An IQ percentile tells you the share of the population that scores below you: an IQ of 100 is the 50th percentile, 115 is the 84th, 130 is the 98th, and 145 is the 99.9th. Because every modern IQ test is standardized to a normal distribution with a mean of 100 and a standard deviation of 15, any score can be converted to a percentile with simple math. The full conversion table below covers IQ 70 through 160, followed by an explanation of how the numbers are computed and why percentiles say more than the raw score ever can.
IQ to percentile conversion table
The table assumes the standard scale used by the WAIS, Stanford-Binet and most other modern tests: mean 100, standard deviation (SD) 15. The rarity column shows roughly how many people you would need to gather before expecting one person at or above that score.
| IQ score | Percentile | Rarity (at or above) |
|---|---|---|
| 70 | 2.3rd | — |
| 75 | 4.8th | — |
| 80 | 9.1st | — |
| 85 | 15.9th | — |
| 90 | 25.2nd | — |
| 95 | 36.9th | — |
| 100 | 50th | 1 in 2 |
| 105 | 63.1st | 1 in 3 |
| 110 | 74.8th | 1 in 4 |
| 115 | 84.1st | 1 in 6 |
| 120 | 90.9th | 1 in 11 |
| 125 | 95.2nd | 1 in 21 |
| 130 | 97.7th (~98th) | 1 in 44 |
| 135 | 99.0th | 1 in 102 |
| 140 | 99.6th | 1 in 261 |
| 145 | 99.9th | 1 in 741 |
| 150 | 99.96th | 1 in 2,331 |
| 155 | 99.99th | 1 in 8,139 |
| 160 | 99.997th | 1 in ~31,600 |
If you want the classification labels that go with each band — "high average", "superior" and so on — see the IQ score chart, which pairs every range with its percentile and description.
How percentiles are computed from the bell curve
IQ scores follow (by design) a normal distribution — the familiar bell curve. Two numbers fully describe it: the mean (100) and the standard deviation (15). To convert a score to a percentile, you first express it as a z-score: how many standard deviations it sits from the mean. An IQ of 130 is (130 − 100) ÷ 15 = 2 standard deviations above average. The cumulative distribution function of the normal curve then gives the share of the population below that point — for z = 2, about 97.7%.
A few landmarks worth memorizing: roughly 68% of people score within one SD of the mean (IQ 85–115), and about 95% score within two SDs (IQ 70–130). That leaves only around 2.3% above 130 and 2.3% below 70. Because the curve is symmetric, an IQ of 85 (16th percentile) is exactly as far from average as an IQ of 115 (84th percentile).
How rare is a high IQ? The 1-in-X view
Percentiles compress dramatically at the top of the curve, so the rarity framing often communicates better than yet another decimal place:
- IQ 130 — 98th percentile, about 1 in 44 people. The classic gifted and Mensa threshold.
- IQ 140 — 99.6th percentile, about 1 in 261. The pop-culture "genius" line, though psychologists don't use that label formally.
- IQ 145 — 99.9th percentile, about 1 in 741. Three full standard deviations above the mean.
- IQ 160 — 99.997th percentile, roughly 1 in 31,600. At this altitude, standard tests run out of norming data and scores become genuinely hard to verify — which is why claims about the highest IQ ever recorded deserve skepticism.
Notice the pattern: each 15-point step doesn't change rarity linearly, it multiplies it. Going from 130 to 145 makes a score about 17 times rarer.
Why percentiles beat raw scores
A raw IQ number only means something because of the population behind it. Saying "my IQ is 124" is really shorthand for "I scored higher than about 94.5% of the norming sample." The percentile is the actual information; the IQ number is just a convenient encoding of it. Percentiles have three practical advantages:
- They are scale-independent. Some tests historically used SD 16 (older Stanford-Binet) or SD 24 (Cattell). An IQ of 148 on the Cattell scale and 130 on the Wechsler scale are the same 98th-percentile performance — the percentile exposes the equivalence that raw numbers hide.
- They map directly to real-world selection. Admissions committees, gifted programs and high-IQ societies think in terms of "top 2%" or "top 0.1%", not points.
- They discourage over-reading small gaps. The 5-point difference between IQ 100 and 105 spans 13 percentile points, while the same 5 points between 145 and 150 spans fewer than 0.1 — near the extremes, small score differences are mostly measurement noise.
For a plain-English tour of what each band means in daily life, our guide to the IQ scale walks through every classification from below 70 to above 145.
Common percentile cutoffs
| Cutoff | Percentile | IQ (SD 15) |
|---|---|---|
| Mensa International | 98th | ~130 |
| Typical gifted-program threshold | ~98th | 130 |
| Intertel | 99th | ~135 |
| Triple Nine Society | 99.9th | ~146 |
Mensa is the best-known example: membership requires a score at or above the 98th percentile on an approved, supervised test. Note that Mensa itself defines the bar as a percentile, not an IQ number — precisely because different tests use different scales.
A worked example: converting IQ 122 to a percentile
Suppose you scored 122 and the table above doesn't list it. The conversion takes three steps. First, compute the z-score: (122 − 100) ÷ 15 = 1.47 standard deviations above the mean. Second, look up (or compute) the cumulative probability for z = 1.47 on the standard normal curve, which is about 0.929. Third, read that as a percentile: roughly the 93rd— higher than about 93% of the population, or about 1 person in 14 scoring at or above. The same recipe works for any score on any test, as long as you know that test's mean and standard deviation.
Percentiles below 100 work identically in the other direction. An IQ of 88 gives z = −0.8, which is about the 21st percentile — higher than roughly 21% of people, lower than 79%. There is nothing special about the lower half of the curve; it is the same distribution mirrored. This matters clinically too: a score of 70 (2.3rd percentile, two SDs below the mean) is the conventional starting point for evaluating intellectual disability, though a diagnosis always requires far more than one number.
Reading your own percentile honestly
Two caveats keep a percentile meaningful. First, it describes the whole adult population, not your peer group — the 84th percentile nationally can feel thoroughly average in a graduate seminar or an engineering office. Second, the number is only as good as the test: online tests can give a useful estimate of where you fall, but they are not clinical instruments, and a formal percentile for educational or medical decisions requires a professionally administered assessment. Where the average itself comes from — and why it is pinned to 100 — is covered in what is the average IQ, and if you are wondering whether your percentile counts as impressive, what is a good IQ score puts the ranges in context.
The bottom line: your IQ score and your percentile are two views of the same fact — where you stand on the bell curve. The percentile is simply the view that's hardest to misread.