Hi ,

​

In 17th-century Europe, black swans were used as a metaphor for impossible events.

Before 1697, Europeans had only ever seen swans with white feathers, so couldn't imagine anything different.

Then Dutch explorer Willem de Vlamingh arrived in Western Australia and became the first European to encounter the unthinkable - a living, breathing black swan.

Since then, "black swan" has taken the meaning of any "high-impact event that is difficult to predict under normal circumstances, but in retrospect appears to have been inevitable," and has been used to describe events ranging from the invention of the internet, to the 2008 Global Financial Crisis and the COVID-19 pandemic.

Yet, unlike the original black swan sighted by de Vlamingh, not all the events currently thought of as "black swans" are entirely unprecedented. The COVID-19 pandemic, for example, wasn't the first pandemic the world has faced and certainly won't be the last.

Many modern-day "black swans" are merely plausible, but highly unlikely events that most people failed to consider because they're outside their lived experience.

Here's the thing...

A major earthquake in the Melbourne CBD is a highly unlikely event that's outside most people's lived experience. Yet, when calculating property insurance premiums, insurers make allowance for it.

If an event is plausible but improbable, allowance can still be made.

Just because an unlikely event isn't represented in your data doesn't mean you should act like an ostrich by burying your head in the sand and treating it as completely impossible.

As a data scientist, dealing with black swan events can be extremely difficult. But any consideration you make of plausible, improbable events is better than none at all.

But what about events that *are* genuinely unimaginable - until they're not? How do you deal with the truly unknown, unknowns? 

Well, that's a topic for another day.

Talk again soon,

Dr Genevieve Hayes.
 

All past emails are available in the archive.