't believe there is a system that can produce constantly good results. I don't blame you, I wouldn't either but here it is and ... works! In general we say the same thing in different words.

        Relowe, you state a good point here and the same matter was discussed by RJOh. He said that more than 13 numbers in a full wheel is impractical and I disagree with him. I currently use full wheels with 30 numbers (100000+ tickets). This ensures me I have a much greater chance to have all the winning numbers among them and then perform what I call "intelligent filtering". This allows me to bring my final tickets down to 50-150 (I think this is much more affordable). No need to say, I have much better chances to win (the top prize) that way compared to the common approach which is to try to have all winning numbers in a set of 13 at most and then wheel them.

        BobP, you mention a good feature (arrange the point numbers). This is good practice in case of partial coverage wheels where some numbers might appear more often than others. In case of well designed abbreviated wheels, this feature does not provide any important benefit (these wheels are well balanced). I say that because if you can feel what numbers might be in a ticket (and arrange the wheel accordingly) you should play them without a wheel!

        The fact that we want to use wheels of 24if3if6 etc proves the need to pick more numbers (even if we want to participate in lesser prizes). Now there is a good way to do this!

        If you have something to do, at least do it well...

          RJOh's avatar - chipmunk
          mid-Ohio
          United States
          Member #9
          March 24, 2001
          19828 Posts
          Online
          Posted: July 19, 2004, 4:50 pm - IP Logged
          Quote: Originally posted by apagogeas on July 19, 2004



          Relowe, you state a good point here and the same matter was discussed by RJOh. He said that more than 13 numbers in a full wheel is impractical and I disagree with him. I currently use full wheels with 30 numbers (100000+ tickets). This ensures me I have a much greater chance to have all the winning numbers among them and then perform what I call "intelligent filtering". This allows me to bring my final tickets down to 50-150 (I think this is much more affordable). No need to say, I have much better chances to win (the top prize) that way compared to the common approach which is to try to have all winning numbers in a set of 13 at most and then wheel them.






          Apagogeas,

          A demonstration might make your point clearer.  Why not pick 30 numbers which you think are most likely to come up in one of the next jackpot games, wheel them and with some "intelligent filtering" reduce the wheel to 200 or less lines and post them and see what the results are.  I doubt if your results would be any better than using a RNG with the same 30 numbers and some general parameters to generate the same amount of lines.

          RJOh

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

            BobP's avatar - bobp avatar.png
            Dump Water Florida
            United States
            Member #380
            June 5, 2002
            3104 Posts
            Offline
            Posted: July 19, 2004, 5:02 pm - IP Logged

            Even a well balanced wheel offers nothing more then its percentages of prize tier coverage. 

            You know the old chestnut example the lottery experts give to prove the need to wheel, where dumb guy makes up six tickets for his twelve numbers and gets all six among the twelve and wins 16 3# prizes, while smart guy wheels his and get a 4# win and a bunch of 3# wins.

            Well the prize both wanted to win was a jackpot, but they put zero effort into winning more then 3# and 4# prizes.  They trusted to luck the wheel would play above its guarantee and they were wrong as higher prizes are much harder to cover.

            There is a certain amount of logic to making up some jackpot shot tickets out of your numbers, perhaps as the first tickets on the wheel, then the wheel acts as insurance when you're wrong.  Even wheeling 24 numbers, you should wheel the other 24 on four lines with a 2if guarantee for insurance.

            What I hate is wheeling that takes my numbers and usurps control over where they fall on the wheel.

            Of course it depends on what the software is doing.  If the software uses wheels and filters, I expect to be able to import wheels of my own making such as front loaded and to position my numbers with the better choices up front.  Like Lottery Director does.

            If the software builds a filtered wheel on the fly out of my numbers like Lottohat or Lotwin, then I accept the loss of control over where the numbers fall and add my own extra plays for a jackpot shot.

            BobP

              lottoarchitect's avatar - waveform

              Greece
              Member #2815
              November 18, 2003
              502 Posts
              Offline
              Posted: July 19, 2004, 5:31 pm - IP Logged
              Quote: Originally posted by RJOh on July 19, 2004



              Quote: Originally posted by apagogeas on July 19, 2004



              Relowe, you state a good point here and the same matter was discussed by RJOh. He said that more than 13 numbers in a full wheel is impractical and I disagree with him. I currently use full wheels with 30 numbers (100000+ tickets). This ensures me I have a much greater chance to have all the winning numbers among them and then perform what I call "intelligent filtering". This allows me to bring my final tickets down to 50-150 (I think this is much more affordable). No need to say, I have much better chances to win (the top prize) that way compared to the common approach which is to try to have all winning numbers in a set of 13 at most and then wheel them.






