What Does a Negative Z-Score Mean? (It's Not a Mistake)

A negative z-score means one simple thing: the value is below the mean. The size of the number tells you how far below, in standard deviations. z = −1 is one standard deviation below average. z = −2.5 is two and a half below. That's the complete meaning — no hidden penalty, no error flag.

And yet this trips people up constantly, because the minus sign feels like failure. Let's kill that feeling with arithmetic.

Where the minus sign comes from

The formula is z = (x − μ) / σ. The standard deviation in the denominator is always positive. So the sign of your answer is decided entirely by the subtraction:

• x bigger than μ → positive difference → positive z
• x smaller than μ → negative difference → negative z
• x equal to μ → zero

Example: a test with mean 75 and SD 5. A score of 70 gives (70 − 75) / 5 = −5 / 5 = −1.00. A score of 65 gives −2.00. The arithmetic is unavoidable — half of every normally distributed group scores below the mean, so half of all z-scores are negative by design. If you never see negatives in your answers, that's when something's wrong. (New to the basics? Our explainer on what a z-score is starts from zero.)

Reading a negative z-score as a percentile

Negative z-scores map to percentiles under 50, and the curve's symmetry makes them easy to get: look up the positive twin and subtract from 100 (in probability terms, from 1).

• z = −0.5 → percentile = 100 − 69.15 = 30.85th
• z = −1.00 → 100 − 84.13 = 15.87th
• z = −1.50 → 100 − 93.32 = 6.68th
• z = −2.00 → 100 − 97.72 = 2.28th

So a student at z = −2 scored at or above only about 2% of the class. That's a weak result on a test — but notice the number itself stayed perfectly meaningful. If your printed z score table only lists positive values, this mirror method is exactly how you're meant to handle negatives; the z score calculator handles them directly.

Three everyday negative z-scores

A cold day. A town's January daily high temperatures average 38°F with SD 8°F. A day that reaches only 30°F: z = (30 − 38) / 8 = −1.00. Colder than about 84% of January days — a legitimately cold day, described precisely, without anyone claiming the temperature was "negative."

A shorter-than-average adult. In a group with mean height 170 cm and SD 10 cm, someone 160 cm tall has z = −1.00. One SD below average: unremarkable, millions of people share that spot.

A fast sprinter. School 100 m times average 14.0 seconds with SD 1.0. A runner finishing in 12.0 seconds has z = (12 − 14) / 1 = −2.00. Only about 2.28% of runners post a time that low — the negative z is the best result in this paragraph. Golf scores, lap times, and error counts all work this way: negative is the direction of winning.

How far negative is "too far"?

Use the same bands as the positive side, mirrored. Between −1 and 0: ordinary, the bulk of below-average values. Between −2 and −1: notably low. Beyond −2: unusual — only about 2.28% of values get there. Beyond −3: genuinely rare (0.13%), and worth a data check before you trust it, in either direction. Our guide to interpret z-scores lays out both sides in one table.

The one genuinely common real error here: flipping the subtraction and computing (μ − x) / σ, which flips every sign and quietly inverts a whole assignment's conclusions. If all your below-average values are coming out positive, that's the bug. The formula subtracts the mean from your value — x first, always.

Frequently asked questions