Learning Science

    What Are You Measuring? Accuracy, Speed, and Consistency

    2026-05-18

    Compare two players.

    A: 97% accuracy on a sight-read, but at 70% of the original tempo. B: 92% accuracy at 100% tempo.

    Who is better?

    This is a trick question. Regardless of the answer, comparing "sight-reading ability" as a single number is the mistake. Sight-reading has three independent dimensions.

    Piano practice Piano practice. Source: Wikimedia Commons, Public Domain

    Step 1: The Three Dimensions

    Accuracy

    The most common metric. How many of 100 notes were correct.

    Pro: Easy to measure. Objective.

    Con: Limited. Even 99% accuracy depends on which 1% was missed. Missing the first melody note vs. missing one inner harmony note are not equivalent.

    Tempo

    Played speed vs. the original tempo. Expressed as a percentage — "60% tempo."

    Pro: Musically meaningful. Pieces only become themselves at 100%.

    Con: Meaningless without accuracy. 50% accuracy at 100% tempo is just fast mistakes.

    Consistency

    How stable the results are across five attempts of the same passage. Measured as variance.

    Pro: Best indicator of actual learning. A lucky run differs from reliable performance.

    Con: Hard to measure. Requires five passes through one piece. Time-consuming.

    Step 2: The Trade-Offs

    These three are not independent. They pull against each other.

    Accuracy ↑ vs Speed ↑: Raising accuracy to 99% requires slowing down. Pushing tempo to 100% forces some accuracy loss. This is a hard cognitive load ceiling.

    Consistency ↑ vs Sight-Reading New Material: Improving consistency requires repeating the same piece. But repetition does not build sight-reading ability for new pieces. A trade-off.

    Speed ↑ vs Consistency ↑: At higher tempos, consistency drops. At slower tempos, consistency improves.

    Because of these trade-offs, maximizing one metric is almost always wrong. You watch the balance.

    Step 3: Priority by Stage

    The weight of each metric shifts with experience.

    Beginner (0–6 months)

    Priority: Accuracy

    The other two are meaningless. Identifying notes correctly is item one. Speed and consistency come after accuracy is stably above 80%.

    Intermediate (6 months–2 years)

    Priority: Consistency

    If your average is 90%, the next question is whether it's consistently 90% or alternating 95% and 85%. The latter is mostly luck. Consistency leads.

    Advanced (2+ years)

    Priority: Speed

    Accuracy above 95%, consistency stable. Now you push original tempo. Accuracy will temporarily dip when you raise speed. That is normal. It stabilizes at a new equilibrium.

    Step 4: How to Measure

    How do you actually capture these?

    Accuracy

    Record. Compare to score. Count wrong notes.

    Or use a digital tool. MIDI keyboards with score-analysis software automate this.

    Self-measurement is biased. "Close enough" creeps in. Use objective tools where possible.

    Tempo

    Compare to a metronome. If the original is 120 BPM and you played at 95, that is 79%.

    Consistency

    Play the same passage five times. Record each attempt's accuracy. Compute standard deviation.

    Under 3% — stable. Above 5% — unstable. Above 10% — luck-dependent.

    You only need to do this once a week.

    Step 5: Build Your Personal Dashboard

    Pick a piece each week and record:

    WeekPieceAvg AccuracyTempo %Std. Dev.
    1Czerny 599-388%60%8%
    2Czerny 599-391%65%6%
    3Czerny 599-393%70%4%
    4Czerny 599-393%80%5%

    What this shows:

    • Accuracy and tempo rising together while consistency improves. Good learning curve.
    • Week 4 pushed tempo, consistency dropped slightly. Normal.

    Without this data, you cannot assess progress. Feelings are not reliable.

    Noteflex and the Three Metrics

    Noteflex automatically measures accuracy and tempo. End-of-session statistics show both. Consistency is indirectly estimable from accuracy variance across sessions.

    Pairing this data against the user's self-perception is core to Noteflex's value. You may feel "today went well," and the data may show "5% below last week's average." That objective feedback is what builds metacognition.

    Closing — No One Maxes All Three

    No one is perfect across all three dimensions. Not conservatory entrants. Not concert pianists.

    The difference is which dimension is stronger, and which trade-offs are made consciously. Players who know their own strengths and weaknesses — and who choose trade-offs deliberately — are the ones who go further.

    Image Sources

    • Piano practice photo: Wikimedia Commons, Public Domain

    References

    1. Lehmann, A. C., & Ericsson, K. A. (1996). Performance without preparation: Structure and acquisition of expert sight-reading and accompanying performance. Psychomusicology: A Journal of Research in Music Cognition, 15(1–2), 1–29. https://doi.org/10.1037/h0094082

    Noteflex는 서비스 개선과 분석을 위해 쿠키를 사용합니다. 자세한 내용은 쿠키 정책 을 확인해 주세요.