[quote=lottoboy]
[quote=stan]
i've just responded to your email.
a summary for other forum members:
* i think your reasoning is wrong and there's nothing to fix in the plugin or in the application itself.
* the count of past draws accepted by a filter is proportional to number of full-package tickets accepted by the same filter only when the filter is applied on a large enough number of past draws. you can't draw any conclusions from back-testing 100 latest draws and comparing it with results of filtering of 500000 package tickets.
[/quote]
Yes , I have received your e-mail and read it carefully. Thanks!
My answer about your reply is as below:
1) In fact, you have not yet understood my idea really since you have refused to do some backtests for that even though these backtests are time consuming. “
Practice is the sole criterion for testing truth.â€
Especially, you have not yet understood the Filter of Number Order totally as you said in your post at 01/05/2010 that:â€
i don't have filter's source code and since i didn't follow this thread i don't even know how that filter works“.
[/quote]
i have read this thread and downloaded joe's filter plugin. i think i know how the filter works now.
2)Yes, my selected tickets for Back-tests are 100 because the results are similar to full winning tickets (more than 2530 tickets for my game). It’s impossible absolutely to filter more than 500000 Winnig tickets like you said since no any game in the world has had about 500000 History Winning Numbers until now!!!
Mr. Stan, don’t forget please : your EL program just is to predict
next result by using some limit history winning numbers (100-200) instead 500000 or more.
sure, but you can statistically correlate results of filtering of 500000 with the results of filtering of 100 draws only
3) I still think we can discover some strong strategies if you and Joe can improve your code to control the relationships between Preview & BT of Winning Orders in Number Order Filter instead of the current relationships between Pre. & BT of Winning Orders are in anarchy as I said in my e-mail and posts. I do hope you can rethink my idea and do some backtests for that. or I can give you more test data I have done if you want.
your back-test data are fine. there's just problem with your interpretation of back-test results.