reprint for translation
Hi Jack
Let me see if I can get Google to translate this better for you. Since quite a few words had spaces removed in the original post (My Error) Google may have become confused during the translation! I've also removed the short form CSN. changed it to "Sorted Lexicographic Numbers" This might help also.
“ Hi Icewynd
That’s fine I do the same thing, update draws, look for patterns to try and make educated guesses .
My latest method is to graph the Sorted Lexicographic Numbers of Historical draw’s, look for a trend (there usually is) and then correlate that, to the 1 to 49draw frequency, paying particular attention to the first number to be predicted in the next draw. Only so many numbers occupy position one and the past frequency of those numbers that occupy position one is what I start with ( BTW ,I’ve read your thread on “eliminating 18 numbers” and find that extremely interesting! ). Anyway, my next step is to take that first number find how many Sorted Lexicographic Numbers starting with it then try to break that down by correlating it once again with overall 1-49 number frequency’s, (researching to find an efficient algorithm to do this, (I’m leaning towards “Goodness of fit” method used to check how well data fits on a trend line graph) but, right now I just use best guess number trend analysis, lol )
I fixate on Sorted Lexicographic Numbers because I feel that if I can get that one variable correct, it will get me extremely close to the major win, if not spot-on. Unfortunately at the moment, I have to do a lot of manual figuring and I’m in the process of retooling my excel spreadsheet(s) to reflect my learning curve (i.e. ever type of statistical variable known to man! Cause I only want to do this once more!
Best I’ve done is 4 correct predictions out of the last 69 attempts.
Since I started to do a lot of reading or I should say listening, (using a text reader, if you don’t have one, get one! It makes going through mountains of PDF’s and web pages a breeze . Best software I think I ever bought, once you shake out the Bugs!)
I’m finding that there is a plethora of methods…
Bayesian Analysis, Poisson Method, Markov Chain/Monte Carlo Method(s), Time Series Analysis Theory, Recurrent Event and Renewal Theory and many, many more.
I’ve learned that the best approach to prediction, is to develop a very strong understanding of all the possible variables, all the possible methods of measure of those variables, how to interpret what type of data we are dealing with, as to the type of method that will be employed to achieve the best predictions.
One benefit out of all research is that I’m starting to understand how to read Probability and Statistical Formula Notation’s. It was just all Greek to me before!
As you have said “it is darn hard to defeat randomness”
Keep sane!
All The Best.
Alex. “
Hope this clarified it better