Greetings to all members and Admins. This is my first post here and my 4th visit to the site.
Greetings to Todd, you have a great web site! btw that test for Dementia was pretty hard... really likes the pics too, keep it coming man!
I've been working alone, for 4 eternal months, in a system to predict the pick-3 numbers, but this pick-3 is slighty different for others....
In my country they have 2 ping-pong-ball machines. A and B. both are used to get one number at 12:00pm and one at 8:00pm, so they get 4 numbers per day (each machine gets two), this is done live on TV (supposedly, you never know...).
If you play lets say number 010 for machine A at 12:00pm and that number comes from B (not A) then you get nothing, pretty hard isnt it?
So if you want to play a number then you have to choose from 4 different options, A or B and 12:00 or 8:00, couldnt be worst....
Im writing my program in Turbo Pascal 7.0 (object oriented) when i feel satisfied with it i will migrate the code to java with a nice interface, menus and stuff... The program tries to predict numbers based on my own filters and show a graphic chart with all the money you earned or lost... in the time played. In my studies i use a two years database from pick-3 rounds at 9 lottery companies (Pretty good collecion).
My program is Based in the following chart:
Machine Last Number it Got Numbers that are very not probably come on next round
A or B 333 333,33X,X33,3X3,etc..
A or B 944 999,994,499,949,etc..
and 1057 more filters...
I've related the current number of Machine A with the past one from the same machine (Dont mix numbers from different machines) they have nothing to do with each other, theyre independant, its a common error (i did). Some relations you can apply are > < + or -, im currently applying 1059 Filters to all the numbers.
They all are called static Filters, and i use them to filter around 250 (they sum 10%! of the total winnings) out of 1000. with the aprox. 750 (90% of the total winnings) numbers left you use your favorite stategy, im still working on this stage. I Will keep you informed.
Im based on the fact that its easier for me to find more than 900 number that are not probaly come out than to find less than 100 that are probably. Maybe im wrong maybe not.
well i wanted to post this for some reasons, the ones i can think of are:
1) Moral support, its been 4 months, all started with a handwriting statistic...and im doing this alone with no support and i think i need a little to keep trying. So if you have the time let me know what you think of my work, any kind of comment is very welcome. You can email me or write your reply here (with the permission of Todd of course), i will be happy to read it. My emal addy is cd_shopper@hotmail.com
2) Hear other strategies. Do you have a strategy you think i can use in my program? let me know so i can evaluate that strategy with the two years database i have from pick-3 rounds. Im open to any suggestion, dont be shy, i'll never reject the help you can give me.
3) Take advice from regular members who have more time working on this than me.
Well i will reply to messages asap, be patient because i dont have internet at home.
And thanks again to Todd for letting me post here, take care and Merry Xmas to all!