Prefer position encoding over color encoding for time-series extrema and range comparisons

Encode values with position when readers must find interval maxima, minima, or ranges in a time series. For example, use a line-based or box-plot position display instead of a colorfield or color-stock display when the question is which interval contains the highest point, lowest point, or widest within-interval span.

Point comparisons require readers to recover specific values before they compare intervals. Position supports that recovery more faithfully than color saturation, which is less accurate for exact point lookup.

Mechanism: Position makes local highs, lows, and gaps easier to read directly, so readers do less inferential work before making the interval comparison.

Evidence: In the time-series study collated by the review, position-based displays outperformed color-based displays on maxima, minima, and range questions, and the review summarizes the same paper as showing position-based charts beating color-based charts for find-extremum and determine-range tasks. (Albers et al., 2014; Zeng & Battle, 2023)

• User Goal: Pick the interval with the highest point, lowest point, or largest within-interval range.
• Task: Point comparison across repeated time intervals.
• Data: Ordered time series partitioned into repeated intervals such as months.
• Chart Setting: Static time-series display where values could be encoded by either vertical position or color saturation.
• Success Criterion: Higher answer accuracy on interval-level extrema or range questions.

Break it when: The primary question is interval average or spread rather than point maxima, minima, or range. Why: the same study found summary-oriented designs performed better on those summary tasks than raw point extraction alone.

Sacrifice: Some fast field-level summarization that color-based views can support. Risk: A position-first view can be a worse fit for average and spread judgments. Mitigation: Switch to interval summary encodings when the task changes from point lookup to summary comparison.

Mistake: Keep a colorfield as the main value encoding for a max, min, or range task. Why it fails: Readers must infer exact point values from color, and accuracy drops on point comparisons.

Failure Sign: Reviewers hesitate or disagree about which interval contains the highest point, lowest point, or widest range when reading a color-encoded series. Quick Check: Show the same series once with position encoding and once with color encoding, then ask three forced-choice interval questions on max, min, and range. Stronger Test: Keep the position version if it yields more accurate answers on brief interval extrema and range questions.

• Replace color-saturation value encoding with vertical position for views whose main job is max, min, or range comparison.
• If interval summaries must stay visible, add position marks that expose interval highs and lows directly.
• Remove color-only value views from tasks that depend on exact point extraction.