Having been involved in a couple of lawsuits as an expert probability witness and having observed that a prudent skepticism is often less prized than… - John Allen Paulos

I’ve argued that the set of standard questions journalists ask and readers want answers should be enlarged. Besides Who, What, Where, When, Why, and How, it should include How many? How likely? What fraction? How does the quantity compare with other quantities? What is its rate of growth, and how does that compare? What about the self-referential aspects of the story? Is there an appropriate degree of complexity in it? Are we looking at the right categories and relations? How much of the story is independent of its reporting? Are we especially vulnerable to the availability error or to anchoring effects?
If statistics are presented, how were they obtained? How confident can we be of them? Were they derived from a random sample or from a collection of anecdotes? Does the correlation suggest a causal relationship, or is it merely a coincidence? And do we understand how the people and various pieces of an organization reported upon are connected? What is known about the dynamics of the whole system? Are they stable or do they seem sensitive to tiny perturbations? Are there other ways to tally any figures presented? Do such figures measure what they purport to measure? Is the precision recounted meaningful?

The moral, again, is that some unlikely event is likely to occur, whereas it’s much less likely that a particular one will...The paradoxical conclusion is that it would be very unlikely for unlikely events not to occur. If you don’t specify a predicted event precisely, there are an indeterminate number of ways for an event of that general kind to take place.

Always be smart; seldom be certain.