Ridge PlotStacked Density PlotArea Plot
About Ridge Plot
Ridge plots — also known as ridgeline plots or joy plots, named after the iconic Joy Division album cover designed by Peter Saville — stack multiple density curves along a shared axis, creating a layered mountain-range effect that reveals how distributions shift, spread, or cluster across dozens of categories at a glance. Unlike small multiples or faceted histograms, ridge plots overlap their curves just enough to save space while preserving each distribution's shape, making them one of the most visually striking and information-dense ways to compare distributions in scientific publications, data journalism, and exploratory analysis.
Ridge plots excel when you need to show temporal evolution (e.g., monthly temperature distributions over a year), group comparisons (e.g., gene expression across cell types), or any scenario where the audience must quickly spot shifts in modality, skewness, or variance across an ordered set of categories. Their compact, self-explanatory layout makes them a favorite in fields ranging from climate science and genomics to economics and social sciences.
Browse our curated collection of ridge plot examples drawn from peer-reviewed, open-access literature to find inspiration for your next figure. When you're ready to create your own, head over to ai.plottie.art — Plottie's AI-powered editor lets you generate publication-quality ridge plots in seconds by describing what you need in plain language.
When to Use Ridge Plot
- Comparing distributions across many categories or groups (e.g., expression levels across 20+ cell types) where individual violin or box plots would be too crowded
- Visualizing how a distribution evolves over time, such as daily stock-return distributions month by month or seasonal climate patterns
- Highlighting shifts in modality, skewness, or spread across an ordered variable like experimental conditions or geographic regions
- Presenting survey-response distributions across multiple demographic segments in a compact, reader-friendly layout
- Replacing faceted histograms or small-multiple density plots when vertical space is limited and overlap between curves aids comparison
Key Features
- Stacked, partially overlapping density curves that create a layered mountain range effect for intuitive visual comparison
- Shared horizontal axis ensures all distributions are plotted on the same scale, enabling direct comparison of location and spread
- Category ordering along the vertical axis can encode meaningful variables like time, latitude, or experimental dose
- Compact layout that scales gracefully from a handful of groups to 50+ categories without overwhelming the reader
- Overlap between adjacent ridges is adjustable, letting you balance readability against space efficiency
- Color mapping can encode an additional variable — such as mean value or group membership — adding a fourth dimension to the visualization
Frequently Asked Questions
What is the difference between a ridge plot and a violin plot?
A violin plot shows a mirrored density curve for each category side by side, while a ridge plot stacks single-sided density curves vertically with partial overlap. Ridge plots scale better when you have many categories, whereas violin plots are often preferred for a smaller number of groups where precise width comparison matters.
Why are ridge plots sometimes called joy plots?
The name joy plot comes from the cover art of Joy Division's 1979 album Unknown Pleasures, which featured stacked radio-pulse waveforms in a style visually identical to modern ridge plots. The term ridgeline plot is now more commonly used in academic contexts to keep the naming neutral and descriptive.
How do I choose the right bandwidth for a ridge plot?
Bandwidth controls how smooth or detailed each density curve appears. A bandwidth that is too narrow produces noisy, jagged curves, while one that is too wide can obscure important features like bimodality. Start with the default kernel-density estimate and adjust iteratively, checking that genuine peaks are preserved without introducing artifacts.
Can I use ridge plots for discrete or categorical data?
Ridge plots are designed for continuous distributions, but you can adapt them for discrete data by using a narrow Gaussian kernel or by plotting histograms instead of smooth densities along each ridge. For purely categorical counts, however, a grouped bar chart or heatmap is usually a clearer choice.