Video 2 · 4:53 · Aug 2026 · 503 words
Z-Scores Explained Simply: How Rare Is 2 Metres?
One ruler for every bell curve. A z-score says exactly how unusual a measurement is — we use it to answer how rare a two-metre person really is, and to clear up what most people get wrong about outliers: a flag, not a verdict.
Open on YouTube ↗ Quiz for this video (3 questions) Watch the series in order ↗
What is a z-score?
A z-score says how many standard deviations a value sits from the mean, which turns every bell curve into the same ruler. A z of +2 means two standard deviations above average, whatever you are measuring. That is how you compare a height with an exam score, and how you say how rare something really is.
What you take away
- z = (value − mean) ÷ standard deviation.
- z = 2 is roughly the top 2.3% of a normal distribution; z = 3 is about the top 0.13%.
- Rare for one person is common in a crowd: 1 in 1,000 still means thousands of people in a country.
- An outlier is a flag to investigate, not a verdict to delete.
Chapters
Jump to a part of the transcript; each heading links to that moment on YouTube.
Full transcript
Word for word what is said in the video — the same text as the subtitles. Times are minutes:seconds into the video.
0:00The question: how special is 2 metres? ▶
Last time, I asked: how special is someone who is two metres tall? Today we work it out — exactly. And along the way, we'll clear up what most people get wrong about outliers.
0:14The z-score: one ruler for every bell curve ▶
Here's our bell curve from last time. The mean height, μ (mu), is 170 centimetres — the orange line. The standard deviation, σ (sigma), is 8 centimetres — the typical distance from the mean. Now watch this. As someone grows taller, we can state their height in centimetres… or in sigmas above the mean. That second number is the z-score. z equals x minus mu, divided by sigma. Positive means above the mean; negative means below. At 178, z is exactly one. At 186, z is two. And the beautiful part: this ruler isn't just for heights. Birth weights, test scores — every bell curve gets the same ruler.
1:19Climbing the ladder of rarity ▶
So, our two-metre man. 200 minus 170 is 30 centimetres above the mean. Divide by sigma: 30 over 8 gives z = 3.75. How rare is that? Remember the bands from last time — 68, 95, 99.7? Careful: those count both sides of the curve. For "taller than", we only need the right side. So let's climb. Above z = 2, you'll find about 1 in 44 people. Above z = 3: 1 in 741. And above 3.75 — our two- metre man — just 1 in 11,310. The computer looks these up for us; no tables needed.
2:21Rare for a person, ordinary in a crowd ▶
1 in 11,310 sounds like "basically nobody". Now zoom out. A country of eighteen million people holds about 1,600 people standing two metres or taller. Rare for one person — ordinary in a crowd. Hold on to that thought: extremely unusual is not the same as impossible. Or wrong.
2:52Outliers: a flag, not a verdict ▶
Points that far from the mean get a label: outliers. A common rule of thumb: beyond two sigma is unusual; beyond three, an outlier. Useful — but it's a convention, not a law of nature. Because extreme does not mean wrong. A recorded height of 20 metres? That's a typo — fix it. A recorded height of 2 metres? That's a real person. That tail is so real that a whole sport is built on it: professional basketball is the far right of this curve, making a living. So treat an outlier as a flag to investigate, never a licence to delete. Check the source. Measure again. Decide then. And if you quietly delete real extremes, sigma shrinks — and yesterday's "normal" becomes tomorrow's "outlier". A vicious circle.
4:05In one picture ▶
So, in one picture. The z-score measures distance from the mean, in sigmas. One ruler fits every bell curve. z = 3.75 means 1 in 11,310 — yet about 1,600 in a country of eighteen million. And outliers are rarely wrong. They're just rare.
4:33What's next: percentiles & the IQR ▶
Next time, we drop the bell curve entirely: percentiles and the interquartile range can call something "extreme" with no assumptions at all. And tell me below: have you ever thrown away an outlier… you should have investigated? Until then: happy learning.