Scatter plot maker
Paste pairs of numbers to see whether they move together. Add a least-squares line of best fit and get its equation, r and R² instantly.
Two columns of numbers: x then y (a header row is optional). Add more y columns to compare groups. Or use Label, X, Y.
Labels & options
| Equation | y = 5.0447x + 46.3239 |
|---|---|
| Correlation r | 0.987 — very strong positive linear relationship |
| R² | 0.9743 |
| Points (n) | 12 |
| Mean of x, mean of y | 4.875, 70.9167 |
| Standard error of estimate | 2.1469 |
| RMSE | 1.9599 |
| Standard error of slope | 0.2593 |
Write a caption for this chart
Get a plain-English summary with the key numbers — handy for slide notes, reports and alt text. Your data is sent to the AI only when you click.
How to make a scatter plot
- Paste two columns: x values, then y values (a header row sets the legend name).
- Optional: add more y columns to compare groups, or use “Label, X, Y” to name points.
- Tick “Line of best fit” to draw the regression line with y = mx + b and R².
- Name both axes, then download PNG or SVG.
Drag a point, watch the best-fit line react
The line is placed so the total area of the yellow squares — the squared vertical errors — is as small as possible. Drag points up or down (or focus one and use the arrow keys).
y = 0.84x + 1.193 · r = 0.95 · squared error = 1.33
What a scatter plot tells you
Each dot is one thing measured twice — a student’s study hours and score, a house’s size and price. The pattern of dots shows the direction (up = positive, down = negative), strength (tight line vs loose cloud) and shape (straight or curved) of the relationship, plus any outliers.
The line of best fit
The line drawn is the ordinary least-squares line: the one straight line that makes the total of the squared vertical distances from each point to the line as small as possible. Its slope tells you how much y changes, on average, for each 1-unit increase in x. R² is the share of the variation in y that the line explains. Full working is on the linear regression calculator, and r alone on the correlation coefficient calculator.
Correlation isn’t causation: ice-cream sales and drownings rise together because both rise in summer.
Have a third number per point? Use the bubble chart maker.
Questions
How do I add a trend line to a scatter plot?
Tick “Line of best fit”. The least-squares line is drawn with its equation and R² in the corner, and the full statistics appear below the chart.
Can I plot several groups on one scatter plot?
Yes. Put x in the first column and each group’s y values in its own column with a header. Each group gets a colour and its own trend line.
How do I predict a y value from the line of best fit?
Type an x value into “Predict y at x =” in the statistics table. It plugs x into y = mx + b. Predictions far outside your range of x values are unreliable.
What is the standard error of estimate?
The typical vertical distance between the points and the line, in y units: the square root of the sum of squared residuals divided by n − 2. RMSE divides by n instead.
Can I upload a CSV file?
Yes. Click “Import CSV” and pick a file with x and y columns. It is read in your browser and never uploaded.
Which variable goes on the x-axis?
The explanatory variable (the one you think influences the other, or that you control) goes on x; the response goes on y.
Why does my plot not start at zero?
Axes start at zero when all values are positive and “Start axis at zero” is ticked. Untick it to zoom in on the cloud of points.