Home Theories Regressive bias

Regressive bias

Reviewed by expert Scientifically proven

High values and possibilities are overvalued (overestimated), whereas low values and possibilities are discounted (regressive bias). This ideological bias may be seen in advertising new products that may or may not appeal to the broader public. The regressive or regression fallacy is an informal fallacy. It implies that something has returned to normal due to corrective actions taken while it was abnormal. Variations in the natural world aren't taken into account. Almost often, it's a kind of post hoc fallacy.

Table of contents:
  1. Explanation
  2. Examples
  3. Misapplication
  4. Regressive bias example


Golf scores, global temperatures, and persistent back pain are all-natural variations that revert to the mean. A logical fallacy is making assumptions that expect exceptional achievements to continue as if they were typical (see Representativeness heuristic). People are most likely to act when variation is at its peak. When the findings return to normal, people believe their behaviours were to blame for the shift when they were not.

In an article titled "Regression Toward Mediocrity in Hereditary Stature," published in 1885, Sir Francis Galton coined "regression." He proved that children born to be very short or very tall grew up to standard height. A stellar performance, on the other, may not match a fantastic performance on one variable in any situation where two variables are not connected. Because height is not 100% heritable, the distribution of their children's heights will be centred somewhere between the parents' average and the population's average. As a result, even if a child is more radical than their parents, the odds are set against them.


When his suffering grew excruciating, he went to the doctor, and the aching began to disappear. As a consequence, he benefited from the doctor's therapy. Pain diminishing after it has grown worse is best explained by regression toward the mean. It's a mistake to presume the discomfort was caused by the doctor. I docked the student's grade because of his poor performance last semester. He did substantially better this semester. Punishment is effective in improving students' grades. Regression toward the mean may better explain the shift in performance because more normal ones typically follow remarkable successes. As a result of reasoning similar to this regression fallacy example, multiple studies have shown that people may develop a systematic bias for punishment and against reward.

The number of accidents on the road fell once a speed camera was installed. As a result, the speed camera has aided in the enhancement of highway safety. Speed cameras are usually implemented when a route has many incidents, and the value soon diminishes (regression to mean). Many supporters of speed cameras attribute the reduction in accidents to the cameras without considering the overall trend. According to some authors, extraordinary feats are likely to be followed by less extreme ones, and athletes are chosen to appear on the cover of Sports Illustrated only after extreme performances. As some athletes appear to believe, the regression fallacy can be perpetrated by attributing this to a "jinx" rather than regression.


On the other side, dismissing legitimate grounds may compound the situation. For example, once the Western Allies stormed Normandy, opening a second central front, German supremacy in Europe waned. The Western Allies and the Soviet Union collaborated to drive the Germans back.

Fallacious evaluation: "Regression toward the mean can explain the withdrawal of German forces from occupied territories as a purely random fluctuation that would have occurred without any intervention on the part of the USSR or the Western Allies, given that counterattacks against Germany occurred only after they had conquered the greatest amount of territory under their control." This, however, was not the case. This is because political power and territorial occupancy are not primarily driven by random occurrences, rendering the regression toward the mean idea useless (on a large scale). In other words, if regression toward the mean is misused, all occurrences can be reduced to a just-so tale with no cause or effect.

Regressive bias example

We have a tendency to remember low values higher than they actually were. In this example Sephora is stating we often miss our foundation actual color. This is meant to have us question ourselves about it. Because of the regressive bias, chances are, we will overestimate the number of time this happens and click through the ban.

Regressive bias is used by Sephora as can be seen on the screenshot

A new cognitive bias in your inbox every week

You'll get to learn more about CRO and psychology. You'll be able to take experimenting to a whole new level!

* We send our mails on Monday morning btw.

Example of hypothesis applied

If we exaggerate one of our users habit, then click though rate for banner on all devices will increase, because of Regressive bias.

Frequently Asked Questions

Will you use psychology for your experimentation process?

Are you curious about how to apply this bias in experimentation? We've got that information available for you!

Join over 452+ users

  • Lifetime access to all biases
  • Filter on metrics, page type, implementation effort
  • More examples and code for experimentation

Choose your subscription!

Pay with Stripe

Lifetime deal PREMIUM

Get access to the search engine, filter page, and future features.

I want this

Lifetime deal

Get access to the search engine and filter page.

I want this

Do you think you know enough about CRO?

Join our monthly mailing to continue learning more and more about CRO and psychology.