Deliberate Practice and Getting Good

I remember being pretty good at the arcade version of Street Fighter II when it first came out. I had a really solid Guile that I played in an unconventionally aggressive style and a pretty good Blanka. I could beat most of the people at the two arcades I went to and thought I had mastered the game. Then, one day this guy I had never seen before came in and totally wrecked me. And he did it with multiple characters as I kept feeding quarters into the machine. He had impossible to escape fireball traps, wakeup dragon punches, impeccable zoning, and more. It wasn’t even close. He was not only better at this game than anyone I had ever encountered, but better than I even thought was possible.

This was well before YouTube, Twitch, or any kind of video on the Internet so I had never had a chance to see how good someone could get at a video game outside of urban legends or people blustering on Usenet. Now, the growing popularity of streaming and e-sports means that we can easily seem what truly godlike, top tier players look like. People who are so quick, so practiced, and who appear to do amazing things so effortlessly. They seem to be as far removed from the rest of us players as professional athletes or concert violinists are from my cat. He can’t even hold a violin properly.

For me, this begs the question: How do you get that good at games?

Looking at the Bobby Fishers (chess guy), Yo Yo Mas (cello dude), and Serena Williamses (tennis lady) of the world, it’s tempting to assume that gaming superstars are gifted with some rare kink in their DNA that sets them apart from other mortals. They just have a natural affinity that gets them most of the way there. 1 But I’m not so sure. I suspect it has a lot to do with how they practice, which we probably don’t fully understand yet.


This is largely because I recently read the book Peak: Secrets from the New Science of Expertise by Anders Ericsson and Robert Pool.2 Ericsson in particular has done a lot of research about how people practice in order to master a skill. Superauthor Malcom Gladwell popularized the work of Ericsson and his collaborators in the book Outliers where Gladwell discussed what he called “The 10,000 hour rule.”3 The idea is that it takes about that many hours of practice to become a master at something –chess, violin, basketball free throws, cs_dust, whatever. Ericsson pushes back against this oversimplification in Peak, saying that the number of hours is far less important than the way in which people practice a skill.

Instead of a using a stopwatch (stopcalendar?) to chart our path to mastery, Ericsson calls for what he calls “deliberate practice.” This isn’t just practicing a skill by doing it over and over again. I mean, it is that, but it’s also more. The authors summarizes the principles of deliberate practice thusly:

  1. It develops skills that other people have already figured out how to do and for which training methods have already been figured out.
  2. It happens outside of the trainee’s confort zone
  3. It relies on well-defined, specific goals and targets very specific things for improvement (for example, in basketball think “making 90% of free throws” vs. “making more shots”)
  4. It requires full focus and attention. No zoning out or practicing on autopilot even if things are repetitive and boring.
  5. It involves getting accurate feedback and deliberately acting on it.
  6. It creates and relies on accurate mental representations of what the trainee is doing
  7. It focuses on improving existing skills in a step-by-step approach. It focuses on incremental improvement.4

The authors go on to describe how this can not only apply to sports, music, and chess, but also other areas of life like work. And, I think, video games.

It would be fascinating to levy this kind of deliberate training program at getting better at games. Not just playing games, but playing them deliberately in very specific ways. It would, of course, be tricky and maybe not a perfect fit to the ideal deliberate practice regimens. Team-based games like League of Legends would be hard to create deliberate practice routines for because they rely so much on cooperation and synergy from teammates. But I could easily imagine a non-team game like Starcraft where players use mods or custom maps to drill through optimal build orders or micromanaging units and casting spells just right. And then watching replays (which are conveniently recorded by the game) to refine your mental model of what’s happening in the game and extracting feedback on your actions.


One of the other things Ericsson and Poole emphasize is having access to a really good coach that’s appropriate to your current level of skill. Beginners can benefit from any kind of coaching, but as you progress you need someone who knows more and can customize training activities to correct more subtle mistakes. Like Olympic athletes who retire to coach a new generation, I wonder if there is a market for aging e-sports stars to take up coaching gigs. Or to charge by the hour in order to review recordings of matches and assign deliberate practice drills.