              Apagogeas,

              A demonstration might make your point clearer.  Why not pick 30 numbers which you think are most likely to come up in one of the next jackpot games, wheel them and with some "intelligent filtering" reduce the wheel to 200 or less lines and post them and see what the results are.  I doubt if your results would be any better than using a RNG with the same 30 numbers and some general parameters to generate the same amount of lines.

              RJOh





              Doubt and criticism is welcome otherwise the world would be rather boring! What you suggest would be nice but:

              1) Currently I have so much work to do to even develop a system to fulfil your desire.

              2) The program will be available in a few days to download (I hope) so you can see by yourself what this "intelligent filtering" is and can do. So, no need to demonstrate something here.

              3) Even if I demonstrate something right now and fails, it proves nothing. I never said I have a system that hits the top prize in 50 lines on every tested draw. I've tested the power of my system and goes beyond anything I've seen so far in other programs. My proposal of 30 numbers (full wheel) is just a much better way to hit the top prize compared to wheeling only 13 numbers at most. Do you know why? If you begin with bad numbers, you end with nothing. If you begin with good numbers (30 numbers provide this), you can possibly end with something (or nothing). It is just another strategy and from my point of view (and testing) provides better results than trying to guess what 13 numbers to pick. Besides that do you perform some filtering on your tickets to eliminate some of them? If you say yes then this is my answer to the RNG matter too.

              If you have something to do, at least do it well...

                Avatar
                New Member
                Jacksonville, Florida
                United States
                Member #5685
                July 16, 2004
                9 Posts
                Offline
                Posted: July 19, 2004, 6:15 pm - IP Logged
                Quote: Originally posted by BobP on July 19, 2004

                (...)

                Of course it depends on what the software is doing. If the software uses wheels and filters, I expect to be able to import wheels of my own making such as front loaded and to position my numbers with the better choices up front.



                BobP






                Each time we've had an opportunity to look at front loaded wheels I've agreed with you on them. They make good sense, because the player usually had some priority in the way the numbers were chosen.



                For example:



                   ... If the player chooses 24 numbers out of a 49-number field, he's not just "shotgunning" some numbers. There must be some reason why he picked those 24 and none of the other 25.



                   ... To pick 24 out of 49, he had to use some 'method' for picking those numbers. Call it anything for this example; the name isn't important. Even if it's something as basic as a "hot" list (not a good term), the 24th number in that list couldn't have been as "hot" as the 1st number, and so on.



                So the idea in front loading is, if the player had a preference in picking the numbers to start with, then his preference ought to carry all the way through his wheeling. In other words, the "preferred" numbers ought to be wheeled together as much as possible, while still keeping the wheel consistent with its win guarantee.



                - - - -



                We've agreed on front loaded wheeling several times already. I just wanted to review it (above) for anyone who's not so familiar with them, in order to offer a different suggestion (coming up).



                You haven't been real happy with "Zone" or "split" wheels, for several reasons. Not to get into them here, because it's way outside the scope of this suggestion.



                There's a case where a certain kind of "Zone" wheel can give you an alternative to front loading. That's what I wanted to offer as a suggestion.



                If the idea is get maximum coverage of some preferred numbers, just put them together into one Zone in the wheel. Make that Zone have its own win guarantee, and then let those "Zone" numbers work with the rest of the numbers.



                For example:   Suppose the player has 24 numbers, including 8 that are more strongly "preferred" than the rest. Put those 8 numbers into a Zone. Let them get wheeled together in something like "2-to-4" numbers per line, with the rest of each line taken up by combining the other 16 numbers.



                The main idea is, if your "preferred" numbers happen to hit you'll know you have them on most of the lines in the wheel, not just in the front part of it.



                It will probably start out as a more expensive wheel than a front-loaded Abbreviated or open cover, but you could consider some filtering to reduce it. It looks like you might have had filtering in mind, in your original post, so this could work.



                Long post. Hope it's helpful.



