Common mistake 1 with Area chart

Overlaid rather than stacked areas with more than two series. The fills hide each other and the reader cannot tell what is behind what.

Common mistake 2 with Area chart

Reading a middle band's change off the chart. Its top and bottom both move.

Common mistake 3 with Area chart

Stacking series that are already percentages of different bases.

Common mistake 4 with Area chart

Using a stacked area for series that can cross in importance. Rank changes are exactly what the format hides, because each band's position depends on everything below it.

EconStatsCraft › Area chart

Area chart

Stacked, it shows composition changing over time: the outline is the total and each band is a share. The right chart for structural transformation, and the wrong one for comparing two bands with each other.

What it needs

Time or order
Components - Two or more that add up to something meaningful.

The intuition

An area chart is a line chart with the space beneath it filled, and the fill is a claim: that the quantity accumulates, so the area under the curve means something. For a flow - sales per month, energy per day - it does. For a level such as a price or a temperature, the area under the line is a meaningless number and filling it merely adds ink.

When to use it

When both the total and its breakdown matter - sector shares of GDP, spending by category, generation by fuel. If only the total matters use a line; if only the shares matter, a stacked bar or a share chart is clearer.

Where the estimator comes from

The mark is the region between the line y(t) and a baseline, usually zero. Its area is \int y\,dt, which is why the encoding suits a RATE: the area under a monthly sales line really is total sales. When the baseline is not zero the shaded region no longer corresponds to that integral, so an area chart on a truncated axis is not merely exaggerated - the fill encodes a quantity that does not exist.

What it assumes

The components add up to something meaningful. Stacking unrelated series produces a total that means nothing but looks authoritative.
Everything is non-negative. A negative band overlaps its neighbours and the stack stops being a total.
The ordering of the bands is a deliberate choice. The bottom band is the only one read accurately, so whichever series sits there is the one the chart argues about.
The categories are stable over the whole span. A component that appears halfway through makes every band above it jump for reasons that have nothing to do with those components.

Statistical properties

Area implies accumulation. Use it for flows, not for stocks or for anything where a cumulative total is meaningless.
The baseline must be zero, or the shaded area has no interpretation. This is a stronger requirement than for a line chart, where a truncated axis is a presentation choice rather than a broken encoding.
Overlapping filled areas hide each other. Two series where one is sometimes larger and sometimes smaller cannot both be seen unless the fills are transparent, and transparency makes the colours ambiguous.
A negative value breaks the metaphor entirely: the fill crosses the baseline and the eye adds two regions that should subtract.
Against a line chart, the fill adds no information at all for a single series. It buys emphasis and costs the ability to overlay a second series clearly.
For parts of a whole over time, a stacked area works but inherits every problem of the stacked bar, plus the harder task of judging a band whose baseline moves continuously.

How to read the output

Only the bottom band has a flat baseline. Every band above it is read against a wobbling floor, which is why you can see a band growing but cannot compare two middle bands with each other. Order the bands so the one that matters is at the bottom.
The outline is the total, and it is the most reliable thing on the chart.
If the question is purely about shares, normalise to 100% first: a band can shrink as a share while growing in level, and the unnormalised chart hides that.
The area itself encodes nothing - only the band thicknesses do. A tall thin band and a short wide one covering the same area are not equivalent quantities.

A worked reading

Monthly electricity generation as an area chart is a good use: the shaded region between March and June really is the energy generated in that quarter, and the eye can compare quarters by area. The same chart drawn for electricity PRICE is not: the area under a price line has units of currency-times-months and corresponds to nothing anyone wants. The line is identical in both charts; only the fill is wrong, and only in one of them.

Symptom → diagnosis → what to do

If you see this	It means	Do this
Area chart of a price or an index	Area encodes accumulation, which is meaningless here	Use a line
Non-zero baseline	The fill no longer means the integral	Zero-base it or switch to a line
Two areas overlapping	One is hidden	Use lines, or facet
Negative values	The fill crosses the baseline	Use a line or a bar chart
Stacked areas, middle band hard to read	Its baseline moves	Put the steadiest series at the bottom, or facet
Reader asked to compare two bands	Only the bottom band has a flat base	Facet, or plot each against its own axis

Common mistakes

Overlaid rather than stacked areas with more than two series. The fills hide each other and the reader cannot tell what is behind what.
Reading a middle band's change off the chart. Its top and bottom both move.
Stacking series that are already percentages of different bases.
Using a stacked area for series that can cross in importance. Rank changes are exactly what the format hides, because each band's position depends on everything below it.

Where to read this properly

Cleveland and McGill (1984), "Graphical Perception", JASA 79(387).
Few, Show Me the Numbers (2012), ch. 6.
Wilkinson, The Grammar of Graphics (2005), on marks and their implied claims.

All analyses

Data & variables

Each example is simulated from a known model, so you can check whether the estimator recovers the truth - and nothing here is anyone's private data.

Filter variables

Drag a variable into a slot on the right - or just click it.

Choose an example dataset above to begin.

Choose a transformation - it opens in the data editor, with a live preview.

Apply to

Multiply by

Columns ctrl-click for several

Group by ctrl-click for several

Within groups

Statistic

Bands

How many

Cut points

Keep rows where

Value

and

Fill with

Value

Also mark which rows were filled

Rules

Base period = 100

Which part

Percent at each tail

Threshold

Survey weight optional

Direction

Names column

Values column

Largest first

Delete them, not just count them

Read this column as

Tidy the text

New name optional

EconStatsCraft rigorous, plain-spoken econometrics

Econometrics that explains itself.

Regression, panel data, time series and the whole battery of tests. Every result is followed by a verdict you can defend.

Load a dataset to see it here.

Area chart

What it needs

The intuition

When to use it

Where the estimator comes from

What it assumes

Statistical properties

How to read the output

A worked reading

Symptom → diagnosis → what to do

Common mistakes

Where to read this properly

See also

Related Charts

All analyses

Charts

Maps

Explore

Regression

Time series - one series

Time series - several series

Panel data

Multivariate statistics

Marketing research

Causal

Finance & risk

Actuarial

Tests & breaks

Machine learning

Econometrics that explains itself.