---
title: "Friday the 13th and the illusion of a pattern"
url: https://erroneo.us/daily/friday-the-13th-illusory-correlation/
description: "Large studies find no extra harm on Friday the 13th. The 'bad luck' is illusory correlation: you notice mishaps on the 13th and forget the other Fridays."
type: daily-entry
section: "Thinking"
date_published: 2026-11-13
author: erroneo.us Editors
date_modified: 2026-11-13
publisher: erroneo.us
license: All rights reserved. Quote with attribution and a link.
---

# Friday the 13th and the illusion of a pattern

> Today is a Friday the 13th. Whatever goes wrong will feel like proof. Here's why that feeling is a predictable glitch, and what the studies actually show.

*Error of the Day for November 13, 2026 · Thinking · by erroneo.us Editors*

- **Wrong:** Of course the printer jammed. It's Friday the 13th.
- **Right:** The printer jams on plenty of Fridays. You only keep score on the 13th.
- **Memory trick:** A pattern only counts if you also tallied the days nothing happened.

## What people get wrong

Today is Friday, November 13. Somewhere a coffee will spill, a train will be late and a laptop will refuse to update, and someone will say, "Well, it *is* Friday the 13th." The error isn't the superstition itself. It's the evidence people think they have for it.

Psychologists call this **illusory correlation**: seeing a relationship between two things when there isn't one, or seeing a much stronger one than exists. Loren Chapman described the effect in a 1967 paper, ["Illusory correlation in observational report"](https://doi.org/10.1016/S0022-5371(67)80066-5). A decade later, David Hamilton and Robert Gifford showed how the same mechanism helps produce [stereotypes about groups](https://doi.org/10.1016/S0022-1031(76)80006-6). Calendar superstitions are a gentler case of it.

## Why it happens

To know whether Friday the 13th is unlucky, you'd need four numbers: bad things on the 13th, good or neutral things on the 13th, bad things on other Fridays, and good or neutral things on other Fridays. Most of us only keep one: the bad things that happened on the 13th. Those are memorable because they "fit." A jammed printer on Friday the 6th isn't a story, so it's forgotten.

Expectation then fills in the rest. You go into the day watching for trouble, find some (there's always some), and file it as confirmation.

## What the studies show

The most quoted study is a cheerful one. A 1993 paper in the *BMJ*, ["Is Friday the 13th bad for your health?"](https://pmc.ncbi.nlm.nih.gov/articles/PMC1697765/), compared Fridays the 13th with Fridays the 6th in one region of England. It found fewer cars on part of the M25 motorway on the 13th, but more hospital admissions from transport accidents: 65 versus 45. The authors concluded, deadpan, that "staying at home is recommended." The numbers were small and the paper ran in the journal's year-end holiday issue, but it's still cited as if it settled the matter.

Larger studies haven't backed it up. A Finnish study of traffic deaths from 1971 to 1997 reported a higher risk for women on Friday the 13th. [Radun and Summala (2004)](https://bmcpublichealth.biomedcentral.com/articles/10.1186/1471-2458-4-54) then checked the much larger national database of injury crashes for 1989 to 2002, comparing 21 Fridays the 13th with the Fridays before and after. They found "no significant differences in any examined aspect," for women or men. They did add a fair caveat: without data on who was actually on the road, anxious people staying home could mask a small effect.

Hospitals have tested it too. A population-based study of 19,747 adults who had common operations on Friday the 13th or on the flanking Fridays the 6th and 20th ([Ranganathan et al., 2024](https://pmc.ncbi.nlm.nih.gov/articles/PMC11175938/)) found that patients operated on the 13th did not fare worse. Rates of death, readmission and complications at 30 days, 90 days and one year were essentially the same.

## The fix

Wrong: I got a flat tire today, ~~which proves Friday the 13th is cursed~~.

Right: I got a flat tire today, which proves tires go flat; I don't count the Fridays they didn't.

Why: One memorable hit isn't a correlation. You need the misses too.

- **Fill in the whole table.** When you suspect a pattern, ask what happens on the days you *aren't* watching. Lucky socks, full moons and "this always happens when I wash the car" fail the same test.
- **Watch out for small samples.** 65 versus 45 sounds dramatic until you remember it came from one region and a handful of days.
- **Notice when expectation is doing the looking.** If you went searching for bad luck, finding some tells you nothing.

Enjoy the day. If something goes wrong, blame the usual suspects: traffic, software updates and the law of large numbers. More of the brain's favorite shortcuts are in our [cognitive biases](https://erroneo.us/thinking/cognitive-biases/) guide.

## Sources

1. [Journal of Verbal Learning and Verbal Behavior: Chapman, Illusory correlation in observational report (1967)](https://doi.org/10.1016/S0022-5371(67)80066-5)
2. [Journal of Experimental Social Psychology: Hamilton & Gifford, Illusory correlation in interpersonal perception (1976)](https://doi.org/10.1016/S0022-1031(76)80006-6)
3. [BMJ via PMC: Scanlon et al., Is Friday the 13th bad for your health? (1993)](https://pmc.ncbi.nlm.nih.gov/articles/PMC1697765/)
4. [BMC Public Health: Radun & Summala, Females do not have more injury road accidents on Friday the 13th (2004)](https://bmcpublichealth.biomedcentral.com/articles/10.1186/1471-2458-4-54)
5. [Annals of Surgery Open via PMC: Ranganathan et al., Superstition in Surgery (2024)](https://pmc.ncbi.nlm.nih.gov/articles/PMC11175938/)

## Go deeper

- https://erroneo.us/thinking/cognitive-biases/

---
Source page: https://erroneo.us/daily/friday-the-13th-illusory-correlation/ · More: https://erroneo.us/llms.txt
