🧠 Rarest Personality Types

📐 Why You Cannot Multiply the Four Letters

The standard explanation for INFJ's 1.5% is a stack of "four less-common preferences." Two of them are not. In the Myers & Briggs Foundation's US estimates the four scales split Introversion 50.7%, Intuition 26.7%, Feeling 59.8% and Judging 54.1%, so Feeling and Judging are the majority sides and only Intuition is genuinely scarce. INFJ is rare despite two common letters, which is why the obvious arithmetic fails.

Multiply those four shares and INFJ comes out at 4.4%, nearly three times the published figure; the same sum gives ISFJ 12.0% against a published 13.8%. Add up the sixteen published type percentages one scale at a time and you get back 50.9% Introverted, 26.9% Intuitive, 59.9% Feeling and 54.3% Judging, which match the scale figures to within rounding. What multiplication throws away is the link between scales: Sensing goes with Judging and Intuition with Perceiving, so SJ and NP types are more common than chance predicts, and NJ and SP types less common. Treating the letters as independent therefore overstates all eight NJ and SP types and understates the other eight. The four rarest types (INFJ, ENTJ, INTJ and ENFJ) are all NJ, so the shortcut inflates all four, INFJ by the widest margin of all sixteen.

👑 ENTJ Women Are Tied for Rarest, Not Alone

The 0.9% for ENTJ women comes from a split of the same US estimates by sex. That split puts INTJ women at 0.9% as well, so by its own source ENTJ women are at best tied for rarest.

The tie matters less than the sample size. The study usually cited for these figures, Hammer and Mitchell's 1996 paper in the Journal of Psychological Type, surveyed 1,267 US adults, so each of these cells holds roughly a dozen women or fewer and a tenth of a percentage point is about one person. A 0.9% reading there is compatible with a true rate anywhere from 0.4% to 2%. The bottom of that ranking is too close to call: these types are clearly rare, but their order is noise.

🧮 The Big Five Figure Contradicts Its Own Definition

The familiar Big Five figure and its own definition cannot both be right. If scoring in the top 5% on Openness had nothing to do with scoring in the bottom 5% on Neuroticism, the two would overlap in 0.05 × 0.05 = 0.25% of people. That is one in 400, not the one in 33 that "roughly 3%" implies. The two traits are in fact only weakly related: a meta-analysis of 212 samples covering 144,117 people put the corrected Openness-Neuroticism correlation at just -0.17, lifting the expected overlap to about half a percent. Reaching a genuine 3% at those cut-offs would take a correlation near 0.88, close to collapsing the two traits into one. For unrelated traits, 3% would instead fit cut-offs near the top and bottom sixth of each scale. Either the definition or the number has to change.

⚖️ One Number Was Sampled, the Other Volunteered

The MBTI's 1.5% and the Enneagram's 4.8% are quoted in the same breath but rest on unequal evidence. The MBTI figure goes back to a published survey of US adults. For the Enneagram, no nationally representative survey of type frequencies turns up in the research literature. The distributions in circulation come from people who chose to take an online test, so each describes a site's own visitors. Read 4.8% as a rough ranking, not a precise measurement.

Your own type is also less fixed than any of these tables implies. Psychologist David Pittenger's 1993 review reported that 39% to 76% of people were given a different four-letter type when retested after only about five weeks. McCrae and Costa's 1989 study in the Journal of Personality found no support for the idea that the MBTI sorts people into truly distinct types. Someone who scores near the middle on Feeling versus Thinking can come out INFJ on one sitting and INTJ on the next: 1.5% one day, 2.1% the other. The finding that these types are rare holds up far better than any one person's claim to belong to one.