Should You Play a New Score First—or Study It First?

Building a Piano Piece · Article One

When an unfamiliar score first reaches the music desk, the most natural impulse is to put both hands on the keyboard and begin at bar one. If the music continues, keep going; if something goes wrong, stop and repeat it. The score appears to be an instruction sheet, so playing seems like the obvious way to read it.

The opposite approach can become equally rigid. A player may feel obliged to analyse the key, harmony, form, themes and fingering before sounding a note, as though complete intellectual understanding were the admission price for touching the keyboard.

Each approach contains part of the truth. Neither is a good complete method. When the aim is to learn a piece over days or weeks, the most useful first encounter is usually:

Build a brief map of the score, then make a controlled trial at the keyboard to obtain auditory and movement information that the page alone cannot supply.

Study first, but not until every question has been answered. Play next, but do not treat the first attempt as a performance. Its purpose is diagnosis.

First identify the task

“Seeing a score for the first time” describes at least three different tasks.

In a sight-reading examination, preparation is deliberately short. The player must preserve pulse and continuity while converting as much notation as possible into music. Stopping to repair each wrong note would defeat the task.

In quick study, perhaps twenty minutes or a day is available before a performance. The challenge is to identify the largest risks and allocate limited practice efficiently.

In repertoire learning, the aim is a stable interpretation that can be reproduced in different conditions after sustained work. This article is about the third task.

The distinction matters because a useful sight-reading instruction—keep going—can become wasteful when applied throughout repertoire learning. A first run-through may reveal the general shape of a piece, but repeated undiagnosed run-throughs can turn practice into a series of reactions rather than a plan.

Why beginning with the first note is not enough

Music reading is not simply a chain in which the eye identifies one note and a finger responds. Research on the eye–hand span in sight-reading shows that the eyes commonly move ahead of the note currently being sounded. This enables a player to process upcoming pitches, rhythms and movements. The span is not a fixed prescription of so many notes or beats; it changes with expertise, tempo and score complexity. A review of the research connects skilled sight-reading with pattern recognition, working memory, processing speed and adaptive visual strategies.

The practical implication is modest but important. Notes should not always be read as isolated dots. Scales, chords, arpeggios, rhythmic cells, voices and recurring textures are larger units of information. If the eye can only chase the fingers, a clef change, leap or altered repetition arrives as a surprise. A brief survey does not remove the difficulty, but it makes the difficulty visible before it interrupts the music.

Evidence about practice behaviour points in the same direction, although it must be interpreted carefully. In a study of 55 string players aged six to eighteen, less advanced players often played repeatedly from beginning to end and frequently left errors uncorrected. More advanced players were more likely to identify difficult sections, concentrate on them, and reintegrate them into the whole. Hallam’s study was not a piano experiment and does not establish a universal causal rule. It does, however, expose a general practice problem: repetition cannot target a problem that the player has not identified.

A much smaller study observed five tertiary piano students learning two quick-study pieces. The higher-ranked performances were associated with planning, interpretive goals, score study, score marking, pauses for thought and listening to recordings of one’s own playing. The weaker practice was more reactive. Because the sample was only five, the study should not be treated as proof of one ideal sequence. It is better read as close evidence that productive practice includes non-playing decisions as well as keystrokes.

Why analysis cannot replace playing

The page does not disclose everything. A pattern that looks comfortable may expose a hand-position change at the keyboard. A neat fingering may fail when the passage is connected to its neighbours or brought closer to performance tempo. Voicing and pedalling cannot be settled solely through visual analysis because the answer depends on the sound produced, the decay of the instrument and the room.

The purpose of preliminary study is therefore not total explanation. It is a minimum viable map: enough information to know what the first trial should test.

Mental-practice research confirms that non-playing preparation can involve genuine musical and motor planning. In one experiment, sixteen professional pianists mentally rehearsed an unfamiliar technical passage by imagining keys, posture, movement and sound. The study investigated changes in motor anticipation. Its small expert sample and specialised material do not justify replacing physical practice with imagery. They do support a narrower conclusion: a short silent preview can be active preparation rather than idle delay.

Study and playing produce different evidence. Looking reveals structure and foreseeable risks. Playing tests sound, movement and fingering. Reliable learning develops through a cycle between them.

A fifteen-minute first-contact protocol

The following protocol is an evidence-informed editorial synthesis, not a fifteen-minute formula validated as a whole in a controlled trial. Its timing can be expanded or compressed. The useful elements are mapping, imagining, testing and recording what was learned.

1. Verify the score: about one minute

Check the work, movement, edition and page order. Locate the key and time signatures, upbeat, tempo indication, repeats, first and second endings, Da capo, Dal segno, coda signs, clef changes, octave signs and page turns. Where the edition explains its practice, distinguish editorial fingering or pedalling from the composer’s markings.

