If I test a method and it beats Zumma 1600 shoes, I then pass it along to my friend David. If he can beat 100 000
RNG shoes with no coding or logic error, and then later tests it against millions of shoes and still gets a positive result, then that is the holy grail. Naturally we have not been able to do this yet, but one time we *thought we had a winner. It beat 10 million
RNG shoes, but was a programming error. That day really sucked.
Hey Luck, yeah that would be a bummer. This system checker your friend David has, what specifically does the resulting data look like? I'm sure he would be able to program it to give results by the shoe no? Can you give me an example (other than the programming error) of some tests that have failed. Were the methods flat betting/progression? 100,000 shoes is a huge number.
We know that any system will work for certain shoes but bomb on
others. How do you reconcile the variance with that number of shoes. Just playing ten shoes at a session is a huge investment in time. So, what if the system works for the ten shoes then goes south when you quit the session?
Suppose ten shoes of the 100,000 lost forty units each, but the rest of
the shoes showed a three unit average win? What I'm trying to
determine is: what constitutes a losing method? Each system being
different, just trying to understand what criteria is used to determine a
system failure. Thanks T101
Bookmarks