Models for deliberate practice: music, chess, and sports

To structure my practice in more detail, I’m using the deliberate practice models that Geoff Colvin defines in Talent is Overrated. Colvin makes a distinction between practice itself, for which these models apply, and two other related tools for developing expertise: practicing in the work, and deepening knowledge.  I’ll discuss those ideas in future posts.  In the next couple of posts after this one, I’ll outline my actual practice routine and how I apply these models to Chinese.  But for now, I want to describe the models as a reference.

The Music Model

The first practice model is a model of music practice.  In this model, a large problem is broken down into small chunks.  In each of these chunks flaws are analyzed, solutions found, and the chunk perfected on its own before trying the next chunk, and before stitching chunks together. This might be called the top down dynamic programming model of practice.  As an amateur bluegrass musician, I know this is the way to practice, but I often don’t do it, and instead spend time perfecting my flaws.  I remember a conversation I had once with my friend Greg Liszt, one of the most innovative banjo players on earth.  We were talking about how we practiced, and I asked him “when you practice alone, do you jam?”  Greg’s an understanding person with a good poker face for ignorance, but he was clearly astonished at the question. I should have known better: Greg essentially isolated himself in practice while inventing a new way of approaching bluegrass banjo.

This mode of practice is the most straightforward for language learning, and can be applied in many different scenarios.

The Chess Model

The chess model is built on case studies: identify a position, study how it was played historically, choose the move you would make and compare it.  For Chinese, I think I could use this model for translation.  Get a text that’s available in both English and Chinese, and compare my translation from English to Chinese to a professional translation.    I think this model becomes more important the higher one goes with the language.  Higher Chinese often includes historical references, idioms, and lyricism that are marks of sophistication; beginning Chinese is just about not sounding like a complete noob.

The Sports Model

The sports practice model is really two approaches.  One approach is conditioning: developing (or maintaining) the capabilities required to play.  In sports, this might be lifting weights for strength, interval training for speed, or some other capability depending on the sport.  The other approach is to develop critical skills: ball handling skills in soccer, for example.  Playing the game is a lousy way to develop ball handling skills, which probably accounts for why the Dutch train so many great footballers while the Americans don’t: Dutch football training is drills, drills, drills, while American soccer leagues throw kids out on the pitch and let them play.

I think I used the conditioning model by accident.  When I started learning Chinese, a friend of mine told me that the basic grammar was pretty simple, but that the sounds were damn hard for Westerners. There aren’t many phonemes in Mandarin, he said, and to a western ear many of them sound the same at first.  Your ear can’t tell them apart, and your tongue can’t make them distinctly.  So when you start, he advised, ignore the grammar, and train yourself to hear and say the sounds and tones correctly.  On the basis of my friend’s advice I started with the Complete Pimsleur Mandarin series: 90 lessons of audio using spaced repetition and recall to build up chunks of the language.  Gaps in the recording are opportunities for you to say the correct answer to a question or fill in a phrase.  I went through the entire Pimsleur series in about 4 months on the commute to work, wearing headphones and talking to myself on the bus. I wound up with an extremely polite, vaguely northern accented Chinese with a limited vocabulary.  But I could speak.  I didn’t know a single bit of Hanyu Pinyin, the widely used romanization, and I couldn’t recognize a single character.  But I had developed the capabilities, using something very much like the sports model, to hear clearly and say recognizably the sounds that had baffled me at the start.

3 thoughts on “Models for deliberate practice: music, chess, and sports

  1. Interesting read Greg. I knew about Pimsleur audio series and I had started using the German one but it was before reading this and other related books and I stopped in part because I thought I had no talent for it, hehe :). I have been doing deliberate practice with algorithms since a few months ago. I will eventually write a post about it.

    Do you use mnemosyne for spaced repetition? It has worked great for me since college.

    I just published a 1-page summary/cheatsheet for the book that I read daily and that I use to try to get new ideas from: http://slnc.me/deliberate-practice-cheatsheet/

  2. Juan,

    That cheat sheet is quite excellent! Your blog, as well. I’d be very interested in how you are structuring the SRS for algorithms.

    I use Anki, as I mention in my latest post, but mnemosyne looks quite good as well. Have you tried both?

  3. Hey Greg,
    This question may be a little off-topic, Pretty much all of your practice will incorporate a single or a few youth soccer drills, but the solution to results is to know which kinds to use and how to use the soccer drills to get the most out of them. You will have a selection of drills available in the market and you will might need to know which ones are utilised for a variety of coaching factors. But just using hundreds of several soccer drills will not make you a significantly better coach, or essentially aid your crew play more desirable. The Coerver Coaching Strategy is deemed the #1 coaching method for children in the world. I have ordered the videos and implemented them extensively in my coaching.
    Thx.

Leave a Reply

Your email address will not be published. Required fields are marked *