IQ 125: how high is it, and is it gifted?

The top tenth of the population

The Superior band covers roughly the top 10% of scores, and at 125 — 1.7 standard deviations above the mean — about 4.8% of people are at or above you. This is the range where the advantage over the average starts to be consistently visible in work that depends on absorbing complex, unfamiliar material rather than on applying skills already learned.

It is also the point where the arithmetic of rarity starts to matter more than the label. "Superior" describes a ten-point band, and its two ends are not equivalent: 120 is about one person in eleven, 129 about one in thirty. The percentile carries that difference where the label does not.

What an IQ of 125 looks like in real numbers

A score of 125 or above is about 1 in 20. In a secondary school of a thousand pupils that is roughly 48 people; scaled up to a city of a million, somewhere around 48,000.

Those numbers are model predictions from the normal curve, not headcounts — but at this distance from the mean the model is still resting on plenty of real data, which is not true further out. That distinction becomes important above 145.

125 is about the last score you can expect to meet in an ordinary room

A score of 125 or above is around one person in twenty. That has a concrete consequence: in a class, a team meeting or a bus of thirty people, you would expect roughly one or two — so someone at this level is unusual without being isolated.

A few points further up that stops being true. At 130 the figure is about one in forty-five, and by 140 it is around one in two hundred and sixty, which is more people than most of us meet in a week. 125 sits at the edge of the range where rarity is still something you can picture in a room rather than something you have to imagine across a city.

The gap to 130 is mostly administrative

130 is the conventional gifted threshold on this scale and the figure Mensa's 98th-percentile rule corresponds to. At 125 you are 5 points below it. With a standard error of roughly ±4 points, that gap is at the edge of a single test's noise, so it reflects a small real difference rather than a change in kind.

The more important point is what the line is. Nothing in the distribution changes at 130: the curve is perfectly smooth through it. The number was adopted because it is a round two standard deviations above the mean and because selection processes — society memberships, school programmes — need a written cut-off. It is an administrative convenience that acquired the authority of a scientific finding. The Mensa requirements guide covers how the rule is actually applied.

Where scores in this range concentrate

Not evenly. Occupations and courses that select hard on academic performance end up with higher average measured reasoning than the population, simply as a by-product of the selection — which is why this range is over-represented in research, engineering, medicine and law relative to its 10% population share.

That is a fact about selection, not a requirement. It says nothing about whether a given individual in one of those fields scores here, and nothing about whether someone scoring here belongs in one of those fields. Reading a group average as an entry criterion is the most common mistake made with numbers like these.

Which strengths carry a composite this high

Composites in the 120s are rarely flat. They usually rest on two or three genuinely strong abilities with the remainder closer to average — commonly strong fluid reasoning or pattern recognition alongside more ordinary processing speed. Two people at 125 can therefore have quite different profiles, and the shape is often more useful than the total.

At the midpoint of the Superior band, a 125 composite typically shows two or three strands clearly in the strong range with the remainder nearer average — the profile is where the practical information sits, since the total alone cannot distinguish a broad thinker from a specialised one.

What five points buys at this position on the curve

Five points is worth different amounts of rank depending on where you stand. Around 100 it moves you about thirteen percentile places; around 125, about three; around 150, a fraction of one. So 120–125 and 125–130 are both five-point intervals, and neither is comparable to five points near the average. This is precisely what gets hidden when a score is reported without its percentile: the point value stays constant across the scale and what it means does not.

What a supervised test would add

An unsupervised online test cannot verify conditions, cannot confirm nobody helped, and typically uses far fewer items than a clinical battery. At 125 that matters: the further from the mean a score sits, the fewer test-takers there were nearby in the norming sample to calibrate it against, so precision drops as the number rises. A supervised assessment on a full battery narrows the confidence interval and produces a subtest profile rather than a single figure — and it is the only kind of result that selection processes accept.