Your OWN statement indicates this is true for HALF. Pointing out that it might not be true for the other half is just an effort to dodge the salient point.
What percentage of our population do you regard as an acceptable loss?
No, "half of the likelihood" does not translate to "HALF of our population."
Whatever it translates too, it represents a significant number of casualties. What percentage of our population do you regard as an acceptable loss?
There is no direct relationship between percentage of likelihood accounted for by genes, and the range of values for the likelihood itself. It's entirely possible for one subgroup to have a 25% likelihood of addiction, a different subgroup (with different nongenetic factors but the same proportion of genetically susceptible people) to have a 1% likelihood of addiction, and for genetic factors to account for about half of the likelihood of addiction. (Details of calculation available on request.)