Interactive Scatter Plots

When to reach for a scatter plot

A scatter plot answers “how do these two numeric variables relate?” Each point is one row of data, positioned by two measurements at once. Add color for a third variable (a category) and size for a fourth (a number) — a scatter plot can honestly encode more information per point than almost any other chart type.

Basic scatter

Interactive scatter

import pandas as pd
import plotly.express as px
 
df = pd.DataFrame({
    "hours": [1, 2, 3, 4, 5, 6, 7, 2, 4, 6],
    "score": [50, 55, 65, 70, 75, 78, 85, 52, 69, 77],
    "group": ["A", "A", "A", "A", "A", "A", "A", "B", "B", "B"],
})
 
fig = px.scatter(df, x="hours", y="score", color="group", title="Hours vs score")
fig.show()

Every point is hoverable by default — Plotly shows x, y, and color for whichever point the cursor is nearest, with zero extra code.

Add trend line

Trend line

import plotly.express as px
 
fig = px.scatter(df, x="hours", y="score", color="group", trendline="ols", title="Trend line")
fig.show()

trendline="ols" fits an ordinary-least-squares regression line per color group and draws it on top — a quick way to see if a relationship looks roughly linear (requires statsmodels installed).

sketch Scatter plot — hover finds the nearest point p5.js Pause Restart

Move your mouse near any dot; the closest point locks on and shows its (x, y) values, like a Plotly tooltip.

Tips

• Use hover_data=[...] to show more context (e.g. a name or ID column) in the tooltip.
• Trend lines can be sensitive to outliers — a single extreme point can tilt the whole fitted line.
• If points overlap heavily, lower opacity or switch to a 2D density/contour plot.

Analyst’s tip

Correlation shown by a trend line is not causation — and a fitted line can look convincing even on random data with a small sample size. Always check the point cloud around the line, not just the line itself, before trusting the relationship it implies.

diagram Reading a scatter plot mermaid

  flowchart LR
  A["One data row"] --> B["x position
(variable 1)"]
  A --> C["y position
(variable 2)"]
  A --> D["color
(category, optional)"]
  A --> E["size
(magnitude, optional)"]
  B & C & D & E --> F["One point,
hoverable for exact values"]

Each visual channel of a scatter point can carry a different variable.

Next

Continue to: Bubble Charts to add a size dimension to your scatter plots.

🧪 Try It Yourself

Exercise 1 – Plot Two Variables

pch.exerciseTag pch.exerciseTitle

Exercise 2 – Color by Group

pch.exerciseTag pch.exerciseTitle

Exercise 3 – Add a Trend Line

pch.exerciseTag pch.exerciseTitle