This prevents avoidable waste: learning the wrong octave, missing a repeat or treating an editor’s suggestion as immutable source text.

2. Build a structural map: three to five minutes

Scan the whole piece before reading every note. Ask where the texture changes, which patterns recur, where the music cadences or returns, and where leaps, dense chords, double notes, hand crossings or accidentals cluster. Similar-looking passages deserve special attention because a small difference is easy to overwrite with the more familiar version.

Mark broad sections with A, B and A′, or use ordinary labels such as opening, contrast, return and coda. A beginner does not need a Roman-numeral analysis of every chord. Bass movement, phrase endings and recurring shapes may be enough. Analysis should reduce uncertainty in the next action, not become a separate examination.

3. Hear and simplify: two or three minutes

Tap or count a representative rhythm. Sing or imagine the melodic contour. Follow the bass alone. Reduce a rapid figuration to its underlying chords or hand positions. Simulate a leap or exchange of hand positions on a closed keyboard lid.

If a passage cannot yet be heard internally, that is useful diagnostic information. During the first trial, the goal there should be to discover the sound, not to achieve speed.

4. Make a controlled trial: about five minutes

Choose a tempo at which pulse can be maintained while attention remains available. A short, moderate piece may permit one continuous overview. A long or technically advanced piece may be sampled by section; there is no prize for forcing an unprepared complete run.

Do not stop at every accidental slip during the overview, because that destroys information about continuity. Equally, do not label every failure “unfamiliarity”. Decide whether it is a decoding, rhythm, fingering, coordination, leap, voicing, pedal or memory problem. The categories suggest different next actions.

Treat the first fingering as provisional. Test crucial choices in context and at more than one tempo. A fingering that is comfortable on a static chord may be poor for the phrase that enters or leaves it.

5. Make a problem map: about two minutes

Before playing the whole piece again, mark three to five important locations. A simple code is enough: A for material that can already be read; B for something requiring a slow test; C for a point that breaks continuity and needs separate work. Add the cause—left-hand leap, uncertain rhythm, awkward substitution, buried melody, or pedal to be tested.

At the end of the first encounter, answer three questions:

1. What are the principal sections and returns?
2. Where are the three largest difficulties, and what causes each one?
3. What will be the first task in the next practice session?

If none can be answered, reaching the final bar has not yet produced a useful first reading.

An example: Bach’s Prelude in C major

The Prelude in C major, BWV 846, from Book I of Bach’s Well-Tempered Clavier, is a clear example. The score is publicly available through IMSLP.

Much of the work is written in continuous broken-chord figuration. Read note by note, the page appears to contain a large number of separate semiquavers. Read by pattern, it reveals a recurring figuration whose bass and harmony change over time.

A first survey should identify the persistence of that pattern and the points at which the texture or sonority changes. The player can then follow the bass and reduce each bar to a harmonic shape before slowly testing the opening at the keyboard. The first trial should listen for direction within each group, continuity of the upper line, bass movement, physical ease and clarity of harmony—not merely whether every note was struck.

A well-known case study of a concert pianist learning the Presto from Bach’s Italian Concerto showed that expert memory involved more than motor and auditory familiarity. Formal structure provided a retrieval organisation, while fingering, phrasing and interpretive decisions served as performance cues. Chaffin and Imreh studied one elite pianist, so their findings cannot prescribe what every learner must do. They nevertheless demonstrate how attention to structure can give later memory more than one point of support.

Should a recording come first?

There is no strong basis for a universal rule that every score should be heard before it is played, or that independent musicianship requires avoiding recordings.

A recording can reveal broad tempo, style and proportion when the notation or idiom is unfamiliar. It can also anchor imitation before the learner has read the score independently. A balanced procedure is to make an initial survey and trial first when independent reading is the goal; consult a reliable recording after that if style or notation remains unclear; compare more than one performance where possible; and return to the score to separate compositional information from a performer’s choices.

Completion does not mean reaching the last bar

In a study of eleven professional pianists, accurate sight-reading did not guarantee equally successful memory after a short period of practice. Memory outcomes varied considerably. The small study cautions against equating an initially fluent performance with organised, durable learning.

A productive first encounter ends with four reusable products: a structural map, a classified list of problems, provisional fingering at critical points and a plan for the next session.

The answer to the title question can therefore be stated precisely:

Study first so that you know what the first playing should investigate; then play so that you can discover what score study has not yet told you.

The goal is not to conquer an unfamiliar piece at first sight. It is to transform an unfamiliar score into a set of musical and technical questions that can be tested one by one.

Primary sources and further reading

• Review of eye–hand span research in music sight-reading
• Susan Hallam on the development of instrumental practice strategies
• Play and non-play strategies in tertiary piano practice
• Mental practice and motor anticipation in skilled pianists
• Chaffin and Imreh on piano performance as expert memory

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.