Watching the news gives you an inverted sense of risk.
We fear bad things that we’ve seen on the news because they make a powerful emotional impression. But the things rare enough to be newsworthy are precisely the things we should not fear. Conversely, the risks we should be concerned about are the ones that happen too frequently to make the news.
Moreover, much news are from events which ocurred far from your location. For example, plane crashes from all over the world are often reported on tv. This may lead one to overestimate the actual probability of being involved in a plane crash.
Not really, I should really worry about those with high probability times impact events.
For example, I shouldn’t be worried too much about compile error – that happens almost every hour in my coding, despite it has high probability. This isn’t really cost anything but a few minutes of time.
But I should worry about getting bugs in my code. Depending on how my code is used that could impact a hell lot of stuff. That might be a lesser probability than I made a compile error, but impact much larger.
On the other extreme I wouldn’t worry too much about the whole Internet is down. Despite its impact, the probability that happens is rather vanishing small.
@Andrew Au
Certainly, we should care about the expected loss of the events, not only the probability of ocurrence. Interestingly, this is precisely the definition of a risk function related to a decision rule in statistical decision theory.
It should also be pointed out that there are a handful technical definitions of risk.
For a fixed value of “bad” you can just consider probabilities. :) But yes, expected harm matters too and not just probability of harm.
Hiking in the mountains: I have a higher chance of being killed in an car accident on the way to a hike than by a bear on the hike.
I help teach open-water (ocean) swimming to beginner triathletes. Simply saying the word “shark” is enough to keep some folks out of the water, and news of a shark attack chases some folks completely away.
So I did the math: Compare the risk of dying from swimming in the ocean for an hour (in a triathlon wetsuit) to the risk of driving an hour in your car. Which do you think is more likely? What multiplier would make the numbers match?
To the closest order of magnitude, that number is 1,000. Per hour, you are 1000 times more likely to die from driving than from ocean swimming.
Upon hearing this, many folks gain perspective, understand the meaning of “acceptable risk” (it’s about the driving, not the swimming), and are able to convince themselves to curb their worry.
The rest respond with something like: “But, it’s still POSSIBLE!”
Sigh.
I’ve had this exact thought before! It’d be neat to have a news channel that spent an amount of time on each topic proportional to how concerned we should be about it.
Kahneman and Tversky, of course, have written in great detail about this phenomenon as “availability bias.”
FWIW, Bruce Schneier had this idea years ago:
https://www.schneier.com/essays/archives/2007/05/virginia_tech_lesson.html
A choice quote from the above –
An important caveat is that some small probability events would be so consequential that it makes sense to worry about preparing for them.