Congrats for keeping it up ;-)
Did you learn much? Did you find a bias?
MJ
(I have post here, can't post further down the thread)
Marmaduke Jinks wrote:
Congrats for keeping it up ;-)
Did you learn much? Did you find a bias?
Frayed knot, nothing remotely approaching significance. It was right/wrong too often to offer any pecuniary advantages.
I've shifted my attention to a tiddly little system that's so simple even Parpacluck might think of it in a couple of lifetimes. It might just be a decent run of good luck or it might be the basis of a competitive WLC
system this year. Fingers crossed.
MJ
(I have post here, can't post further down the thread)
Evil Nigel
"nigel" <useweb@nospam.com> wrote in message news:ueednSvx4oxX9oz4nZ2dnZfqnPSdnZ2d@brightview.co.uk...
Marmaduke Jinks wrote:
Congrats for keeping it up ;-)
Did you learn much? Did you find a bias?
Frayed knot, nothing remotely approaching significance. It was right/wrong >>too often to offer any pecuniary advantages.
I've shifted my attention to a tiddly little system that's so simple even >>Parpacluck might think of it in a couple of lifetimes. It might just be a >>decent run of good luck or it might be the basis of a competitive WLC >>system this year. Fingers crossed.
MJ
(I have post here, can't post further down the thread)
Evil Nigel
Thanks for keeping us entertained :-)
MJ
Congrats for keeping it up ;-)
Did you learn much? Did you find a bias?
MJ
(I have post here, can't post further down the thread)
Marmaduke Jinks wrote:
Congrats for keeping it up ;-)
Did you learn much? Did you find a bias?
MJ
(I have post here, can't post further down the thread)
I've been accumulating (at a very leisurely pace) further analysis.
Comparing the 'best match' predictions with the actual outcomes had a
-9.5% correlation.
Comparing an average of the best 10 predictions with the actual outcomes
had a +11% correlation!
Comparing an average of the 2nd to 9th best predictions with the actual outcomes had a +15.1% correlation.
That's over a six week time period so none of those correlations are significant. In any case, a profitable lottery system would probably
need something like a 90% correlation.
Fun, innit!
Nigel
Sysop: | Keyop |
---|---|
Location: | Huddersfield, West Yorkshire, UK |
Users: | 300 |
Nodes: | 16 (2 / 14) |
Uptime: | 90:00:31 |
Calls: | 6,697 |
Calls today: | 2 |
Files: | 12,232 |
Messages: | 5,348,481 |