Use position encoding for interval point comparisons

Encode value with position when viewers must compare individual points or point-derived ranges across time intervals. For example, use a line graph or modified stock chart instead of a colorfield, woven colorfield, or event-striped color view when the task is to find the interval with the highest point, lowest point, or largest range.

Point comparisons require readers to extract specific values before they compare intervals. Position supports that extraction more reliably than color when the chart must preserve individual highs, lows, and spans.

Mechanism: Position makes it easier to read and compare exact point locations, while color supports coarser summaries and makes single-point lookup less precise.

Evidence: In the maxima, minima, and range experiments, position-based encodings generally outperformed color-based encodings, supporting the prediction that position is better for point comparisons in time series aggregation tasks (Albers et al., 2014).

• User Goal: Identify which interval contains the highest value, lowest value, or largest max-minus-min gap.
• Task: Compare point-level extrema across known time blocks such as months.
• Data: Quantitative time series grouped into discrete intervals.
• Chart Setting: A static time-series display where value could be encoded by position or by color.
• Success Criterion: Higher accuracy on interval extrema and range judgments.

Break it when: The primary task is comparing interval summaries such as average or spread rather than extracting individual points. Why: The paper found that color-based aggregation and explicit interval summaries can better support those summary judgments than raw position alone.

Sacrifice: You give up some support for fast visual summarization of whole intervals. Risk: A plain line can still leave average or spread judgments hard if those summaries are not encoded explicitly. Mitigation: Add task-relevant interval summaries when summary comparison matters alongside point lookup.

Mistake: Encode value only with color for tasks that ask readers to find interval maxima, minima, or ranges. Why it fails: Readers must recover specific point values from color, and the study showed that this lowers accuracy on those point-comparison tasks.

Failure Sign: Reviewers can describe broad color differences but struggle to name the interval containing the extreme point or largest range. Quick Check: Show the same series once with position encoding and once with color encoding, then ask which interval contains the highest point, lowest point, or largest range. Stronger Test: Repeat the A/B check on both smoother and noisier series to see whether the position version remains more reliable.

• Switch the primary value channel from color to vertical position.
• Replace a colorfield-style view with a line-based position view for extrema and range questions.
• Add discrete interval markers for highs and lows if readers must compare ranges across intervals.