It is a hypothetical that one couldn't give a hard number to. I would say the 7.5% figure given earlier is impressive. If we think of 80% of people having the problem, and fewer than 10% of them actually manifesting it within prison, well, that's progress.
To get to zero, you would probably need to choke off a vast amount of access to community. So probably the cost of getting to zero would include other aspects of things you would want to do, like maintaining relationships with key people in the community, having exposure through rehabilitative opportunities, and preparing people for reintegration in the community. That means at some point in the process you'd have to ease off on the restrictions and allow people to make choices for themselves, which they would do well in some cases and badly in others. If you don't let people have a chance to make a mistake, they will never have the chance to test out their ability to resist making mistakes. Any system of treatment and rehabilitation has to have a graded re-exposure so people have to manage the learning they've achieved. That can mean some access to the community. That means some re-exposure to risk. Achieving zero percent drug and alcohol use within prison is theoretically possible if it is hermetically sealed, but in terms of dealing with the problems of reintegration of people back into the community, you'll be losing at that end if you try to aim that high.
Would it be fantastic to take all the things you've talked about--the standover tactics, the manipulation, the violence that comes from people fighting over access to drugs, and the things that people will do to get drugs--out of the equation within prisons so that people would have the opportunity to direct and find better ways of dealing with large problems? You bet. I don't think that's measurable, however. But there is that balance between needing eventual community reintegration to be a reality and trying to get the amount of drugs in prison down to as low as possible.