What is regression to the mean?

Regression to the mean is the statistical phenomenon where extreme measurements on one variable tend to be followed by less extreme measurements on a second, correlated variable. It was first described by Francis Galton in 1886 when studying the heights of parents and children.

Why does regression to the mean happen?

It happens because extreme scores are partially due to chance (noise). When you select for extreme values on one measurement, you are also selecting for above-average luck. On the next measurement, luck is likely to be closer to average, pulling the result back toward the mean. The mathematical formula is E[Y|X=x] = mu_Y + rho*(sigma_Y/sigma_X)*(x - mu_X).

What is the sports curse and how does regression explain it?

The Sports Illustrated curse (or sophomore slump) describes athletes who perform exceptionally well, gain fame, then appear to decline. In reality, their first performance was partly luck, and their second is closer to their true ability. Regression to the mean, not a real curse, explains this pattern entirely.

How does correlation rho affect regression to the mean?

The strength of regression to the mean depends directly on the correlation rho. When rho = 1 (perfect correlation), there is no regression. When rho = 0 (no correlation), extreme first scores always regress fully back to the population mean.

Where does regression to the mean appear in medicine?

Patients often seek treatment when symptoms are at their worst. Even without treatment, symptoms would tend to improve due to regression to the mean. This is why clinical trials need control groups.

What is a bivariate normal distribution?

A bivariate normal distribution describes two jointly normally distributed random variables X and Y with means, standard deviations, and correlation coefficient rho. The conditional expectation E[Y|X=x] forms a straight line, and the slope rho*(sigma_Y/sigma_X) determines the degree of regression.

Can regression to the mean cause false conclusions in research?

Yes. If researchers select the lowest-performing students for an intervention and then see improvement, that improvement might be entirely due to regression to the mean rather than the program itself.

How is regression to the mean related to linear regression?

The term regression in linear regression comes directly from Galton's original work. He noticed that the best-fit line relating parent heights to child heights had a slope less than 1, showing that tall parents had children closer to the average height.

Does regression to the mean only go downward?

No. Regression to the mean works symmetrically: extremely low scores on variable X tend to be followed by less extreme (higher) scores on Y. If you select the worst students, they will tend to improve due to regression, not their effort.

How can I avoid being misled by regression to the mean?

Use randomised controlled trials with proper control groups. Avoid selecting subjects based on extreme pre-test scores without accounting for regression effects. Use statistical methods that model measurement error explicitly.

Why does a lower correlation rho produce stronger regression to the mean?

The purple regression line drawn on the scatter plot is E[Y given X] = rho times X, so its slope is exactly rho. At rho = 1 this line coincides with Y=X and there is no regression at all; as rho shrinks toward 0, the line flattens toward the horizontal, meaning any extreme X value predicts a Y value pulled almost all the way back to the overall mean of zero.

Home▸Statistics▸Regression to the Mean

📉 Regression to the Mean

Explore why extreme performances tend to be followed by more average ones. From sports to medicine, this statistical phenomenon shapes decisions. Interactive bivariate normal scatter plot.

Statistics3DEasy60 FPS

About this simulation

This simulation generates N pairs of correlated scores (x, y) from a bivariate normal distribution with correlation ρ, using a Box-Muller transform to produce standard normal draws and y = ρ·x + √(1−ρ²)·z. It then selects the top-performing slice on X (by threshold %) and shows how their average Y score falls back toward the population mean — the real statistical phenomenon behind sports slumps, medical placebo effects, and mistaken program evaluations.

🔬 What it shows

A scatter plot (or histogram) of period-1 versus period-2 scores for up to 600 simulated individuals, with the top-X% performers on X highlighted in yellow. A purple line plots the true conditional expectation E[Y|X] = ρ·X, a grey dashed line marks Y=X for comparison, and a red arrow shows the gap between the selected group's average X and their actual average Y — the regression effect itself.

🎮 How to use

Drag Correlation ρ (−1 to 1) to change how strongly period-1 and period-2 scores relate, Sample size N (30–600) to change how many points are drawn, and Selection threshold (5–50%) to change how extreme the highlighted group is. Switch between Scatter, Histogram and All views, hit Resample for a fresh random draw, or load the Sports Curse, Medicine and No Correlation presets to see pre-tuned examples of the effect.

💡 Did you know?

Francis Galton first documented this effect in 1886 studying parent and child heights, calling it "regression towards mediocrity" — the origin of the word "regression" in statistics. When ρ = 0 (the No Correlation preset), any group selected for extreme X scores regresses all the way back to the population mean on Y, because the two measurements share no real information at all.

Frequently asked questions

How does the simulation actually generate the correlated data?

It draws two independent standard normal values z1 and z2 using a Box-Muller transform, sets x = z1, and sets y = ρ·z1 + √(1−ρ²)·z2. This is the standard construction for a bivariate normal pair with correlation ρ: when ρ is close to 1, y tracks x almost exactly, and when ρ is close to 0, y is essentially independent noise.

