A misleading graph does not need a false number to distort the truth. Truncating the y-axis, cherry-picking the data range, using an uneven scale, or adding false 3D depth can all make an ordinary difference look dramatic, or hide a real one. Learn to recognize each trick.
How a graph can mislead without a single false number
A misleading graph uses true numbers but presents them in a way that gives a false impression. No number on the graph has to be wrong — the distortion comes from choices like the axis scale, which values are included, or how the graph is drawn.
The truncated axis: the most common trick
The most common way a graph misleads is a truncated y-axis — starting the vertical axis somewhere above zero instead of at zero. This stretches small differences into what looks like a huge gap.
Same two values, 82 and 86. Starting the axis at 80 instead of 0 makes an ordinary difference look dramatic.
In the left graph, value B looks almost twice as tall as value A — but the real difference is only 4 out of roughly 85, a difference of less than 5%. The honest graph on the right, with an axis starting at 0, shows the bars nearly the same height, which is the truthful picture.
Other ways graphs mislead
Cherry-picked range: showing only the months or years where a trend looks favorable, and cutting off the rest.
Inconsistent scale: using unevenly spaced intervals on an axis so equal-looking gaps represent unequal amounts.
3D distortion: tilting or adding depth to bars or pie slices, which visually inflates some segments over others.
How to spot one
Always check where the y-axis starts, whether the full relevant time range is shown, and whether the scale is even. This is one of the checks covered in critiquing data presentation. Comparing this to the advantages and disadvantages of different graphs also helps: even the right graph type can still be drawn dishonestly.