I mean, forget e-sports. That’s already old news. E-coaches are where it’s probably going to be at.


1. The rest is presumably having the right gaming mouse and snacks.
2. Ericsson, A., & Pool, R. (2016). Peak: Secrets From the New Science of Expertise. New York: Houghton Mifflin Harcourt.
3. Gladwell, M. (2011). Outliers: The Story of Success. New York. Back Bay Books.
4. Ericsson & Pool, pp 99-100

7 thoughts on “Deliberate Practice and Getting Good

  1. Interesting article. I don’t participate in E-Sports but I am a speedrunner and I have noticed that I get much better much faster at the games I run when I put more focus on practice. I approach it like a marathoner approaches running. A marathoner doesn’t go out and run 26.2 miles everyday until they get good at it. They have a training regimen that is varied and focuses on improving individual aspects that will make them good.
    When I started speedrunning I basically spent the entire time actually doing full runs of the game. This is basically the equivalent of throwing yourself into a wall a million times. Eventually you will break through but it’s gonna be a pain getting there. I see a lot of speedrunners doing this because “practice is boring”.
    Now, a lot of my focus is on practice. This partially started because I didn’t have enough time to stream attempts on Twitch. Some of my time is spent doing a hard trick over and over to have a better understanding of the trick. Most of my practice is spent breaking the run into smaller chunks. That way I can experiment with different strategies and do tricks that may be outside my comfort zone when compared to a full run. The thing with speedruns is that since a lot of people reset before finishing a run they tend to see the beginning a lot more than the end. By breaking the game down like this I have allowed myself to be comfortable with the game from start to finish.

  2. Your bullet points 1-7 appear to describe what happens in certain martial arts traditions. The kata exist not because they are secret killer technique, but because they train specific skills and have the required clear feedback points. Not always, though.

    A lot of what you’re saying is also discussed in a book I like very much called “Talent Is Overrated” by Geoff Colvin. Highly recommended. He points out that all the hours in the world will not lead to any improvement at all if the student is not specifically focused on learning something during those hours.

  3. Very interesting article, I really like how Ericsson described the effective usefulness of deliberate practice. There is quite a lot in common with Csikszentmihalyi’s theory of Flow, which is also something I have experienced many times during practicing playing music, writing essays or well, playing video games. I do agree that if the majority of people would not get discouraged by early failure when trying new things, and were able to focus a bit more we would have an overall much higher skill level in the general public.
    Also, and I want to make this its own point in this comment, THANK YOU for making sure that the 10’000 hour rule gets the proper citation as Ericsson intended, not as Gladwell popularized it as a yardstick. Too many do not make the distinction and sell the 10’000 hours at face value to others.

  4. Great article! I can see how deliberate practice helps in a number of fields from languages to music. I have no doubt it can also help e-athletes get to the top of their game (pun very much intended). I do wonder if this is the only area of games where it can be applied. I’m a game designer, and I’ve always felt that game design is a field that does not benefit from the same structure as others. Maybe its because we’re too young and new to have a solid body of knowledge or maybe because there isn’t enough experts to coach new generations to go around, but we do lack the the tools and methodologies to measure and promote our own skill. We don’t even have a consistent series of exercises that we can repeatedly use to practice!

    Don’t get me wrong, game designers do practice quite a bit. Analyzing new games, participating in game jams and working on design challenges, helps hone useful skills. However, none of these are objectionably measurable and as such cannot be used for deliberate practice. Some say that subjectivity is inherent in game design, and as such it can never be practiced deliberately; but I’m not completely sold on this. Music, dance and painting all have varying degrees of subjectivity, and yet they can be practiced. I’m very curious to know your thoughts this.

  5. All the things you mention have been very common in Starcraft since the Brood War days. People have always used custom maps as a way to drill specific game skills like build orders, unit micro, etc. As for coaching, this has always been common too! Lots of high-level players offer coaching, at rates that are comparable to, e.g. piano lessons. I can’t speak to other e-sports but Starcraft is way ahead of you here 🙂

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.