Someone please tell me what I'm doing wrong. I wrote a program to go through the past history of 2 of our jackpot lotteries here in ON. I wanted to find out just how often there is a match on 5, 4, or 3 numbers (for a 6-of game).
For 6/45: there have been 1446 draws.
42.6% of the time, a draw had only matched ONE number in a past draw.
27.8% of the time, a draw had not matched any numbers in a past draw.
23.3% of the time, a draw had only matched TWO numbers in a past draw.
That right there accounts for 93.6% of the time. So the majority of the time, a draw did not have 3 or more matching numbers from previous draws. (So I'm wondering why use a past draw to choose a new draw)?
Anyway, only 4.56% of the time, a draw did match 3 numbers in a previous draw.
And less than 1% of the time, a draw matched 4 or 5 or 6 numbers in a previous draw. So it doesn't look like playing previous draws offers any advantage.
The numbers were similar for ON 6/49 (which has had 2300 draws):
42.7% of the time, a draw only matched ONE number in a past draw.
31.4% of the time, a draw did not match any numbers in a past draw.
20.8% of the time, a draw matched only TWO numbers in a past draw.
This accounts for 94.9% of all the draws for ON 6/49.
So I tried to slice it another way. In ON 6/49, there was only 1 draw that matched 5 numbers in 4 previous draws. There were only 6 draws that matched 5 numbers in 3 previous draws. Every other draw that matched 5 numbers only did so twice or less. So by theory, I should be playing these 7 lines (1+6) for 40 weeks (along with the highest ones that hit 4 numbers and 3 numbers). However, in the entire history of the game, these 7 lines have only matched 5, 4, or 3 numbers a total of 637 times. Which sounds like a lot, but in 2,643,850 passes, that is less than 1% of the time. ((2299+1)*2300 draws/2).
And it looks interesting when you see that 12,233 times a current draw matched a previous draw on 4 numbers. But when you consider the 2,643,850 comparisons it took to get to that number, it still only amounts to 0.46% of the draws.
Finally, yet another way to look at it:
If I were to play the 10 "best performing" strings in ON 6/49, they would be:
Draw# H5 H4 H3
2293 1 10 131
2284 1 10 120
2137 2 10 115
2268 2 13 115
2275 1 15 114
2144 1 8 113
2125 1 14 113
2074 1 10 113
2279 2 13 110
1930 2 6 109
However, these strings have hit 5, 4, or 3 numbers a total of 1273 times, which only works out to .048% of the time. So based on history, these strings should not be expected to match on 5, 4, or 3 numbers 99% of the time.
The numbers simply don't seem to add up for me, when you look at the entire history of the game and calculate how many times there was not a match on 5, or 4, or 3 numbers.
I'm wondering if everyone's success with this strategy is due to the strategy itself, or due to the other filters that have been used along with it (i.e. high/low, odd/even, sums, etc)?
If someone can shed some light on this for me, please do. Thanks!