Self-attribution bias
Also called self-serving bias
Crediting successes to one's own skill and blaming failures on bad luck or outside factors. It feeds overconfidence because winning trades seem to prove ability.

- Level
- Intermediate
- Reading
- 2 min
- Sources
- 3RegulatorResearch
- Updated
On this page· 5 sections
The winning trade was a good read. The losing one was a fake-out, a news shock, a bad fill. Each explanation can be true on its own. Applied every time, the pair builds a record in which skill never loses.
That split has a name. The FCA describes self-attribution bias as the tendency to credit successes to one’s own ability and blame failures on bad luck, and lists it as one source of overconfidence[1].
How it works
Psychologists documented the pattern in the 1970s[3]. People tend to credit successes to their abilities and failures to bad luck or the actions of others, and they make self-enhancing explanations after success more often than self-protective ones after failure[3].
Simon Gervais and Terrance Odean showed in 2001 how this can make a trader overconfident. In their model a trader does not know their own ability and infers it from successes and failures[2]. By taking too much credit for the successes, the trader becomes overconfident[2]. Expected overconfidence rises in the early stages of a career; with more experience, the trader comes to recognize their ability better[2].
Brad Barber and Terrance Odean then looked at what this might mean for real accounts. They studied 1,607 investors who switched from phone-based to online trading, drawn from 78,000 households at a large discount broker, with records from January 1991 to December 1996[3]:
- before going online, they beat the market by more than 2% a year[3];
- after going online, they traded more actively and more speculatively, and lagged the market by more than 3% a year[3].
The authors’ reading: a run of good results, credited to skill, fosters overconfidence, which leads to more trading and weaker results[3].
Example
Say a trader closes 20 trades: 11 wins and 9 losses.
- All 11 wins get filed under skill.
- Of the 9 losses, 7 get filed under bad luck and 2 under the trader’s own mistakes.
- The “skill record” now reads 11 wins and 2 losses: 11 ÷ 13 = about 85%.
- The actual record is 11 ÷ 20 = 55%.
Nothing in the trades changed. Only the filing did, and the filing is what the trader learns from.
Watch out
Early success is the risky stretch. In Gervais and Odean’s model, overconfidence peaks early, before experience corrects it[2].
Luck works in both directions. The bias lies in the asymmetry: bad luck is invoked for failures, while successes go to ability[1].
It may not be the whole story. Barber and Odean report that self-attribution alone could not explain the results of the online investors, and point to illusions of knowledge and control as well[3].
Sources
- RegulatorFinancial Conduct AuthorityApplying behavioural economics at the Financial Conduct Authority (Occasional Paper No. 1, April 2013)Checked November 9, 2026
- ResearchOxford University Press (Review of Financial Studies)Learning to Be Overconfident (Gervais & Odean, Review of Financial Studies 14(1), 2001)Checked November 9, 2026
- ResearchTerrance Odean, UC Berkeley Haas (author copy)Online Investors: Do the Slow Die First? (Barber & Odean, Review of Financial Studies 15(2), 2002)Checked November 9, 2026
