Skip to content
ML Academy

SQL, data modeling, data quality, EDA, visualization · insight-0069

Start bar axes at zero, and switch to dots or lines when you need to show small differences

Evidence type
documented fact
Difficulty
easy
Roles
Data analyst, Data scientist
Industries
cross-industry

Summary: situation, action, reason, boundary

  1. Situation

    A bar chart of route load factors starts its axis at 75% so the differences look clearer.

  2. Action

    Keep bars anchored at zero. If differences are small relative to the level, plot points or a line on a zoomed axis, or plot the differences themselves (versus target, versus last year) as bars from zero.

  3. Reason

    Readers judge bars and shaded areas by their length or area, so the visible ink must be proportional to the value.

  4. Boundary

    Bar charts, area charts and filled line charts.

Symptom or risk

Bar lengths no longer represent values, so a 2-point gap looks like a doubling and readers draw the wrong conclusion.

Exceptions

Ratios on a log scale should have bars that start at 1, not 0; dot plots and lines encode by position and can use a non-zero axis.

Trade-offs

Zero-based bars can make meaningful small changes look flat; choose a different encoding instead of truncating.

Pitfalls

Shading the area under a line that starts at a non-zero baseline, which recreates the bar problem.

Lessons and concepts

Claims and their sources

  • Bars on a linear scale should start at zero (principle of proportional ink); differences can be shown as bars from zero, and bars for ratios on a log scale should start at 1.

    Fundamentals of Data Visualization