One of my favorite techniques is to estimate the average and, critically, "the worst case if everything that can go wrong does go wrong," and plan on the range.

One of my favorite techniques is to estimate the average and, critically, "the worst case if everything that can go wrong does go wrong," and plan on the range. It's an honest approach, albeit more difficult to plan around.