The Self · Daily Learning Experience
West Berlin, 1991. A practice room above a courtyard, ten past six in the morning — and a sheet of paper cut into ninety-six little boxes.
Act One · The Room
The sheet is not a test. It is a grid. Seven columns for seven days, and every day sliced into ninety-six little boxes — one for each quarter of an hour, midnight to midnight. Write in what you were doing. All of it. Sleeping counts. Breakfast counts. Standing at the window not practising counts.
The man is Anders Ericsson. With Ralf Krampe and Clemens Tesch-Römer he is running a study at the Hochschule der Künste in West Berlin that will be published in 1993 and will, entirely by accident, become one of the most misquoted numbers in the world.
They talked to forty violinists. Ten whom the professors named as having “potential for careers as international soloists” — the paper calls them the best violinists. Ten good violinists from the same academy, matched for age and sex. Ten music-education students training to teach. And ten middle-aged professionals from the Berlin Philharmonic and the Radio Symphony Orchestra, for the long view.
Three sessions each. One interview about a whole life. One diary week. And one question that sounds easy and isn’t: how many hours a week did you practise alone, in every year since you started?
Before we tell you what the sheets said — a number from you. Then we will hand you the same grid.
Ericsson collected both: what people remembered doing, and what a diary said they actually did. The diary was the check on the memory. So: guess first, on your honour, then paint.
You can skip this. Nothing here is scored, and there is no right answer.
Now paint one ordinary day of yours. Pick a colour, then click or drag across the quarter-hours. Midnight is at the top.
Act Two · What the sheets said
The two top groups practised alone an average of 24.3 hours a week. The future music teachers: 9.3 hours. That is the headline, and it is not surprising.
The shape is more interesting than the size. The best groups practised about three and a half hours a day, in roughly 19.5 sessions a week averaging 80 minutes each — and they clustered those sessions in the late morning, between about ten and two. Not one heroic marathon. Chunks, with the day built around them.
Then they added up a life. Working from the year-by-year estimates, by the time each violinist turned eighteen they had accumulated, on average:
Accumulated hours of practice alone, by age 18
Ericsson, Krampe & Tesch-Römer 1993 — averages for each group of ten violinists
| Group | Hours by age 18 |
|---|
Source: Psychological Review 100(3), 363–406. By age twenty the top group averaged about ten thousand hours — hold on to that figure; it is about to escape.
The diaries also recorded sleep. The best and good violinists slept 60.0 hours a week. The future teachers slept 54.6. Naps, in the afternoon, on purpose: 2.8 hours a week against 0.9.
The people doing the most work were also the people doing the most sleeping. Rest is not the opposite of the finding. Rest is inside the finding — and it is the part that gets cut when the study is turned into a slogan.
“…essentially no benefit from durations exceeding 4 hr per day and reduced benefits from practice exceeding 2 hr.”Ericsson, Krampe & Tesch-Römer, 1993, p. 370
Ericsson meant something narrow by deliberate practice, and almost everyone who quotes him means something wider. His version needs four things at once: a goal just past what you can already do; your full attention; feedback fast enough to matter; and another go, corrected. Drag each card into a tray. Then see how he would have sorted them.
Disagreeing is allowed and interesting. This is a definition, not a law of nature — and the boundary is exactly what researchers are still arguing about.
This dial describes a task, not a person. It cannot tell you anything about who you are; it can only tell you whether the bit you keep repeating is currently too easy, about right, or coming apart in your hands.
You can skip this. Nothing here is scored, and there is no right answer.
Set the shape of a week you could really keep — not a heroic one. The page will do the sum the internet loves to do, and then, in a moment, we will show you why the sum is a trap.
Act Three · The number that got away
Malcolm Gladwell publishes Outliers. Chapter two is called “The 10,000-Hour Rule.” It opens with the Berlin violinists. Within a few years the figure is on classroom walls, in business books, in speeches, in songs. A study about what kind of hour changes you has become a study about how many.
In 2016, in Peak and in an essay of their own, Ericsson and Robert Pool set about taking it back. Their objections, in their own words:
“There is nothing special or magical about ten thousand hours.”Ericsson & Pool, 2016
Gladwell, they wrote, “incorrectly claimed that all the violinists in that group had accumulated over ten thousand hours” — when in fact “half of the ten violinists in that group hadn’t actually accumulated ten thousand hours.” Ten thousand was an average at age twenty, and at twenty those students were, as Ericsson put it, nowhere near masters of the violin. And an hour on a stage is not an hour of deliberate practice.
Then the counting caught up with the story.
How much of the difference between people does practice explain?
Percentage of variance in performance accounted for by deliberate practice, by domain
| Domain | Variance explained |
|---|
Source: Macnamara, Hambrick & Oswald, Psychological Science, 2014. Overall correlation r = .35 — about 12% of the variance across all domains pooled.
And in 2019, Brooke Macnamara and Megha Maitra went back to Berlin’s question with 39 violinists, thirteen per group, and had the skill ratings made double-blind — the raters could not see anyone’s practice hours. The gradient did not come back. The group rated good had logged 9,844 hours; the group rated best had logged 8,224. The difference was not statistically significant (p = 0.364), and it pointed the wrong way.
Ericsson’s reply deserves to be on the page too, because it is fair: the later studies, he argued, counted a looser thing than he had defined. Anything anyone called practice went into the total. His whole point had been that the hours are not interchangeable.
So what is left standing? Nobody in this argument — nobody — thinks you can skip the work. The fight is about how much of the gap between two people the work explains, and what else is in the room: teachers, money, bodies, luck, age of starting, and whether anyone ever showed you what to fix.
Six wobbles
Every skill has one. The bar you skate over. The word you avoid saying. The angle you never shoot from. It is almost always the smallest thing in the room and it is almost always the thing that moves.
You can skip this. Nothing here is scored, and there is no right answer.
Nothing below is a score, a type or a diagnosis. It is your own writing handed back, plus one pattern we think we can see. You are the only person who can say whether it is true.
Does that land?
The long argument
The graded bit
This part is scored — and it is scored on the story and the evidence, never on anything you wrote about yourself. Your answers upstairs have no right or wrong. These do.
Try this at home
Here is the question to walk away with. If the hours are not interchangeable — if one kind of hour moves you and another kind only keeps you company — then what would you find out about yourself by counting, for one week, only the hours of the first kind?