Observations vs. Models

Note to future self…

Good arguments (“sound bites”) tend to be very persuasive. Today I thought about how people’s predictive arguments about anything can be divided into two groups:

Observations: is a statement that explains a phenomenon.
Example: Electric cars are selling a lot. This is the future!

Models: is a thesis that explains a phenomenon and intrinsically makes a prediction about the future.
Example: The rising cost of residential real estate will cause people to move to cheaper cities.

The underlying idea behind a model is this notion of cause → effect. That’s the intrinsic mechanism that drives the a model from explanation to prediction.

Why does this matter?

Some observations will disguise themselves as models. We all have this impulse to create cause-and-effect relationships so sometimes make we make observations that suggest that something will happen in the future based on a certain observation of the present. Typically you can be right about this but not always.

The problem with these observations is that it makes it hard to pin-point the variable or set of variables that will drive this effect. Does this variable have


All of this is very basic and seemingly trivial. However, until now when I read people’s comments online or a book I don’t tend to stop and think:

Is this person making an observation or explaining their model of reality?

This will automatically generate follow-up questions (e.g. what’s the variable(s) driving the model, can I prove it?, how does it compare to what I already know about the world, etc.). I will give it a try and report back any updates.

That’s it. Now back to work…