Free Scatter Plot Maker with Line of Best Fit

Place each observation on its own line with X first and Y second, separated by a comma, tab, semicolon, or space. Name both axes and confirm that every row represents one paired observation. The maker plots valid pairs immediately and lists rows that need correction.

Keep the Line of best fit switch on. The tool calculates an ordinary least-squares linear regression with an intercept, then draws the fitted line across the visible X range. It also reports the slope, intercept, Pearson correlation, R², and sample size from the same valid pairs.

The equation predicts Y from X using the fitted slope and intercept. R² describes the share of Y variation explained by that linear fit in this sample. It does not measure general accuracy, prove that the relationship is causal, or show that a straight line is the right model.

Yes. Add up to four series and give each a name, color, marker, and its own X,Y pairs. The chart keeps the groups on the same axes while calculating a separate line and statistics for each series, which is useful for control, treatment, cohort, or instrument comparisons.

An unusual point can pull the slope, intercept, correlation, and R² toward itself, especially when it has an extreme X value. Outlier Impact compares the fit with and without a candidate point. Treat that comparison as a diagnostic and inspect the original measurement before excluding anything.

Avoid interpreting a straight line when the points form a clear curve, separate clusters, a funnel, repeated bands, or another structured pattern. Residual Diagnostics plots the remaining error against X and flags strong curvature or changing spread as a practical warning that the linear model may be incomplete.

A residual is the observed Y value minus the Y value predicted by the line. A loose, patternless cloud around zero supports the linear fit; visible curves or changing spread suggest the line is leaving structure unexplained. The automatic message is a screening aid rather than a formal statistical test.

Copy two adjacent numeric columns and paste them into the X,Y pairs box; tab-separated spreadsheet rows are recognized automatically. Comma-, semicolon-, and space-separated pairs also work. Data Health identifies incomplete or nonnumeric rows, while parsing and regression calculations stay inside your browser.