This is a much better explanation of the problem with estimates than my analogy :) Especially this part: Professionals have found it's often easier to refuse

This is a much better explanation of the problem with estimates than my analogy :) Especially this part: Professionals have found it's often easier to refuse to provide estimates than to fight over high-confidence estimates or educate stakeholders in interpreting probabilistic date ranges. Not estimating saves lots of time, too.