A few years ago, I spent about five months trying to improve at chess. In that time, I learned a lot about chess. More importantly, I learned a lot about how to systematically become more effective at something I care about.

In that time, my rating improved from about 1100 (the ~50th percentile of players on chess.com) to about 1400 (the ~75th percentile of players on chess.com). While neither is a particularly impressive rating, I’m proud of the improvement and consider a 25 percentile improvement significant.

This post is the first in a short series where I’m going to explore what I learned in this time: not about chess, but about what expertise is and how to gain it.

Today, I’ll lay the groundwork for this series by answering two questions: what is expertise, and what makes someone an expert?

I hope this series is useful for anyone that wants to be better at something than they are or has found that their improvement in a domain has stagnated.

What is expertise?

My definition of expertise draws heavily from the “Learning How to Learn” online course: expertise is a set of mental models that allow you to effectively perform a task.

You can think of a mental model as just a way of “compressing incoming information” when looking at problems. By compressing the information, your brain can “summarize” what it’s seeing and act on that summarized version of the information instead of the “raw data”. Later, you can use that extra brainpower to apply additional layers of mental models, further increasing your efficacy at the task.

For example, rather than remembering the speed limit on every single stretch of road, most people (in Michigan) know:

• If you’re on a highway with a median, the speed limit is probably 70
• If you’re on a highway without a median, the speed limit is probably 55
• If you’re on a regular thoroughfare, the speed limit is probably 45
• If you’re in a neighborhood, the speed limit is 25

In math, mental models tell you how to answer “What is (x² - 1) times (y³ + 7)?”. To the unfamiliar, the problem might look intimidating. Anyone that’s taken algebra will immediately recognize that we’re just multiplying two polynomials together and the problem can be solved using FOIL.

In both cases, mental models allow you to identify the pattern associated with the incoming information (“I’m on a highway without a median”, “I’m multiplying two polynomials”) which then map to problem solving techniques associated with this pattern (“the speed limit is 55”, “I need to use FOIL”).

What makes someone an expert?

Experts have a set of mental models that allow them to be unusually effective at a given task.

As an example, the follow set of mental models might correspond to each chess rating:

• 600: You know how the pieces move
• 800: All of the above, plus you recognize some common openings and have learned the first few moves of very common openings
• 1000: All of the above, plus you know of and have practiced a few basic tactics (pinning, forking)
• 1200: All of the above, plus you have a basic understanding of how to protect your king and what pawn structures you should try to avoid (backwards pawns)
• 1400: All of the above, but your opening knowledge is deeper (generally can make it 7-8 moves into a game without getting lost)

It’s not that players that have a rating of 1400 are any smarter than players with a rating of 600. They’ve just built more effective mental models that allow them to better approximate how to perform the task “perfectly”.

How do I become an expert?

While I don’t pretend to have all of the answers, the time I spent improving at chess taught me some useful lessons. Competitive games are great for understanding the “meta” of how to gain expertise because there are lots of people striving to do so and discussing their strategies. Unlike in many other domains in “real” life (e.g. driving), competitive games usually give clear feedback about who’s better at the task.

I’ll share more about what I learned in this series: if you want to follow along, feel free to subscribe below.