Agreed. The penalty associated with the risk is important too. Probability theory and statistics can get us the risk fairly easily. But putting a number on the

Agreed. The penalty associated with the risk is important too. Probability theory and statistics can get us the risk fairly easily. But putting a number on the penalty associated with failure is a different problem altogether. If getting sick meant a 1:10 chance of getting a really bad headache for a week, I would take that chance. But if it was a 1:1000 chance of death I would probably not take the chance.