What Is the Highest IQ Score Possible?

Real test ceilings

Most current clinical tests report a maximum Full Scale IQ of about 160. Both the Wechsler Adult Intelligence Scale (WAIS-IV) and the Stanford-Binet Fifth Edition (SB5) use a mean of 100 and a standard deviation of 15, and both cap the Full Scale IQ near that level. A 160 already lands a person at roughly the 99.99th percentile.

Two things in that table matter. First, the WISC-V for children was given extended norms developed by Pearson with the National Association for Gifted Children, which push the reportable range a bit past the usual ceiling using statistical modeling rather than by actually testing millions of rare high scorers. Second, the Cattell scale uses a standard deviation of 24 instead of 15, so a Cattell 160 is nowhere near a Wechsler 160. This is exactly how a person can honestly quote a bigger number: they are on a stretchier ruler. A score only means something when you also know the standard deviation behind it.

Why extreme scores are unreliable

The reason tests stop around 160 is not modesty. It is that the numbers stop being trustworthy. Three forces pile up at the top of the scale.

The norming sample runs out. A test is calibrated by giving it to a representative standardization sample. Even a large sample contains only a handful of people scoring above 145, and essentially none above 160. So the tables that convert a raw score into an IQ at those heights are built by smoothing and extending the curve, not by observing real people at that level. Extrapolation is not the same as measurement.

Measurement error swamps the signal. Professionally administered tests carry a standard error of measurement of roughly 3 to 5 points. Near the ceiling there are only a few of the hardest items left, so one lucky or unlucky answer can swing a reported score by several points. Down the middle of the scale that barely matters. Out in the thin tail, a 5-point swing is the difference between "1 in 30,000" and "1 in 3 million."

The tail thins faster than the numbers suggest. Equal steps on the scale are wildly unequal in scarcity. Going from 130 to 145 moves you from about the top 2 percent to about the top 0.1 percent. The next 15 points is rarer still. Because so few people exist at each higher rung, there is simply not enough evidence to say that a 158 and a 162 are meaningfully different people. A sensible rule of thumb: trust rarity estimates reasonably well between about 70 and 145, and treat anything above that as an approximation rather than a precise count.

How this differs from "highest ever recorded" claims

The famous giant numbers are a separate thing from what a test can measure today. The best-known example is Marilyn vos Savant, listed for years in the Guinness Book of Records with an IQ of 228. That figure came from a childhood test using the 1937 Stanford-Binet, which used the old ratio IQ method: mental age divided by chronological age, times 100. A ten-year-old who tests like a young adult can produce a huge ratio that modern deviation scoring would never generate. Guinness eventually retired its "Highest IQ" category in 1990, concluding that IQ tests were too unreliable to crown a single record holder.

Other legendary figures, such as William James Sidis with claimed scores of 250 to 300, were never verified by any standardized modern test at all. These numbers survive as folklore, not as measurements. The honest takeaway is that a reported 228 from a 1937 childhood formula and a modern WAIS 160 are not on the same ruler, and neither one proves a person is twice as capable as someone at 145.

None of this makes a genuinely high score meaningless. It just means the confident three-digit trophy numbers you see online deserve a raised eyebrow, and that the difference between, say, 150 and 165 is far less solid than the difference between 100 and 130. If you want to know where you stand, the useful information is your band and your percentile, not chasing a personal record at the edge of the curve.

Frequently asked questions

Q: Is there a maximum IQ score?

A: Not in any absolute sense, but tests set a practical ceiling near 160. IQ is a rank on a bell curve, so the formula has no upper limit. What is capped is the instrument: the WAIS-IV and Stanford-Binet 5 both stop reporting Full Scale IQ at roughly 160, because there is not enough data to measure reliably above that.

Q: Can you have an IQ over 200?

A: You can be assigned a number that high, but no standardized modern test can verify it. A 200 implies a rarity greater than the entire human population, so it comes from ratio-IQ childhood formulas, unstandardized high-range tests, or raw extrapolation, not from a validated measurement.

Q: Why do the most respected tests stop at 160?

A: Because above roughly 145 the norming data becomes too thin to trust. Standardization samples contain almost no one at that level, measurement error of 3 to 5 points gets amplified in the tail, and equal point steps represent hugely unequal rarity. Publishers cap the score rather than print numbers they cannot defend.

Q: How can someone claim an IQ of 228 if the ceiling is 160?

A: Those figures usually come from a different scoring method. Marilyn vos Savant's 228 was a childhood ratio IQ from a 1937 test, not a modern deviation score. Different tests also use different standard deviations, so a big number on a stretchier scale is not comparable to a 160 on the Wechsler scale.

Q: Does a higher number mean someone is proportionally smarter?

A: No. The scale is a rank, not a linear amount of ability. The gap between 130 and 145 already spans the top 2 percent down to the top 0.1 percent, and above that the numbers grow less reliable, so small differences at the extreme high end tell you very little.

References

• Wikipedia. "IQ classification." https://en.wikipedia.org/wiki/IQ_classification
• Wikipedia. "Marilyn vos Savant." https://en.wikipedia.org/wiki/Marilyn_vos_Savant
• MDPI Encyclopedia. "Ceiling Effects: The Example of IQ 276." https://encyclopedia.pub/entry/58818

Last updated: July 13, 2026