I haven't read the Vogelstein paper yet, but from the abstract it seems that when its talking about 'luck', its talking about errors in DNA replication. The rate of these errors is pretty constant in healthy cells (about 1 per duplication), so its pretty much a matter of luck whether you acquire a mutation in a place that matters or a place that doesn't, and it doesn't really matter who you are or where you are. He seems to be contrasting this to other mechanisms that cause mutations, like chemicals, radiation, or viruses. Sure there is 'luck' involved at the individual level once you've been exposed, but the point is if you reduce exposure across the population, you reduce incidence rates. With DNA replication, there isn't anything you can do to reduce incidence rates.