Was reading the FAQ on the California Super Lotto Plus web site, which contains a section “How does the Lottery make sure SuperLotto Plus is completely random?” One part of the answer given is:
At least once a month, each solid rubber ball in all six sets is weighed and measured down to 1/1000 of a gram to ensure consistency in weight and measures.
No wonder random people always win.
Music: Miles Davis :: There’s A Boat That’s Leaving
Interesting.. that still doesn’t preclude the possibility that the fractions of a gram beyond the 3rd decimal point will have an effect on the outcome. It’s a basic principle of chaotic (deterministic) systems (which a tossed/juggled/picked up ball is) that they are highly sensitive to initial conditions.. so even if you measure to ‘n’ decimal places, place ‘n+1’ could cause enough of a difference to fuck up your outcome. In otherwords the Lotto is not giving us true randomness, but a better approximation of randomness.
Point for discussion: Is there *ANYTHING* truly random?
(Let’s exclude quantum phenomena to make it easier.)
I say no (but the measurement errors as above make it seem like randomness exists).
A follow-on… check out the LavaRand server:
http://www.lavarnd.org/
The site’s down now for revision, but basically they used the chaos from a digitized image of a lava lamp to seed a “random” number algorithm, making “true randomness”. Still not truly random, though, since the action of the lava lamp’s bubbles is totally deterministic, but it’s close enough for gub’mint work..
I’d say that if someting exceeds our ability to predict, it’s hard to call it deterministic. The laws of physics determine the outcome sure, but there are too many variables for humans (or our computers) to track, so the bouncing balls may as well be random.
Micro-weight diffs affect probability though, and that’s what you’re getting at. But do they affect the randomness enough? Remember that all balls have an equal chance of being out of weight range so fairness is still there.