Skip to main content
ponderloop

Normal Distribution Calculator

For a normal distribution with a mean of 170 and a standard deviation of 10, P(160 < X < 185) = 0.774538: 77.4538% of values fall between 160 and 185, which are z-scores of −1 and 1.5. Enter your own mean and standard deviation and pick the question: the probability below a value, above it, between two values or farther than a set distance from the mean, or the value that sits at a given percentile. Each answer comes as a decimal and a percent with the matching z-scores, a bell curve with the area shaded, the 68–95–99.7 rule worked out for your numbers and a table of common percentiles.

Updated Runs in your browser: nothing you enter is uploadedHow this tool works

What do you want to find?

z_a = (160 − 170) ÷ 10 = −1, z_b = (185 − 170) ÷ 10 = 1.5

P(160 < X < 185) = 0.774538

With a normal distribution with mean 170 and standard deviation 10, the probability that X lies between 160 and 185 is 0.774538, so 77.4538% of values fall in that range.

Probability, P(a < X < b)
0.774538
As a percent
77.4538%
z-score of a
−1
z-score of b
1.5
Probability outside, P(X < a or X > b)
0.225462
Mean (μ)
170
Standard deviation (σ)
10

Bell curve

160185140−3150−2160−11700180119022003xz
Shaded: P(160 < X < 185), 77.4538% of the area under the curve.

For your mean and standard deviation

The 68–95–99.7 rule for a mean of 170 and a standard deviation of 10.
WithinRange of valuesShare of values
1 standard deviation160 to 18068.2689%
2 standard deviations150 to 19095.45%
3 standard deviations140 to 20099.73%
Common percentiles for a mean of 170 and a standard deviation of 10.
Percentilez-scoreValue
1st≈ −2.326348≈ 146.736521
5th≈ −1.644854≈ 153.551464
10th≈ −1.281552≈ 157.184484
25th≈ −0.67449≈ 163.255102
50th0170
75th≈ 0.67449≈ 176.744898
90th≈ 1.281552≈ 182.815516
95th≈ 1.644854≈ 186.448536
99th≈ 2.326348≈ 193.263479

Worked examples

Each result is computed by the calculator from the inputs in the middle column.
QuestionInputsResult
Adults between 160 cm and 185 cm tallmean 170, SD 10, between 160 and 1850.774538 (77.4538%)
Exam scores below 60mean 72, SD 9, below 600.091211 (9.1211%)
Deliveries taking more than 45 minutesmean 35, SD 6, above 450.047790 (4.779%)
Bolts more than 0.3 mm off the target lengthmean 50, SD 0.2, farther than 0.3 from the mean0.133614 (13.3614%)
The height that 90% of adults fall belowmean 170, SD 10, 90th percentilex ≈ 182.815516
The score that only the top 5% reachmean 500, SD 80, 95th percentilex ≈ 631.58829
Within one standard deviation of the meanmean 100, SD 15, between 85 and 1150.682689 (68.2689%)
The middle 95% of a standard normal curvemean 0, SD 1, between −1.96 and 1.960.950004 (95.0004%)

The 68–95–99.7 rule for a mean of 170 and a standard deviation of 10

The share of a normal distribution within 1, 2 and 3 standard deviations of the mean, with the range that covers for the example.
WithinRange of valuesShare of values
1 standard deviation160 to 18068.2689%
2 standard deviations150 to 19095.45%
3 standard deviations140 to 20099.73%

Probability within and beyond a distance from the mean

For a distance of k standard deviations: the share of values within it, and the share beyond it on each side. The same for every normal distribution.
Distance from the meanWithin (both sides)Beyond (both sides)Beyond (one side)
0.5 σ0.38290.61710.3085
1 σ0.68270.31730.1587
1.5 σ0.86640.13360.0668
1.645 σ0.90000.10000.0500
1.96 σ0.95000.05000.0250
2 σ0.95450.04550.0228
2.576 σ0.99000.01000.0050
3 σ0.99730.00270.0013

Percentiles of a normal distribution with mean 170 and standard deviation 10

The value below which each share of the distribution lies, from the inverse normal distribution.
Percentilez-scoreValue
1st≈ −2.326348≈ 146.736521
5th≈ −1.644854≈ 153.551464
10th≈ −1.281552≈ 157.184484
25th≈ −0.67449≈ 163.255102
50th0170
75th≈ 0.67449≈ 176.744898
90th≈ 1.281552≈ 182.815516
95th≈ 1.644854≈ 186.448536
99th≈ 2.326348≈ 193.263479

How to calculate a normal distribution probability

  1. Choose the question. Pick whether you want the probability below a value, above a value, between two values, farther than a set distance from the mean, or the value at a percentile.
  2. Enter the mean and standard deviation. Type the mean (μ) and standard deviation (σ) of the distribution: 170 and 10 in the example. The standard deviation must be greater than 0.
  3. Enter the value or values. For the example, a = 160 and b = 185. Each one is turned into a z-score, z = (x − μ) ÷ σ: −1 and 1.5.
  4. Read the probability. The probability is the area under the bell curve over your range: 0.774538 as a decimal, 77.4538% as a percent. The curve shades that area.
  5. Check it against the tables. Compare with the 68–95–99.7 rule and the percentile table for your mean and standard deviation: a result far from what they suggest usually means a mixed-up mean, standard deviation or limit.

