hmmm,,You mean Artificial Neural Networks(A.N.N.£c?Its relied on Error Back Propagation£¬ BP - - - Its a good try.

I am after a 1-2 punch with ANNs...
1. A pattern classification run, which should spot less-than-obvious patterns in the draw history and then
2. A prediction run, to get the most likely next result, only now armed with the patterns it thinks it sees...
so far no commercial demos do the trick, I guess I will eventually have to code one, but the learning curve is steep...

It's actually simpler than that. The ANN learns the past patterns and given that, "votes" for the most likely next draw. Any pick-6 game result (and history) is just a vector of 6 numbers, pretty simple to implement. You won't see any patterns, since human minds are not able to visualize 6-dimensional spaces. So even though there might be a pattern, and the system might know the pattern, we humans will not be able to see it or make sense of it. The only thing we'll see are the outcomes predicted.

With all the hoopla about it, I am starting to feel challenged to make a project of it and do one. I thought of doing it Internet-based, so others could utilize it, but that would probably triple the time it would take me to do it. So it's more likely that it'll be just a client/server program with a database behind it and not much of a user interface. After all, there is not much to show the users. I wish I wasn't so busy in my real life.

I am also writing a program but its a very"weird" theory.Its based on both frequency of past draws and Randomness.I dont believe ANN could work out a jackpot Because the error will always exist, so I quit the idea long time ago.

I quit the ANN idea in 1990, believe it or not. At the time I was predicting the prices of silver, gold and platinum using and ANN. Lotto seemed like a great idea, but I didn't think there would be a pattern, and no pattern, no predictions. And worse, thinking there is a pattern and basing predictions on it is just self-delusion.

I've been trying,It's hard to get this group to open up. Everybody is pretty much locked into their position/approach. And because of the amount of time and effort they've put into their approach, it's almost(on the edge) of a personnal attack to suggest something different. My suggestion is to work the end game. That being, how you generate reports for your final data. Given the posts I've read, this is an area that needs to be developed more thoroughly. Just a thought for a direction to go.

I've been reading the messages on this board for several weeks and haven't run into any worthwhile suggestions on the PB jackpot strategy. I've been at this a long,long time and I'm convinced the proper procedure is to start by eliminating as many 5# combos as possible. For instance, if I select 38 different #s and eliminate all the 5 # combos contained in those 38 #s, I can discard 501,000 combos in one swell foop. Of course, the trick is to have at least one number of the 5 ball winning set not included in the 38#s. If all five of the winning jackpot #s are included in the 38, then all is lost and you just have to wait for the next drawing and try again. I'm very successful at selecting a group of #s that don't include all 5 of the winning jackpot #s. But here's my problem. Say I am convinced that #1 will be one of the winning #s in the next PB drawing. I would eliminate all the combos contained from the #s 2 through 53. Thats 2,598,600 five # combos. Terrific except there are 2,869,685 combos in the 53 PB #s meaning there are about 270,000 combos unaccounted for and when you multiply that by the 42 PB $s, the odds against you are still over 1,000,000 to 1. Putting it another way, you can eliminate 99% of the possible 120,000,000 PB combos and the odds are still 1 million to 1. Better than what you seem to be doing, just taking a chance and depending on luck to overcome the 120,000,000 to 1 odds agsinst you. What to do,what to do? Unless a small group of folks are willing to put a lot of money at risk,...a lot of money... hoping that the odds have been significantly reduced, this all seems meaningless. And there isn't any group crazy enough to risk huge sums of cash against strong odds. Is that the end of it? Not necessarily. I think the proper way to attck it is to go the other way. Instead of a small group putting in large amounts each, assemble a large group to put in very small amounts each. Each isn't really out anything because each was going to play the PB anyway. Still only one member of the group can win but isn't the idea to increase one's chances? This is just a general outline and one of these days, one of these days, I'm sure I'll run into a good organizer to get a group together. As I said, I've given it a lot of thought and this is the only way I think PB can be attacked. If anyone has any thoughts that would change or improve my idea, I'd like to hear it. Incidentally, I'm obsessed with hitting the jackpot and one of these days, one of these days........ Pigskin

Has anyone thought about the MUSL swapping out the White Ball and Red Ball machines after every game or every other game etc? I think I am the only one even thinking this approach but I think in order for the MUSL to keep the game honest, they have to swap out the machines every so many games just to keep the drawings truly random. If you could observe the same machine on every draw, you could truly come up with a pattern. But when the machines are swapped out, you lose the tracking element of the machine. Think of it as counting cards at Poker or Blackjack. The card decks are swapped out after so many games. Besides, these machines have to break down sometime so there has to be spare machines somewhere and they are probably used every once in a while. Hope this gets some people thinking. Thanks.