Remember the Anselm challenge?

[ragnorak]

(03-11-2016 10:49 PM)Frank Wrote:  The problem for only 12 balls on a 659 matrix lottery is that its a double whammy against you for the odds. When we did the Anselm challenge with 16 balls the theoretical 'norm' was for your average score to be 1.959 based on one point per ball hit. Now that the odds are tripled and the ball numbers in your entry are reduced, the average score for your challenge is 1.220.
I think your £100 is pretty safe given the 37% increase in difficulty, according to my calculations. I'm not sure I'd want to play with so few balls. I say this after just completing a challenge in the Google RGL group. In that challenge they had 24 balls, you had to have them in priority order, and I came second bottom out of about a dozen players. If truth be told there is not enough history to the 659 lottery for my system to be any use, and I'd be surprised to hear that anyone elses system can truly work after only 44 draws.

If we've learned one thing from the lottery, it's that no matter how high the odds, if enough people are playing for long enough then someone will win. It's a fair point though that on a smaller scale with a dozen or so people playing, matching 6 out of 12 could take several years as the odds are 48,764 to 1. When designing this game I was trying to make the prize structure nice and simple while maintaining just a small house edge, and of course I didn't want my £100 to be too easy to win by chance! So 12 numbers struck me as the right balance. However I think there's room to make it a bit more interesting so I added an extra prize for matching 5 plus the bonus which includes £75 cash. Already one member has matched 4 numbers after just a few weeks of the game, so I think it's doable!

I'm not so sure that adding 10 balls to the lottery requires a complete reset of statistics, it just means you have to do a bit of work to adjust for the differences. For example if a pair of numbers hasn't appeared for a few years, then for all the draws up to 2065 we can count it as 1 time overdue for every 78.4 draws, and for draws 2066 onwards it's an additional 1 time overdue for every 114.1 draws. Add those together and you've got an item that incorporates both 6/49 and 6/59 results. It may be a little more complicated depending on the type of statistics you're doing of course.