Probabilities and distributions are hard to grasp. We usually avoid the problem by settling on mean values and hope they are somehow representative. But every stat textbook warns students that one must at least look at variances also; large variances undermine the usefulness of means and medians.
Longevity and morbidity data are of particular interest for all the obvious reasons. The good news is that the trends are very favorable if you enjoy living. Visit any old cemetery and note the life spans recorded on the old tombstones and markers. And as average life expectancies increase, so must the associated variances.
I mention all this because I had trouble digesting Dr. Ezekiel Emanuel's "Why I Hope to Die at 75." The man is a scientist and an ethicist. He knows that every statement (generalization) in his essay must to be stated in terms of distributions. He even includes one in the essay.
Even the relevant distributions are conditional; the one that applies to any one individual is contingent on the years already reached -- as well as the state of health, etc.
Whether it is Immanuel Kant or Daniel Gilbert, many wise people have had something to say about our ability to know our state of mind in the future. There is not an epidemic of suicides among the old an infirm because most cling to life and hope for another day. I hope that when Dr. Emanuel reaches 75 that he is able to say "not so bad after all."