                Joe

                CDEX

                  kbcherokee's avatar - buffalo2
                  Virginia
                  United States
                  Member #1712
                  June 20, 2003
                  1549 Posts
                  Offline
                  Posted: July 19, 2004, 6:38 pm - IP Logged

                  Friends:

                   

                  I use Gail's system all the time with good success. BUT, you have to study the numbers and use a chart to track the past draws to see where the numbers are heading ie: Bell-Curve and then make your number selections to wheel.

                  A great tool.

                  Chief 

                  May the warm winds of heaven blow softly upon your house. May  the Great Spirit bless all who enter there. May your mocassins make happy tracks in many snows. And may the rainbow always touch your shoulder.

                    RJOh's avatar - chipmunk
                    mid-Ohio
                    United States
                    Member #9
                    March 24, 2001
                    19828 Posts
                    Online
                    Posted: July 19, 2004, 6:45 pm - IP Logged
                    Quote: Originally posted by apagogeas on July 19, 2004



                    1) Currently I have so much work to do to even develop a system to fulfil your desire.

                    2) The program will be available in a few days to download (I hope) so you can see by yourself what this "intelligent filtering" is and can do. So, no need to demonstrate something here.

                    3) Even if I demonstrate something right now and fails, it proves nothing. I never said I have a system that hits the top prize in 50 lines on every tested draw...... 






                    Apagogeas,

                    I,m sorry I misread your response, I thought you had a system up and running that you were presently using that created wheels of over 100,000 lines and reduced them to less than 200 lines to play.  I'm presently using a RNG in which I put all the numbers that I want to play along with the group format(5 or 6 numbers), some parameters (sums, gaps, ranges, positional limits and etc.) and the numbers of lines I want to play (up to 500) which seems to work fine.  I look forward to hearing more about your program when it comes on line.

                    RJOh 

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

                      Avatar
                      Melbourne
                      Australia
                      Member #5731
                      July 18, 2004
                      36 Posts
                      Offline
                      Posted: July 19, 2004, 7:52 pm - IP Logged

                      Apagogeas,

                      Thanks for your response.  Well 30 numbers in a wheel and 'intelligently filter' that down to some 150 tickets or so.  That is a big task. Some of those numbers in the big wheel must surely be part of a shotgun approach unless the hot cold due warm dead system is very focussed. I was wondering about some combinational wheel concepts where more than one wheel might be used and then hopefully get the best out of each wheel.  Perhaps where you might have say 15 highly preferred numbers or so and run these through a 6if6 wheel and then for another 15 numbers total 30 (including the original 15) and then run these at say a 4if6 and so reduce the total games to play, but with your highest probablity numbers contained in what I might term a more dense relationship with each other. Would intelligent filtering handle such an approach? Or first I suppose I should ask, can your program handle more than one wheel at a time?

                      many thanks

                      relowe

                        lottoarchitect's avatar - waveform

                        Greece
                        Member #2815
                        November 18, 2003
                        502 Posts
                        Offline
                        Posted: July 20, 2004, 3:26 am - IP Logged

                        RJOh, no hard feelings. Developing a strategy for a particular game is much more than a few mouse clicks activity. You already know it in almost any strategy you may follow. Even picking your initial set of 13 numbers requires some analytical thoughts by yourself to proceed (otherwise your selection will be no better than randomly picked numbers). I want to be able to approach a lottery game in several ways and see what performs best and finally use it. I've designed the system in such a way that allows multiple strategies to be used and I've found that my current strategy (begin with 30 numbers) offer better results than other strategies. Perhaps, If I search a lottery game more, I may find an even better approach to use.

                        Relowe suggests a different strategy. My approach is more demanding to develop and he suggests another approach (using multiple wheels). Well, relowe this feature is supported and you can mix several wheels together of any type and combination. Also, "intelligent filtering" is not a button that analyses your current tickets and remove unwanted ones. It is a strategy that can be used in any approach you may try and hopefully decreases your initial tickets enough without worrying to discard the top prize winning ticket (assuming this was included on your initial tickets prior filtering). It requires the HCD system to do so but it offers very steady results. If more tickets need to be removed (to be in your budget) you can try the native number groups feature and still if more filtering is required, you can proceed in sums/gaps etc (this is the most inaccurate filtering procedure of all).

