ScotthYoung.com Members Area

Day 4: Why the 10,000 hour rule is broken (and how to learn skills faster)

Today I’m going to tell you how to learn skills more quickly. In particular, I’m going to attack a common misconception people have about the so-called “10,000 hour rule”, popularized by Malcolm Gladwell in his book Outliers.

—-

So what’s wrong with 10,000 hours?

In Gladwell’s book, he used the research of Anders Ericsson to observe a trend. Almost all of the world-class experts in fields ranging from athletics to entrepreneurship acquired over 10,000 hours of deliberate practice before their big successes.

The misconception people fall into is believing that, because world-class experts tended to have 10,000 hours under their belt, that becoming really good at something is mostly a matter of putting in the time. Put in your 10,000 hours and you’ll become great. Right?

It turns out that Ericsson’s research says no such thing. In fact, he argues the opposite. Many people will stall at “plateaus” in their skill development that can persist forever. Far from an inevitability, it seems that most practice efforts snag on these plateaus, so the experts who break through them are the exception, not the rule.

—-

Stagnation, not growth, is the default. Hitting plateaus and barely improving your skill level over years is what happens most of the time. Experts are world class, not simply because of their dedication to thousands of hours of practice, but also because they climbed past the plateaus on the way.

There are a couple reasons your skill growth can plateau, and most are insidious in that they sneak up on you.

The first is simply comfort. In the beginning, a skill is hard, frustrating and performance is difficult. Once an adequate level has been achieved, however, habits start to form and the drive to get better goes down.

A second is that practice is not separated from performance. Performance is the concert, practice is the drills to get the chord sequences perfect. In music and athletics the division between performance and practice is well known and understood. But it other domains this important distinction is ignored.

Consider programming as a job. You might write code 8 hours a day, but that doesn’t mean you’re getting 8 hours of practice. Even when you are productive, the time you spend is mostly solving problems you can easily fix. Perhaps only a small percentage of your day is training new abilities from your work.

A final cause of plateaus is lack of timely feedback. If your work doesn’t give you immediate feedback on what you’re doing well and what you’re doing poorly, you may spend a lot of time practicing but still not improve greatly.

Given all these obstacles, it’s not surprising that most attempts at mastery stall at adequacy. Getting to a truly high level of skill attainment requires successfully dodging these traps along the way.

—-

Ericsson found that deliberate practice was the key to avoiding plateaus. Deliberate practice isn’t just any kind of practice, however. It has to have the following features to count:

1. Method, not results. This is the practice/performance distinction. Your practice efforts need to focus on correct execution, not merely whether they get the job done.
2. Targeted, not broad. Practice sessions should improve specific aspects of your performance, not just try to improve it broadly.
3. Guided feedback. Coaching is best, but measured feedback can also work to adjust your progress.

The problem I hear from people is that this idea sounds great, but it’s hard to apply to skills outside of music or athletics. How do you use this to become a better programmer, writer, architect or entrepreneur?

—-

The Breakdown Method

The breakdown method is a way to achieve this. Formed from conversations with Cal Newport, many of the details of these ideas are his.

The basic idea is simple: you break a skill down into parts, figure out how to measure those parts, then set aside practice sessions to build those parts. This achieves most of the ideals of Ericsson’s deliberate practice, but it can be used on any complex skill without needing personalized coaching.

Step One: Break a Skill Into Its Parts

The first step is to decompose your skill into practiceable parts. What I mean by ‘practiceable’ is that the subskill is narrow enough that if you focused solely on improving that in your practice session you could do it rather straightforwardly.

Consider programming. There are thousands of microskills that programming can be broken down into. Syntactical fluency with a languages’ core features. Understanding particular design patterns. Code organization and modularity. Algorithmic efficiency.

Take writing as a different example: headlines, brevity, vocabulary, research, argument structure, tone, humor, grammar and spelling. All of these parts are components of what makes a writer good–different writers have mastered different aspects of the skill and the interplay of different factors gives a writer her distinctive voice.

For the skill you want to learn, break it down into smaller skills until the small skill is something you could devise a way to practice directly.

Step Two: Measure Your Ability on Subskills

Once you’ve gotten a list of skills, you need to find a way to measure your ability. Sometimes this can be assessed directly, such as milliseconds to run a test code benchmarked against a competitor. In these cases, your job is easy.

Other times, you may want to use a proxy. Headline writing isn’t the only factor that influences email open rates, but it explains a lot of the variance in open rates for emails within a list, given a consistent writing schedule. Here, you might use open rates as a proxy measurement for your ability.

If neither of those work, a benchmarking approach may also work. Here, pick people who are in your skill range, but are somewhat more proficient than you are at this particular subskill. Now work at the subskill until you can honestly tell yourself you’re meeting their quality levels.

Step Three: Set Aside Specific Practice Sessions

Next set aside a specific practice session to improve a skill you’ve found, measured and decided was important. If the subskill is small, you may only take a couple sessions to improve markedly. If the subskill is hard, you may take a couple months.

If you’re not sure how much work you need, I would recommend spending one hour practicing it. If you find the exercise fruitful, try doing a 30-day trial to work on that subskill for at least 20 minutes a day.

If you didn’t find the exercise useful, go back and reexamine your subskills. Perhaps they are poorly defined? Perhaps you need to establish a different measuring criteria?

Another possible hiccup could simply be that you don’t know more effective techniques for improving that subskill. In this case, I suggest reading a book or interviewing someone who does have that expertise. Often watching and reading other examples can highlight inefficiencies in your method that can be weeded out with successive practice.

Example: Learning French

Here’s how I would apply this method to a skill I’d like to get better at, speaking French.

First, I’d decompose the skill:

• flexibility with different phrase patterns
• pronunciation and accent
• vocabulary
• prosody and intonation
• listening comprehension
• reading comprehension
• writing
• etc.

Many of these are still too broad to be ideally practiceable, so I might narrow them down further. Vocabulary could be subdivided into slang, expressions, technical terms, cultural terms, synonyms for common words. Pronunciation could be broken down into specific vowel sounds I have difficulty with.

Next I’d find a way to measure this. If it were vocabulary, I could use Anki flashcards to assess performance. If it were accent, I could have French speakers compare my pronunciation to natives and point out words and phrases they thought stood out.

Finally I’d implement a practice schedule. Reading comprehension could be as simple as picking up a book that’s just above my current level. Writing could mean writing essays on randomly assigned topics and getting evaluated using a service like iTalki.com.

—-

Day Four Homework

Unlike the previous three days’ methods, this one isn’t something you can do in ten minutes. Doing a full skill decomposition, measurement analysis and generating improvement could take as much as a month.

So, to get your ball rolling, I want to work on a simplified version of this method for the homework. If you do find it useful, I suggest going back and fully applying it:

1. Pick a skill you’re trying to learn.
2. Break it down into 5 subskills that you think are important. Five is a good starting number because then you won’t get distracted by the thousands of ways you can divide up a skill which often have considerable overlap.
3. Pick one skill and figure out how you would practice and measure it.

Once you’ve done that, hit REPLY and tell me what skill you picked, which subskill you isolated and how you would improve it.

Best,
-Scott

P.S. – Many of you have been asking me when Learning on Steroids will reopen. I’ll be reopening it on August 14th, after the last day of the bootcamp. From there, we’ll hold open registration for one week, until the 22nd, when I’ll have to shut it down for another year. (Don’t worry, I’ll send out an email explaining all the details once we reopen)