ManYue: Read bar-chart baselines before comparing heights

A meeting slide or poster in a comic may show bars whose apparent threefold height does not mean threefold data values. Read coordinates, units and baseline before checking a character’s explanation. Arithmetic uses an invented sample or explicit labels, while ManYue supplies navigation and zoom, not automatic chart reading, pixel measurement or judgement of deceptive intent.

Keep the chart title and nearby dialogue

Choose a panel with legible title, vertical axis and two bars. Record work, edition, chapter, file position and panel number. Retain who presents and explains it without treating a character’s claim as the author’s conclusion. Without a suitable work, draw two panels: counts of forty hats for A and sixty for B, then a character discussion. Label the fictional teaching example, not real population data. If the chart is too small to reveal axes, mark insufficient information rather than inventing unit and total from coloured strips. Limit later judgement to supplied evidence.

Copy axes and measurement scope

Copy A/B categories, vertical unit, tick spacing, starting value and top value, preserving numbers separately from the title. Establish counts rather than ranks, percentages or logarithmic scale; integers alone do not decide. The sample explicitly counts hats on one basis, without comparing group sizes. Record any axis break and explanation. If labels are absent, write unlabelled instead of treating an assumed zero as visible fact. Confirm a shared axis: screen heights on two different axes are not directly comparable.

Calculate the value ratio for a zero baseline

In version one, the vertical axis starts at zero: A forty, B sixty. B exceeds A by twenty, and sixty divided by forty is one point five. On an equally spaced linear axis with a shared zero, bar heights can encode that ratio, provided the drawing matches labels. Twenty hats is not twenty percentage points. Calculate from labels rather than claiming precise values from a phone-screen ruler. If a comic uses decorative schematic bars, retain illustrative or unknown-scale status rather than assuming measurement precision.

Separate visible lengths from values on a truncated axis

Version two retains forty and sixty but starts the bar baseline at thirty. Visible lengths represent ten and thirty, a threefold ratio, while original values remain forty and sixty, ratio one point five. Separate baseline, endpoint value and visible difference rather than changing data because the bars look more unequal. A truncated axis can emphasize differences, but suitability depends on context; a nonzero baseline alone does not prove fraud. Conversely, hiding that baseline does not justify claiming threefold quantities. Identify the comparison that lacks a common zero.

Keep character claims separate from chart evidence

If the sample character says B is three times A, record a character claim, then explain from axis and labels that it substitutes visible-length ratio for value ratio. In real work, establish whether threefold refers to bar height, increase or quantity before calling it wrong. A claim explicitly limited to displayed length differs from threefold count. Attribute narration, character and title separately rather than deciding author attitude for them. The task checks support from the visible chart. Motive, irony and unreliable narration require additional story evidence.

Change only the baseline for comparison

Place both versions side by side in external notes, holding forty and sixty fixed while changing only zero versus thirty baseline. Record their common difference twenty and value ratio one point five. If tick spacing also changes, identify a separate condition instead of combining experiments. Do not alter the original page to prove your interpretation; a labelled self-made sketch may assist. This explains changing appearances for the same data, not how to make more persuasive promotion. Without original labels, do not infer numbers by redrawing; preserve insufficient-data status.

Exclude cropping and layout when details are unreadable

If the starting value is invisible, restore a whole-page or suitable single-page view to exclude screen cropping or reduced paired-page labels. Zoom reveals existing text, not ticks absent from the source. A fuller nearby version may assist only after matching title and edition; do not substitute an unsourced online image. If unclear, mark baseline undetermined rather than assuming zero or deliberate truncation. Feedback identifying the missing axis, number and panel is easier to verify than a generic misleading-chart label.

Recheck three numerical relationships after returning

Leave the reader and return to title, baseline and endpoints. Without the old answers, compute original difference twenty, value ratio one point five and truncated visible lengths ten versus thirty, ratio three. State what each number measures, retaining drawing-scale limits and provenance of character claims. Mark observations not performed; the invented example does not establish the same issue in a real work. Completion explains one specific comparison from chart evidence, not an entire plot’s truth from the bars’ dramatic appearance.

Related guides and sources

ManYue comic infographics: distinguish icon counts, areas and legend units; ManYue testimony versus pictures: preserve contradictions before judging truth.

Checked on November 8, 2026: Apple public version information lists ManYue 2.0.4, requiring iOS / iPadOS 15.0 or later. Feature scope follows the official App Store description and the installed version. This article provides reading and checking methods you can perform; it does not claim step-by-step physical-device or private-server testing. Manual records, external tools and actually visible controls are identified separately, and personal analysis or code observations are not presented as verified automatic capabilities.

For nonzero baselines and visual comparison, see the original research Truncating the Y-Axis: Threat or Menace?. The forty/sixty example is original; it neither reproduces study results nor treats stylized comics as measurement diagrams.