How is the "selected" group and the regression arrow computed?

The simulation sorts all N points by their X value descending and takes the top slice defined by the Selection threshold slider (5–50%). It then computes that group's average X and average Y, and draws a red arrow from the point (avgX, avgX) on the Y=X identity line down to the actual point (avgX, avgY) — visually showing how far the group's real Y average falls short of matching its inflated X average.

Why does a lower correlation ρ produce stronger regression to the mean?

The purple regression line drawn on the scatter plot is E[Y|X] = ρ·X, so its slope is exactly ρ. At ρ = 1 this line coincides with Y=X and there is no regression at all; as ρ shrinks toward 0, the line flattens toward the horizontal, meaning any extreme X value predicts a Y value pulled almost all the way back to the overall mean of zero.

What do the Sports Curse, Medicine and No Correlation presets model?

Sports Curse sets ρ = 0.5, N = 300 and a tight 10% selection threshold, mimicking how only truly outstanding single-season performances get media attention and then regress hard the next season. Medicine uses ρ = 0.7, N = 150 and a 30% threshold, closer to how patients selecting into treatment at symptom peaks naturally improve afterward. No Correlation sets ρ = 0 to show the extreme case where selected performers regress completely to the population average.

Is the regression effect shown here a real causal effect or an artifact?

It is a statistical artifact of selecting on one noisy measurement, not a real force pulling scores down. The simulation's own info panel states this directly: any extreme score partly reflects genuine ability and partly reflects luck, and since luck by definition does not repeat, the next measurement naturally looks more average even though nothing about the underlying ability changed. This is exactly why the model computes a "Regression %" statistic rather than attributing the change to any real intervention.

Frequently Asked Questions

What is regression to the mean?

Regression to the mean is the statistical phenomenon where extreme measurements on one variable tend to be followed by less extreme measurements on a second, correlated variable. It was first described by Francis Galton in 1886 when studying the heights of parents and children.
Why does regression to the mean happen?

It happens because extreme scores are partially due to chance (noise). When you select for extreme values on one measurement, you are also selecting for above-average luck. On the next measurement, luck is likely to be closer to average, pulling the result back toward the mean. The mathematical formula is E[Y|X=x] = μ_Y + ρ·(σ_Y/σ_X)·(x − μ_X).
What is the sports curse and how does regression explain it?

The “Sports Illustrated curse” (or sophomore slump) describes athletes who perform exceptionally well, gain fame, then appear to decline. In reality, their first performance was partly luck, and their second is closer to their true ability. Regression to the mean, not a real curse, explains this pattern entirely.
How does correlation ρ affect regression to the mean?

The strength of regression to the mean depends directly on the correlation ρ. When ρ = 1 (perfect correlation), there is no regression — the second measurement equals the first exactly. When ρ = 0 (no correlation), the second measurement is completely independent, and extreme first scores always regress fully back to the population mean.
Where does regression to the mean appear in medicine?

Patients often seek treatment when symptoms are at their worst. Even without treatment, symptoms would tend to improve due to regression to the mean. This is why clinical trials need control groups — apparent improvement after treatment may simply be regression, not a real drug effect.
What is a bivariate normal distribution?

A bivariate normal distribution describes two jointly normally distributed random variables X and Y with means μ_X, μ_Y, standard deviations σ_X, σ_Y, and correlation coefficient ρ. The conditional expectation E[Y|X=x] forms a straight line, and the slope ρ·(σ_Y/σ_X) determines the degree of regression.
Can regression to the mean cause false conclusions in research?

Yes. If researchers select the lowest-performing students for an intervention and then see improvement, that improvement might be entirely due to regression to the mean rather than the program itself. Pre-post designs without control groups are especially vulnerable to this confound.
How is regression to the mean related to linear regression?

The term “regression” in linear regression comes directly from Galton’s original work on regression to the mean. He noticed that the best-fit line relating parent heights to child heights had a slope less than 1, showing that tall parents had children closer to the average height than themselves.
Does regression to the mean only go downward?

No. Regression to the mean works symmetrically: extremely low scores on variable X tend to be followed by less extreme (higher) scores on Y. If you select the worst students, they will tend to improve — not because of their effort, but because their poor initial performance included bad luck.
How can I avoid being misled by regression to the mean?

Use randomised controlled trials with proper control groups. Avoid selecting subjects based on extreme pre-test scores without accounting for regression effects. Use statistical methods that model measurement error explicitly, such as regression dilution correction or structural equation models.

⚙ Under the hood

Pick extreme performers from a correlated pair of scores and watch their follow-up drift back toward average, the trap behind many sports curses.

regression to the meanGaltoncorrelationsports cursestatisticsbivariate normal

3D · Three.js / WebGL renderer · 60 FPS target · runs fully client-side, no install