All toolsChart Types

Scatter Plot

Plot two variables against each other to reveal relationships, clusters and outliers.

What is a scatter plot?

A scatter plot places one variable on the horizontal axis and another on the vertical, then marks each observation as a point. The resulting cloud reveals whether the two variables are related, how tightly, and in which direction, along with any clusters or outliers. In finance a canonical example is a risk-versus-return chart, where each fund is a point and the shape of the cloud maps the risk-reward frontier.

Risk vs Return by Categoryscatter

Illustrative scatter of forty funds by risk and return, coloured by category. Synthetic data for demonstration.

When you want to know how two quantities relate, nothing beats plotting them directly. A scatter plot turns pairs of numbers into a shape you can read: a rising cloud means positive association, a shapeless blob means none. This tool scatters forty funds by risk and return, coloured by category, so the trade-off is visible immediately.

Reading a cloud of points

The pattern of a scatter carries the message. Points marching from lower-left to upper-right indicate a positive relationship — more of one variable goes with more of the other. A downward slope indicates a negative one. A round, directionless cloud means the two variables are essentially unrelated. The tightness of the pattern signals the strength of the relationship, and any lone points far from the crowd are outliers deserving a closer look.

Read more
Encoding a third variable turns a plain scatter into a richer tool. Colouring points by category — equity versus multi-asset, say — lets you compare two groups on the same axes and see whether they occupy different regions of the risk-return space. Sizing points by a fourth variable, such as assets under management, adds yet another dimension without adding a chart.

Correlation is not causation

A scatter plot can strongly suggest that two variables move together, but it cannot prove that one causes the other. A tight upward cloud might reflect a genuine driver, a shared hidden cause, or pure coincidence over the sample. The discipline of scatter analysis is to treat a visible relationship as a hypothesis to investigate — with theory and further data — rather than as a settled conclusion about cause and effect.

How Quadesto computes it

Quadesto renders a scatter from any two numeric columns, with an optional third column to colour points by group and a fourth to size them. It can overlay a regression line and report the correlation and R-squared. The risk-return scatter above is built from a table of risk, return and category values, and the chart embeds into research and factor reports.

Build this with your own data

Upload a CSV or connect a live source, and Quadesto renders this exact chart — styled, computed, and embeddable in your reports and newsletters. Free to start.

Create free account

Related tools

Beta Coefficient Chart

Open tool

Correlation Matrix

Open tool

Returns Distribution Chart

Open tool

Heatmap Chart

Open tool

Explore all free financial tools, browse the finance glossary, or see how teams use Quadesto.

Frequently asked questions

What does a scatter plot show?
It shows the relationship between two variables by plotting each observation as a point on x and y axes. The shape reveals whether they are positively related, negatively related or unrelated, how strong the relationship is, and where clusters and outliers sit. It is the go-to chart for exploring bivariate data.
How do you read correlation from a scatter?
An upward-sloping, tight cloud indicates strong positive correlation; a downward-sloping tight cloud indicates strong negative correlation. A round, diffuse cloud means little to no linear relationship. The tighter the points hug an imaginary line, the stronger the correlation between the two variables.
Can a scatter plot show more than two variables?
Yes. Beyond the x and y axes you can encode a third variable as point colour — often a category — and a fourth as point size. This lets a single scatter compare groups and weight observations, packing up to four dimensions into one readable chart.
Does a scatter plot prove causation?
No. A scatter can reveal that two variables move together, but correlation is not causation. The apparent link could stem from a shared hidden cause or coincidence. Treat a visible relationship as a hypothesis to test with theory and more data, not as proof that one variable drives the other.
Can I make a scatter plot from my own data?
Yes. Upload a table with two numeric columns to Quadesto, optionally add a category column for colour and a metric for point size, and the engine renders the scatter. Add a regression overlay to read correlation, then embed the chart in reports.