Believing you are above average on most positive traits — smarter, kinder, more skilled — even when that's statistically impossible for everyone.
Imagine a classroom where the teacher asks every kid to raise their hand if they think they're a better-than-average student. Almost every hand goes up. But 'average' means half the kids should be below it — so a lot of those raised hands are wrong. That's illusory superiority: almost everyone thinks they're in the top half, even though that's impossible for everyone at once.
Illusory superiority is a pervasive self-enhancement bias in which individuals systematically rate themselves as superior to the average person across a wide range of attributes — intelligence, driving skill, health behaviors, moral character, interpersonal skills, and professional competence. The effect is robust across many life domains and persists even when people are given specific, well-defined attributes to evaluate rather than vague ones. Importantly, the bias is not simply overconfidence in an absolute sense but specifically involves inflated comparative self-assessments: people believe they are better than most others, which is mathematically impossible for the majority of any population. The strength of the bias varies with how personally important the trait is to the individual and is modulated by cultural norms around self-enhancement versus modesty.
The same glitch looks different depending on the terrain. Finance, medicine, a relationship, a team — same mechanism, different costume.
Investors consistently overrate their stock-picking ability relative to other market participants, leading to excessive trading volume and below-market returns. This pattern persists even among professionals, with the majority of fund managers believing they will outperform benchmarks despite evidence that most do not.
Physicians tend to overestimate their diagnostic accuracy relative to peers, which can reduce motivation to seek second opinions or consult decision-support tools. Patients similarly overestimate their adherence to treatment plans and healthy behaviors compared to others, leading to underreporting of risk factors.
The vast majority of professors rate their teaching as above average, creating resistance to pedagogical training and peer evaluation. Students likewise overestimate their understanding of material relative to classmates, which can reduce study effort and produce surprise at poor exam performance.
Partners tend to believe they contribute more to household labor, emotional support, and relationship maintenance than their significant other does. This asymmetric perception of contribution is a common source of resentment and conflict in relationships.
Developers and designers frequently overestimate the usability and quality of their own code or interfaces compared to industry benchmarks, leading teams to skip user testing or dismiss usability feedback. Product managers may believe their feature prioritization instincts are superior to data-driven approaches.
Employees consistently rate their own performance, teamwork ability, and leadership skills above those of their colleagues. This inflates expectations around promotions and compensation, and creates friction when feedback or rankings do not match self-perception.
Citizens tend to believe they are better than average at detecting misinformation, political bias, and propaganda, which paradoxically makes them less vigilant consumers of news. This self-assessed superiority in media literacy correlates with reduced fact-checking behavior.
The term 'illusory superiority' was coined by Van Yperen and Buunk in 1991. The phenomenon was documented as early as 1976 by the U.S. College Board. Ola Svenson's 1981 driving study became a landmark demonstration. Vera Hoorens provided a comprehensive theoretical overview in 1993, and Kruger and Dunning (1999) identified a specific subtype in low-competence individuals.
Moderately positive self-illusions likely provided survival advantages by promoting the confidence needed to pursue mates, compete for resources, take calculated risks, and persist through adversity. Ancestral individuals who slightly overestimated their abilities may have been more likely to attempt challenging hunts, assert dominance in social hierarchies, and recover from setbacks — all traits that would enhance reproductive fitness in uncertain environments.
Language models trained on human-generated text can inherit the better-than-average framing embedded in how people describe themselves and their work, potentially generating overly positive self-assessments or recommendations. Recommendation algorithms that rely on user self-reports inherit inflated competence ratings, skewing personalization. AI hiring tools trained on self-evaluation data may calibrate poorly if the training signal is systematically biased upward by applicants' illusory superiority.
50 cards are free to preview. Buyers unlock the rest of the deck plus the interactive training — Spot-the-Bias Quiz unlimited, Swipe Deck with spaced repetition, My Blindspots, Decision Pre-Flight, the Printable Deck + Cheat Sheets, and the Field Guide e-book. $29.50$59.
