The ranking is an accuracy ceiling; alignment and mark size erode it

What it is

Realized graphical judgment under presentation constraints separates a channel's relative potential under controlled conditions from performance in a finished product. An empirical ranking is a useful starting point for choosing an encoding, but not a fixed mathematical ceiling: alignment, mark size, density, display conditions, and reader differences can amplify error. Labels, gridlines, interaction, and training can also alter the complete task rather than merely making one channel worse.

Why it happens

When a shared baseline is broken apart, the task may change from comparing positions on a common scale to comparing nonaligned positions or lengths. Very small marks encounter pixel quantization and visual thresholds; dense marks introduce crowding, occlusion, and visual-search competition. Low contrast, zoom, viewing distance, and device quality further weaken discriminable cues. Chart literacy, vision, domain knowledge, and judgment strategy affect how a reader uses those cues. The same channel name therefore does not guarantee the same stimulus, task, or reader condition.

Studying it

Use factorial experiments or controlled product evaluations to vary baseline alignment, mark and chart size, spacing or density, gridlines, and display conditions, while measuring error distributions, time, and failure modes. In the same study, Heer and Bostock replicated classic graphical-perception experiments and separately tested chart size and grid spacing, illustrating why presentation variables require measurement rather than deduction from an abstract ranking. Report final pixel or physical dimensions, resolution, viewing context, and sample composition; model condition interactions and individual differences instead of universalizing one breakpoint or device average.

Where it stops holding

“Ceiling” is a cautionary design metaphor here. Classic rankings describe relative performance for particular stimuli and magnitude judgments, not a theoretical maximum for every visual-analysis task. Direct labels, tables, zoom, filtering, and tooltips change both available information and interaction cost; high density can sometimes help reveal global shape. Exact reading, safety decisions, and accessible use require verifiable values and equivalent paths, not merely a larger chart.

Applying it

• Inspect final rendering, not only design files, at target breakpoints, zoom levels, exports, print sizes, and projection conditions; record actual size and display assumptions.
• Preserve a shared scale and baseline for consequential comparisons where possible. When alignment must break, add reference guides, direct labels, or a comparison view, then verify that error falls.
• Establish minimum distinguishable size and acceptable density through task testing rather than a universal pixel threshold. Aggregate, facet, filter, or disclose on demand when crowding wins.
• Provide text, tables, keyboard operation, and assistive-technology access to critical values; revalidate when devices, audiences, or content density change.

Related

• Same group: U1.01.1 Visual channels have a stable empirical accuracy ranking · U1.01.2 The ranking comes from magnitude-judgment experiments, not design intuition · U1.01.3 Quantitative and categorical data need two different channel rankings · U1.01.4 The most important variable should get the highest-ranked available channel
• Adjacent: U1.02.1 Position along a common scale is the most accurate channel · U1.02.2 Losing the shared baseline sharply degrades position judgment
• Search terms: chart size · alignment effect · display conditions