Parallax Distance Calculator

Convert between a star's parallax and its distance exactly, watch the geometry that makes it work, and see how measurement uncertainty propagates — and when the simple inversion stops being trustworthy.

Parallax / distance calculator

d(pc) = 1 / p(″) — exact by the definition of the parsec. A star's parallax is the tiny back-and-forth shift in its apparent position, measured against distant background stars, as Earth orbits the Sun. Smaller shift means a farther star.

Parallax (p)

Distance (d)

Astronomical units268399.23 AU

Parsecs1.3012 pc

Light-years4.2441 ly

Kiloparsecs0.001301 kpc

Schematic — the star's distance and the apparent shift are both exaggerated for visibility (real stellar parallaxes are far too small to draw to scale); the direction and relative size of the effect — smaller shift, farther star — are accurate.

Every other distance-measuring method in astronomy — standard candles, redshift, the whole cosmic distance ladder — ultimately calibrates itself against one technique that needs no assumptions about what a star is or how bright it “should” be: parallax. Measure how much a star’s apparent position wiggles over a year, relative to the much farther background stars, and you have its distance directly from geometry. It’s also where the parsec comes from — the unit is defined by this relation.

The relation

d(\text{pc}) = \frac{1}{p(^{\prime\prime})}

A parsec is, by definition, the distance at which 1 AU subtends an angle of exactly one arcsecond. That’s not a coincidence dressed up as a formula — it’s why the relation is this clean. A parallax of 100 milliarcseconds is a distance of exactly 10 pc, no approximation involved.

The geometry

The diagram above shows why this works: as Earth orbits the Sun, a nearby star’s line of sight sweeps through a small angle relative to the much more distant background stars, which don’t move appreciably no matter where Earth is. Compare the star’s apparent position from opposite points in Earth’s orbit (classically, observations six months apart) and it appears to have shifted — not because the star moved, but because you did. The parallax angle p is defined as the shift relative to the mean position (i.e., as seen from the Sun), so the full back-and-forth swing between the two extreme epochs is 2p.

When d = 1/p stops being trustworthy

Parallax measurement always carries some uncertainty, and it’s tempting to propagate it the simple way: since d = 1/p, a fractional uncertainty on p becomes the same fractional uncertainty on d. That’s exactly right as a first-order approximation — but it’s only a first-order approximation, because 1/p is a nonlinear transform. For a roughly-Gaussian parallax measurement, the induced distribution on distance becomes increasingly skewed as the fractional uncertainty grows, and the naive point estimate becomes biased. Push it far enough — a parallax consistent with zero within its error bar — and 1/p isn’t just imprecise, it’s undefined or nonsensical (a negative or infinite “distance”).

This isn’t a hypothetical edge case. It’s a real, well-documented issue in modern astrometry: Gaia measures parallaxes for well over a billion stars, and a large fraction of the faintest, most distant ones have fractional uncertainties large enough that naive inversion measurably biases the results. The standard fix, following Bailer-Jones (2015) and the body of work it spawned, is Bayesian: combine the parallax measurement with a prior on how stars are actually distributed in space, rather than trusting 1/p in isolation. As a rule of thumb, this calculator treats fractional uncertainties under 10% as reliable, 10–20% as worth caution, and anything larger as a case where simple inversion shouldn’t be trusted at face value.