|Posted: June 8, 2006, 10:52 am - IP Logged|
Another note. Maybe it's just me but I don't discriminate between midday and evening for Duke
GA last night...458
1. Scenario I: Based on Emily's 1, 9 and 17th: 725778224 (i'm assuming these are the correct draws) .. inputting this string would have produced last night's hit in Duke (sniff right-485)
2. Using the method I suiggested would not have a produced a hit with the evening draws history only ..
3. But if the full Georgia history was used (midday and evening) it would have produced the following
Which would change to (eliminating doubles by incrementing digit by 1)
and Duke would bark out a hit 548 in Sniff left --much earlier than the sniff right (hey I'm left brain--I start from the left)
I try both methods and compare them....
1. I use a combination of the methods...
2. The 1, 9 and 17th draws picks seem to work...I suspect many other picks woudl work too..the idea 9 (i am guessing) is to tie the prediction to recent history in a way that does not simply try to guess one number based on one number--that is it is not too linear...
3. Eliminating the doubles helps to widen the pool in an orderly (and "lottery logic") kind of way
4. Compare and contrast using the straight draws with the double-eliminator method...thru back testing and see...
In the meantime I am getting to like that doggy...