Posted by KM Klitser Mahlmberg on March 29, 2001.
In Reply to: Re: A possible killer lotto strategy? Open debate posted by BigJer on March 28, 2001. See also the 2005 update at the end of this material.
To all:
I once ran the deltas in Ion's MDIEditor and got some interesting results in the any/ver filter tables. One would have to do it all manually, but perhaps it would be instructive or at least interesting to do the same with those values. There might be statistical tendencies in the skips that might be generalizable enough to allow one to eliminate certain wheeled combinations.
KM
: Ion and fellow lotto players
: I have been looking at not the numbers but the skips
: of the numbers that hit and the sum of those skips which has a good bell shaped curve. I have not been able to project that as a tool as yet . Any help here?
: BigJer
: : Ion:
: This may not be particularly what you're looking for, but it has been helpful in those instances when I've had far too many potential wheeled lotto combinations.
: : I freely admit that this strategy has its faults. However, since I cannot afford to play thousands of combinations, I had to come up with a criteria that at least decisive to cull down the herd, so to speak. It is as follows:
: : Utilizing your FFG I get a pool of potential numbers and wheel them with one of your wheels that automatically eliminates unlikely pairs, all odds, all evens and etc..
: : Unfortunately that still leaves thousands of numbers that I cannot afford to play.
: : Basically what I do is copy and paste the Mdieditor file into SPSS stats analysis software and filter out combinations from the original pool that do not meet the following criteria:
: : 1. Do wheeled combination numbers in each position conform to historical positions based on entire history of lottery draw? If yes, then retain. Does the difference between the last and first lotto number, spread of line, conform to historical 80% statistical trend? If yes retain.
: : 3. Does the number of even and odd numbers in the wheeled lotto combination match the statistical history of what combo. of even/odd happened in similar past drawings? If so retain.
: : 4. Does the number of high and low numbers in the wheeled combination match the statistical history of hi/lo that happned in past similar lotto drawings? If so retain.
: : 5. Does the sum range of a wheeled combination fall within an expected range that occured at a statistically high degree in past similar drawings? If so retain.
: : Thus filtering the original pool of numbers through each step, one hopefully will retain a group of lottery combinations that have characteritics VERY congruent with winning lottery combinations of the past based on statistical likelihoods in similar draw situations.
: : Granted this is far from an ideal wheel-filtering strategy, but when faced with literally thousands of possible combinations one has to risk being wrong for the sake of being decisive. I can live with that.
: : One final note. Generic neural nets overfit the data with sample sizes that are too small and have attrocious Rsquared and average error values when sample sizes are larger. Even if the numbers seem good, neural nets generalize horribly with lotto data. After all, they were never meant for pseudo random numbers anyway.
• Update 2005
Klitser et al:
I love to give people names I believe are suited for book characters. Cyber space is a book; we only deal with characters, we see nobody. I hope you don't mind; therefore feel free to address me as Jaqk Fowuru (a.k.a. Parpaluck).
• The greatest random number/combination generator for lottery ever — IonSaliuGenerator v7.0 — makes extraordinarily good usage of deltas and also standard deviation. Run it in all confidence. See the link in the footer of this page (Odds Calculator & Lottery Number Generator).
• Deltas are considered, although in a limited form, across MDIEditor and Lotto WE v1.3.
• I don't hide it. I write lottery software especially for myself; call it in–house software. It is very natural to keep a number of things as your trade secrets. My in–house software makes extensive usage of deltas in multiple forms. Also, the software applies the three fundamental forms of dispersion (deviation, fluctuation). They are: standard deviation; the average of the deviations disregarding the sign; average deltas (absolute differences).
There is a multitude of considerations for not releasing each and every piece of lotto software. Writing the documentation is a very difficult task. Then, things change constantly. It is very hard keeping up with updating the software and rewriting the documentation. Besides, a share of selfishness has always been a requirement of life. Don't you think?
Resources in Lotto, Lottery Software, Wheeling.
See a comprehensive directory of the pages and materials on the subject of lottery, lotto, software, systems, and wheels. The link is displayed in a new window. If the page content is of interest to you, copy-and-paste to that lottery page of yours: SaliuLottery.txt or something similar. Copy as much as you can in that text file. By editing that page in your own words will result in a very effective lottery manual.
Remember to check the free software downloads site:
Socrates Home 
Search Site 
Help on Lottery Software 
Random Generator, Odds Calculator 
New Writings, Pages 
Sitemap, Content