A quick search of the inestimable comp.risks archives revealed this: http://catless.ncl.ac.uk/Risks/15/80#subj2 http://catless.ncl.ac.uk/Risks/15/80#subj2 Mon

A quick search of the inestimable comp.risks archives revealed this: http://catless.ncl.ac.uk/Risks/15/80#subj2 http://catless.ncl.ac.uk/Risks/15/80#subj2 Montreal -- Daniel Corriveau said he hopes that his 'victory over the system will give hope to others.' The computer analyst and his family received more than $620,000 [1C$ = U$0.75], including interest, from the Montreal casino yesterday, weeks after they overcame odds of one in six billion and beat an electronic keno game three times in a row." The author explains the following key points: o Corriveau used an "antique 286" computer to analyse 7,000 combinations from the keno game, [which uses an electronic pseudo-random number generator]. o Corriveau noticed that the electronic game was repeating numbers in a predictable pattern. o Corriveau and several family members bet on what they predicted would be due to come up; they won three times in succession. I had originally seen an article speculating someone power cycled a keno machine after recording the winning numbers, with the assumption or knowledge that the random number generator reseeded with a 0 on cold boot. I'm not sure if that's just me mis-remembering the details of this case, or another one altogether. If anyone has a link to the second case please let me know.