When a Difficult Passage Keeps Failing, Should You Repeat It or Change the Task?

Building a Piano Piece · Article Eight

After a passage has failed ten times, two reactions are common. One is to repeat with greater force until the fingers finally “remember”. The other is to attribute the problem to talent, hand size or an excessively difficult work and avoid it.

Both conclusions may be premature. Repeated failure first indicates that the current task has not created conditions in which correction can be learned. Repetition is useful when the error can be identified and a change can be executed. When every attempt repeats the same confusion, additional trials merely increase familiarity with confusion.

The first principle is:

Do not begin by asking how many more repetitions are needed. Rewrite “I cannot play it” as an observable type of failure, then change one condition that may be causing it.

Changing the method is not avoidance. It converts an outcome description into a causal hypothesis.

A wrong note is an outcome, not a diagnosis

The same wrong pitch can have many causes: an accidental was not read; the rhythm was grouped incorrectly; fingering changes each time; the eyes locate a landing too late; the preceding movement never releases; a bimanual alignment is unclear; the tempo exceeds current organisation; or attention is captured by another voice.

Different causes need different tasks. Reading errors call for renewed score study. Rhythm may need counting or tapping. Location needs departure, path and landing. Oversized slow movement may require a near-tempo burst. Bimanual problems need a simplified combined task. Memory needs structural cues and varied starting points.

An interdisciplinary framework for musicians’ error management distinguishes the role of errors in exploration, declarative understanding, procedural learning and creative work. Experts are not error-free; they often detect, classify and recover errors earlier. The article by Kruse-Weber and Parncutt is a conceptual framework, not a validated piano treatment, but it makes the essential point that practice develops error management as well as error reduction.

Find the first deviation, not the final wrong key

Look earlier than the audible mistake. When did the plan first diverge? Did the eyes remain on the current note? Did the hand fail to travel? Had pulse already slowed? Did a voice become unexpectedly loud? Did the body brace in anticipation?

Audio recording reveals timing and sound; video can help inspect gaze direction, hand position and excess movement. Slow-motion viewing is not a medical diagnosis, but it can distinguish a wrong landing from a preparation that failed two beats earlier.

Write a testable statement:

not: this bar always goes wrong
but: the right hand looks towards the bass only after releasing the previous chord,
so the leap begins one beat too late

The next trial can change gaze timing, travel or release rather than attacking the wrong note again.

Seven ways to change the task

1. Change information load

Remove one voice, separate the hands, omit an ornament temporarily, or block broken chords. The simplification is diagnostic, not a permanent arrangement.

2. Change time

Slowing supports observation, but if very slow movement differs from target organisation, use a short near-tempo burst, rhythmic variants or planned pauses. Place pauses at genuine movement groups rather than arbitrary points.

3. Change the spatial demand

For a leap, rehearse silent travel, play only departure and landing, vary the starting distance, or direct gaze to the target before moving. A pilot experiment compared fixed, variable and spaced practice of wide left-hand piano leaps and found only partial support for variability benefits. The study involved thirty-six music students and one specific task. It does not prove that every leap should use random distances; it suggests that modest variation can test whether learning belongs only to one memorised path.

4. Change attentional focus

Closing the eyes is not universally safe or useful. More controlled changes include looking earlier at a landing, moving attention from fingers to sound, or monitoring release instead of the final note. Change one focus at a time.

5. Change the starting point

Beginning on the difficulty may remove its real entry conditions. Start from a preceding beat, harmony or phrase. Also start after the difficulty so that its ending does not become a stop signal.

6. Change feedback timing

Do not restart the whole section after every accidental slip. Complete the defined short unit, then evaluate how the error affected continuation. When a known dangerous error repeats, reduce the task immediately rather than proceduralising it.

7. Change physical load

Reduce dynamic force, shorten work rounds, and inspect seating and arm path. Piano biomechanics contains diverse, individual movement strategies, and a recent systematic review found substantial variation in tasks, samples and measurements. It does not support one universal “correct movement”. Pain, tingling or persistent discomfort is a stop signal, not a normal breakthrough phase.

When should the original repetition continue?

Maintain the task while error is falling, the cause remains consistent, and each correction can be implemented. A leap may become progressively less variable; a ratio may stabilise after counting; interference from an old fingering may diminish.

Use a small round—perhaps three attempts—then assess the trend. One unexplained success among two failures is not evidence that the passage is about to be “ground into place”.

When may the material exceed current capacity?

Changing method does not guarantee that every passage can be solved now. If slowing, simplifying, sectioning, revising fingering and obtaining teacher guidance still do not produce the underlying movement without pain and loss of control, a related, easier task may be needed first.

Judge this at the level of the foundation, not final tempo. If isolated slow octaves already produce persistent tension, hundreds of continuous octaves are not the bridge. If voices cannot be identified separately, full-texture repetition will not create contrapuntal hearing.

Temporary deferral can include a smaller span, a slower but structurally similar figure, reduced voices, a relevant study, or a teacher-designed exercise. The bridge should address the diagnosed cause rather than add generic technical volume.

Example: the left-hand material of Chopin’s “Revolutionary” Étude

Chopin’s Étude in C minor, Op.10 No.12 is known for continuous rapid left-hand figuration. The score is available through IMSLP. Repeating a long left-hand span from the beginning makes it difficult to distinguish fingering, harmonic position, group transition, thumb use, arm travel and endurance.

Begin by blocking the figuration into chords or positions to map the harmony. Mark each relocation and its preparation. Practise one complete group with entry and departure. Use a brief near-target trial to see whether the slow movement is too large. Only then increase length, observing which error returns first as endurance is required.

If the second group repeatedly fails, beginning every time at the first group merely overlearns the first. Start before the transition, travel through it and continue to the next stable event. The practice object is the positional relationship, not all the notes printed on the page.

A difficult-passage prescription sheet

Use five columns:

The table does not create false scientific precision. It prevents simultaneous changes to tempo, fingering, dynamics and starting point that make it impossible to know what helped.

Stronger evidence that a difficulty is solved

A passage is not solved because it passed once today. Better evidence includes: the former cause can be explained; entry is possible from different points; modest variations of tempo, dynamics or context do not destroy it; it returns after rest; and a minor slip can be survived without restarting the piece.

The final criterion matters. Reliable performance is not a myth of zero error. It is a system with tolerance and recovery. Practice should occasionally allow a small error to continue so that the route onward can be tested.

Persistent error usually calls not for stronger will but for better problem definition. If the task remains unchanged, failure can remain unchanged. Once the relationship being learned changes, repetition becomes meaningful again.

Primary sources and further reading

• An interdisciplinary framework for musicians’ error management
• A pilot study of variable practice in wide piano leaps
• Systematic review of the biomechanics of piano playing
• Using a music microanalysis protocol to enhance practice

Continue reading: Building a Piano Piece series page

If you would like to bring these ideas about listening, understanding, and practice to the keyboard, you might try ScoreFlow, an app I developed to make score reading and daily practice flow more naturally together.