                        BobP and CDEX, I wish you had suggested this feature about wheels when I first asked about it some months ago. I would be able to include it in the program as well because it can be an excellent starting point for a new strategy as well. CDEX, I would like more info on your "Zones" if you can.

                        If you have something to do, at least do it well...

                          BobP's avatar - bobp avatar.png
                          Dump Water Florida
                          United States
                          Member #380
                          June 5, 2002
                          3104 Posts
                          Offline
                          Posted: July 20, 2004, 3:43 am - IP Logged
                          Quote: Originally posted by relowe on July 19, 2004



                          Apagogeas,

                          Thanks for your response.  Well 30 numbers in a wheel and 'intelligently filter' that down to some 150 tickets or so.  That is a big task. Some of those numbers in the big wheel must surely be part of a shotgun approach unless the hot cold due warm dead system is very focussed. I was wondering about some combinational wheel concepts where more than one wheel might be used and then hopefully get the best out of each wheel.  Perhaps where you might have say 15 highly preferred numbers or so and run these through a 6if6 wheel and then for another 15 numbers total 30 (including the original 15) and then run these at say a 4if6 and so reduce the total games to play, but with your highest probablity numbers contained in what I might term a more dense relationship with each other. Would intelligent filtering handle such an approach? Or first I suppose I should ask, can your program handle more than one wheel at a time?

                          many thanks

                          relowe




                          The hyper geometric distribution of numbers tells us you will average roughly two to three of the winning numbers for every set of 15 you choose to wheel unless you have a method of predicting that actually can put five or six winning numbers among your 15 being wheeled. 

                          Many people use four 12 number wheels to cover 48 of the 49 numbers, they generally average zero, one, two or three numbers per set of 12, very rare to get all six in one set.  In other words, sounds good, doesn't work.

                          The trick and you can see this even in the Lottohat demo is to start with a full wheel for all the numbers you want to work with including all of them and then reduce by filter and prize guarantee and finally by percentage of covrage until you hit your budget.  You have to make a consistant chain of correct choices or you lose.

                          http://www.lotto-logix.com/lottosoft.html  go here and download Lottohat's free demo and see how the concept works.  Apagogeas software will have many more filters, but the principle is the same.  BobP
                            Avatar
                            New Member
                            Jacksonville, Florida
                            United States
                            Member #5685
                            July 16, 2004
                            9 Posts
                            Offline
                            Posted: July 20, 2004, 11:39 am - IP Logged
                            Quote: Originally posted by apagogeas on July 20, 2004

                            CDEX, I would like more info on your "Zones" if you can.








                            Open covers, split wheels, zone wheels, and others have been discussed extensively in the Usenet newsgroup "rec.gambling.lottery" (RGL). Something new in wheels comes up several times per month, either as a reply to a request or from someone's new ideas. It is a very large amount of material.



                            It would be hard to cover it all again here, especially in this thread which has for its subject: a software program that can't even handle any of those wheels.



                            If you're about ready to release your own software, you'll probably want to check into that "RGL" newsgroup at some point anyway. (Even if you don't, your customers will.)



                            Suggestion. This is offered in friendship, so please don't take it the wrong way...   



                            If you do go into RGL, be careful about making opening statements like:



                               > It seems that you need some practice in maths!



                            There are maths people and cover designers there who will slice a person into Gyros for a making a statement like that. Lurk for awhile, read the FAQ (posted several times per month), and you'll get the feel of the place.



                            Also be careful about making "win" statements about your software. You'll be challenged to show consistent improvement over probability to the tune of 3SD or greater. Your potential customers will read everything, including the Google Groups archives a year from now.



                            Best of luck with your new software. Keep on the high road for ethics, advertising, and technical support.



                            You owe it to your customers and players for at least two reasons. (1) First, they sent money to you. (2) Afterward, they will be spending their money for playing combinations made with your software. They'll do so according to their expectations. Their expectations, initially, will be based on whatever they saw in your advertising and other statements. Go for the happy customer rather than the big sale, both of you will be proud of what you've done.



                            Cheers,



                            Joe

                            CDEX

                              lottoarchitect's avatar - waveform

                              Greece
                              Member #2815
                              November 18, 2003
                              502 Posts
                              Offline
                              Posted: July 21, 2004, 9:24 am - IP Logged

                              Thanks for the info CDEX. I

                              If you have something to do, at least do it well...