Threaded Mode | Linear Mode

**Frank** · 05-02-2011, 09:18 PM

(05-02-2011 05:22 AM)Peter Wrote: I Frank,
So I've had a crack and come out with this but I'm not sure what I can take from it or even how to interpret it. A little help needed I think lol.

It's based on 6 main balls and no bonus.
See attachment.

I've sort of lumped all 49 balls together. I think that's what you did with your 6.

I didn't do this in excel - don't know where to start. I wrote a quick basic program in qb64 though.

Pete

I'm sorry I got the impression from your charts that they were Excel charts, and mistakenly thought you were using it for your analysis.

Well it took me a while to get my head around what you are showing in your PDF file ( I dislike PDF as a format as you cannot import the data it displays on a web page into a spreadsheet, its read only). Despite the fact that you ignore the bonus ball and also use the convention that 0= direct repeat and hence your results do not mimic mine I do agree with the general trend of your results. They look OK. What I suggest you do is plot them on a chart, that is your percentages on the vertical axis and gap on the horizonta axis to reveal the exponential decay curve. I also took logs of the original 1-49 totals (e.g log to base e of 1224 for a gap of 0, log to base e of 960 for a gap of 1 etc. using your data) and plotted those to reveal a linear graph . For results including bonus ball this told me that the vast majority of gaps are less than 49 hence this sets the max overdue flag at 7 when using overdues to predict a return of a ball in any of the 7 positions. See what you come up with .