Frequently asked questions

How do I find the probability between two values in a normal distribution?
Turn both values into z-scores with z = (x − μ) ÷ σ, look up the area to the left of each, and subtract: P(a < X < b) = Φ(z_b) − Φ(z_a). For a = 160, b = 185, mean 170 and standard deviation 10, the z-scores are −1 and 1.5, and the probability is 0.774538 (77.4538%). The calculator does this for you and subtracts the right-hand tails when the lower value lies above the mean, so that small probabilities keep their digits.
How do I find the value (x) for a given percentile?
Choose “Find the value at a percentile”, enter the percentile and the mean and standard deviation. The calculator finds z with the inverse of the normal distribution function and returns x = μ + zσ. For the 90th percentile with mean 170 and standard deviation 10, z is ≈ 1.281552 and x is ≈ 182.815516. On the standard normal curve the 97.5th percentile is z ≈ 1.959964.
What does P(|X − μ| > c) mean?
It is the probability that a value lies farther than c from the mean in either direction, so below μ − c or above μ + c. With mean 170, standard deviation 10 and c = 15, the limits are 155 and 185, and the probability is 0.133614 (13.3614%). The probability of staying within c is the rest, 0.866386.
Is P(X < a) the same as P(X ≤ a)?
Yes. A normal distribution is continuous, so the probability of landing exactly on one value is 0 and “below” and “at most” give the same answer. For a = 160 the probability below is 0.158655 and above is 0.841345; they add up to 1. If your data are whole numbers (ages, counts), a continuity correction treats “at most 160” as “below 160.5”, which you can do by entering 160.5.
What exactly does the 68–95–99.7 rule say?
It says that about 68%, 95% and 99.7% of a normal distribution lie within 1, 2 and 3 standard deviations of the mean. The exact shares are 68.2689%, 95.45% and 99.73%, and the page works them out for your mean and standard deviation (for the example, 160 to 180, 150 to 190, 140 to 200). Values 2 standard deviations from the mean are not rare: about 4.55% of values are farther out than that.
How is this different from a z-score calculator?
A z-score calculator centres on one number: how many standard deviations a score is from the mean, with its percentile. This calculator starts from the distribution (mean and standard deviation) and answers probability questions about ranges and tails, and the reverse question of which value sits at a percentile. It shows the z-scores along the way, so the two give the same numbers where they overlap.
Can I use it with sample data?
Yes, as an approximation: enter the sample mean and the sample standard deviation as μ and σ. The result then describes a normal curve fitted to your sample, which is only as good as the fit. If the sample is small and you are testing a mean or building a confidence interval, the t distribution is the right tool, because estimating σ adds uncertainty that the normal curve ignores.
What if my data are not normally distributed?
The probabilities here only describe data that follow a bell curve. Skewed data such as incomes or waiting times, or data with hard limits such as percentages, can give answers that are far off in the tails even when the mean and standard deviation are right. A histogram or a normal quantile plot of your data is a quick check before you rely on a tail probability.
How accurate are the results?
The distribution function is evaluated in double precision with W. J. Cody’s error-function approximations and the inverse with Peter Acklam’s algorithm plus a Halley step, and the results are compared in the tests with SciPy’s scipy.stats.norm. Probabilities are shown to 6 decimals, and to 4 significant digits in exponent form below 0.000001: for example the probability of a value more than 5.5 standard deviations above the mean is 1.899e-8.
Why must the standard deviation be above 0?
A standard deviation of 0 would put every value at the mean, so there is no curve to take areas under and z = (x − μ) ÷ σ would divide by zero. The calculator asks for a positive standard deviation rather than showing a made-up number.
What is the difference between the percentile and the probability?
They are the same area read two ways. A percentile is the probability to the left of a value times 100: a value at the 90th percentile has a probability of 0.9 below it. Going from the value to the percentile is the “below” question (for 160 it is 0.158655, the 15.9th percentile); going from the percentile to the value is the “find the value” question.

How it works

A normal distribution is set by its mean μ and standard deviation σ. A value x is standardised as z = (x − μ) ÷ σ, and probabilities are areas under the standard normal density. The distribution function Φ(z) = ½ erfc(−z ÷ √2) is evaluated in double precision with W. J. Cody’s rational Chebyshev approximations of the error function (Math. Comp. 23, 1969), which keep their relative accuracy far into the tails. P(X < a) is Φ(z_a), P(X > a) is the right tail ½ erfc(z_a ÷ √2), and P(a < X < b) is Φ(z_b) − Φ(z_a), or the difference of right tails when a lies above the mean. P(|X − μ| > c) is erfc(c ÷ (σ√2)). The value at a percentile uses the inverse of Φ: Peter Acklam’s rational approximation followed by one Halley step, then x = μ + zσ. These shared numerics are tested against SciPy’s scipy.stats.norm (cdf, sf and ppf) and the calculator’s own results against SciPy values for several means, standard deviations and tails. Probabilities are shown to 6 decimals, or in exponent form below 0.000001; z-scores and values to at most 6 decimals, with ≈ where that rounded. The bell curve is drawn from mean − 4σ to mean + 4σ and a limit beyond that is marked at the edge. The normal model is an assumption: the probabilities describe your data only as far as they follow a bell curve. Everything is computed in your browser.

Sources

More: Statistics, probability and gradesAll math & numbersAll toolsHome & DIY calculators