I have developed failed systems on the forums (most recently "soniq boom" for pick3/4) but I know they don't work... a simple circular reference error made me believe it was better than it actually was... developing out in the open allowed a few fresh pairs of eyes to see the error. I know these systems don't work, and I know I don't have any edge alone... so why not openly develop stuff with a group of like-minded experimenters in an open forum... we KNOW nothing works, so why put a pricetag on it? In the process of open development, everybody gets better at whatever the chosen method is... in the SB case it was excel... everyone involved got a bit better at excel, myself included. would be the same for MS access, basic programming, even pencil and paper tracking like tntea's v-tracs.
I am relatively sure we will never find a programming answer to our basic questions, but it's the thrill of the journey itself that keeps me in it. Most have systems that they keep a guarded secret that don't work... but they don't want to share them "just in case"... I have never held back on any research I have done, it's all been open, shared and most importantly, free. this goes for powerball as well as pick3/4.
I have thrown out my ideas and systems to be analyzed, scrutinized and debunked... they all have and I learned a great deal in the process.
I still stand by my original assessment of the talent pool here... IF we all worked together, we could probably make some serious headway into getting better than random results. Andrew's winHunter program is developed in the open on another forum, he releases source code and helps new users get up to speed with the use of the system... it is far too complex to be of use for anything other than what it was designed to do, develop systems and compare "stacks".
Something along those lines here would be geat... a system that could involve everyone (programmers to help develop it, users to test it with different data and report bugs) some sort of framework that could work with different ideas, different systems... something that could "learn" would be nice but anything as a unified effort would be great... sourceforge is a free site that would host such a project.
I have followed marco's advice and looked up "strange attractors", it would have been better if he explained the theory a bit more, but we run into the dreaded "P" word... Proprietary... I don't think we will ever get to see what makes synch1.xx tick because marco has painted himself into a corner... He charged for the system, now he has a (mostly unhappy) custormer base... he made claims (remember when goose did that?) that were never realized. If he ever did change his mind and decide to open the development, the people who paid for it would be more than a little upset. He could explain the theory without code examples and we all could take it from there, but I don't see that happening either.
I will keep on trying... right now I am working on refining kbcherokee's "bell curve" system... once the bugs are worked out I will release that too (free of course) in excel. I have another system in it's infancy that creates fingerprints of a draw (odd/even, up/down/repeat from last draw, decade, pre-draw column) and that will take a while to finish... but that one will eventually be shared as well... both systems make no claims, they don't pick winning numbers, they just present the same data in different ways... and ONLY by sharing can we gain insight on the interpretation of this data (or any data).
MadDog's PB/MM challenge is always interesting to see, they are making better and better picks all the time (4+1 to date from 15/5 picks of the 55/42 matrix of PB)... would be even more interesting if the contestants explained how they picked the numbers...
dreaming out loud again... mark my words, we will see no real progress until people begin working together... I am sure that soon someone will have used winHunter to nail a 6/49 or bigger jackpot... then that board will be all the rage.
oh well, back to the spreadsheets