Percentile Calculator: Simple and Fast

Enter numbers separated by commas, spaces, or new lines:

Enter percentile to calculate (0-100):

Select calculation method:




Common Percentiles:

Four Calculation Methods

Statisticians don't agree on a single method for calculating a percentile that falls between two data points, so this tool offers four standard options: Linear Interpolation (the most common, and what Python's NumPy and most statistics textbooks default to), Nearest Rank (rounds to the closest actual data point, no interpolation), Excel PERCENTILE.INC (matches Microsoft Excel's default function), and Excel PERCENTILE.EXC (Excel's exclusive variant, which treats the dataset as a sample rather than the whole population).

Worked Example: Linear Interpolation

Dataset: 10, 20, 30, 40, 50 (5 values, already sorted). To find the 90th percentile: index = (90/100) x (5-1) = 3.6. This falls between index 3 (value 40) and index 4 (value 50), with a weight of 0.6 toward the upper value. Result: 40 x (1 - 0.6) + 50 x 0.6 = 16 + 30 = 46. So the 90th percentile is 46, even though 46 isn't one of the five entered numbers - interpolation estimates a value between the two nearest data points.

Why the Methods Disagree

For the same dataset, the 50th percentile (median) via linear interpolation lands exactly on the middle value (30, since its index of 2.0 is a whole number and needs no interpolation) - but Nearest Rank and the two Excel methods can produce slightly different results for percentiles that don't land on a clean index, particularly with small datasets.

Dataset Statistics

Alongside the percentile, the tool reports count, min, max, range, mean, and median for context - useful for spotting whether a requested percentile is even meaningful (a 99th percentile request on a 5-value dataset, for instance, is really just asking for something close to the maximum).