|Posted: March 30, 2006, 3:35 pm - IP Logged|
I know that a lot has been said with Pi and some variations of formulae (lest I say this with the debacles that have taken place here the past week or so?), but was wondering if this really had some merit. Probably would be just as good as random numbers, but oh well.
I have been backtesting WEST VIRGINIA for crying out loud this afternoon and came up with some interesting numbers. Of course, as many of you know, I use a progressive betting strategy using $0.25 box bets for my tests. And the drawdowns on the losses were very minimal to say the least.
What I did was this:
I took the previous 10 results, and divided each my Pi (3.14) and then added 666 to one set of 10 numbers, then added 777 to the same set of 10 numbers, and then added 888 to the same set.
Now, that's 3 different sets of 10 numbers and played them throughout 2005 with the progression to see how they did. I used the same set of numbers until there was a hit. In other words, I did NOT refresh my numbers after each drawing, only after a winning drawing. The results are as follows:
((Set of 10 numbers)/Pi) + 666 = $545.00 profit for 2005
((Set of 10 numbers)/Pi) + 777 = $452.50 profit for 2005
((Set of 10 numbers)/Pi) + 888 = $555.00 profit for 2005
Now, if you played all 30 numbers separately, that's a nice $1,552.50 profit for 2005.
I was thinking about combining all 30 numbers and playing it that way, but I got a little tired and just kind of stopped. However, when I did January, it was working well, with several hits in that month.
Another way you could do it is to take each set of 10 numbers, do the calculations AND then take those generated 10 numbers and play them AND their mirrors for a total of 20 numbers played. When I started on one set of numbers earlier, there were even more hits (which is obvious since there are twice as many plays).
Any thoughts on this, or am I just chasing again?
The North Carolina Education Lottery - so much a joke that here are their mascots: