All models are wrong; the practical question is how wrong do they have to be to not be useful
– George Box

A common flaw in thinking is to think that ”a map is the territory”, i.e. a model accurately represents the reality.

Maps, however, are always abstractions and simplifications of reality – that’s their function.

Think about it: if the map would be the reality, there would be no need for the map. How inconvenient would it be to have a map with a scale of one mile to one mile? You’d need a truck just to carry the map around.

However, maps are useful and we need maps as guides: there’s no other way to understand the world around us. Even though maps can never fully represent reality, they are accurate enough so that we can use them to observe and alter the reality they aim to represent.

Problems with maps

To use the map, you need to interpret it. This can lead to errors.

Our brains also like to take shortcuts whenever they can (also called cognitive ease). This might lead us to use the wrong model, as we tend to overuse prevalent, well-known models, even if they’re not a good fit.

However, in some cases, a map can be effective even when it does not seem to represent reality. For instance, believing that the cold is caused by evil spirits that fly out of your mouth when you sneeze or cough and thus affect other people, can be an effective map for public health even if some don’t believe in spirits.

How should we use a map?

So the problem isn’t that we are using maps – the problem is that we often forget their limits.

You need to know the limits of your map before you use it.

To use a map correctly, we should keep three things in mind:
1) Reality should update the map.
2) Consider why and by whom the map was created.
3) Maps can (but maybe shouldn’t) influence reality.

Relation to other models

We can also relate the map-terrain relationship to other mental models. For instance, in psychology, it is thought that an individual can’t have absolute knowledge about the reality (i.e. the territory), and instead can only access beliefs of reality (i.e. maps) built over time.

In finance, the map-terrain relationship is often used incorrectly in risk analysis. For instance, financial models often use ”the worst-case scenario”  to calculate risk, and these models base the worst-case scenario in past events (i.e. past recessions). However, just because the biggest drop in DOW has been 12%, doesn’t mean it couldn’t drop 80% tomorrow (which is so many deviations from the average that the model would deem it impossible).

This also ties to Warren Buffets’ investing strategy. He doesn’t use any financial models and instead uses back-up systems and margins of safety to secure his assets from the unexpectable volatility of the market.

Books & articles:


I vlog and write about wacky experiments.

My YouTube channel


Look no further! Here are a few carefully selected blog posts for you.

The Unlazy Letter

Probably the most irregular and wackiest newsletter out there. Give it a shot?

Privacy Preference Center