View Single Post
  #18   Report Post  
Posted to uk.d-i-y,uk.tech.broadcast
Pancho Pancho is offline
external usenet poster
 
Posts: 435
Default OT: Latering thinking puzzle "Why do more peoplre die on theirbithday than any other day?"

On 12/06/2021 18:54, NY wrote:
"Pancho" wrote in message
...

The trick is often a nuance in the way the question is asked. If you
don't spot the trick you may reproduce the question wrongly. In which
case people won't be able to answer.

An example might be that we have one birth day, which is often a
traumatic event, compared to the other 30,000 days in an average life.

So perhaps he was tricking you on the difference between a birthday
anniversary and the actual date of your birth.


By saying that we were "overcomplicating things" by trying to eliminate
the effect of neonatal death, he seemed to imply that what he was saying
related only to anniversaries of the date of birth. I got the impression
(though I never clarified it) that he meant "of the adults in this room
now, there is a higher chance of each person dying on the anniversary of
his birth than on any other day" and it sounded as if it was a
natural-causes effect which was outside the person's control ie not
accidents while drunk at your birthday party, and not suicide on your
birthday or the "desperately staying alive till it's my birthday"
effect. And it was "so very obvious". Either a wind-up or some very
weird factor that even https://en.wikipedia.org/wiki/Birthday_effect
doesn't mention.


As I said the nuance is in the specific words used in the question.
These questions are slight of hand. Often when asked by a poor
questioner, the actual question is mangled, incorrect. Which often
happened when we were ten, or was asked by a none to bright teacher. The
smug comment about "overthinking" was also a standard part of the shtick.

But in this case the obvious gimmick is the actual day you were born
rather than anniversary of that day. Even if risk of death were the same
for every day lived (Poisson distribution), the day of birth would be
the most likely stopping time. The probability of surviving to
subsequent dates is monotonically declining, and hence the risk of dying
on the subsequent date is declining.

So without the exact phraseology of the question, and confidence it was
asked correctly, pursuing the problem any further is a fool's errand.

Even mathematicians sometime chase silly/improbable remarks such as
Fermat's last theorem ("The proof was too large to fit in the margin").