Exoplanet detection methods explained

Nearly all of them were found indirectly, and each method is biased toward a different kind of planet.

4 min read · Published 28 August 2026

In one paragraph

Most exoplanets are detected indirectly. The transit method watches for a star dimming as a planet crosses in front, and has found the large majority. Radial velocity measures the star's wobble as a planet tugs it, and was the first successful method. Both are strongly biased toward large planets in close orbits, so early exoplanet catalogues were dominated by hot Jupiters.

The main methods

Method	What it measures	Biased toward
Transit	Starlight dimming as a planet crosses the star	Large planets, short orbits, near-edge-on alignment
Radial velocity	The star's motion toward and away from us	Massive planets in close orbits
Direct imaging	The planet's own light, with the star blocked	Large, young, hot planets far from their star
Gravitational microlensing	Light bending as a system passes in front of a background star	Planets at intermediate distances; not repeatable
Astrometry	The star's position wobbling on the sky	Massive planets in wide orbits

Notice that every method has a bias, and they differ. Any statement about how common a type of planet is must account for which method found it. The raw catalogue reflects what is easy to detect, not what exists.

The transit method

If a planet's orbit is oriented so it passes between us and its star, the star dims slightly and repeatedly. Measuring that dip gives the planet's size relative to the star, and the repeat interval gives the orbital period.

The dips are small. Jupiter crossing the Sun would dim it by about 1%; Earth by about 0.008%. Detecting that requires exceptional photometric precision, so dedicated space telescopes were built for it.

The severe limitation is geometry. The orbit must be nearly edge-on from our viewpoint, which is true for a small fraction of systems, well under 1% for Earth-like orbits. So the method misses the vast majority of planets, not because they're undetectable but because they never cross.

Its great advantage is that transits also permit atmospheric analysis: starlight filtering through a planet's atmosphere during transit carries the signature of its composition, which is how JWST studies exoplanet air.

The transit method's bias means we mostly find planets we'd never want to live on, and I'd read every 'Earth-like' headline with that in mind.

Radial velocity

A planet doesn't orbit its star; both orbit their common centre of mass. The star therefore traces a small circle, moving alternately toward and away from us, which shifts its spectrum by the Doppler effect.

This gives the planet's mass, or rather a minimum mass, since the orbital inclination is usually unknown.

The velocities are minuscule. Jupiter makes the Sun move at about 12 metres per second; Earth at about 9 centimetres per second. Detecting an Earth analogue requires measuring a star's velocity to better than walking pace across light years, which is at the edge of current capability.

Combining both methods is powerful: transit gives radius, radial velocity gives mass, and together they give density: which distinguishes a rocky planet from a gas envelope.

What the biases mean

Early catalogues were dominated by hot Jupiters: massive planets in very short orbits. That was widely reported as surprising, and it was partly a selection effect: those are exactly the planets both main methods find most easily.

Correcting for detection bias suggests small planets are far more common than large ones, and that planets are extremely common generally. Most stars appear to have them.

Two categories with no solar system analogue turned out to be among the most common: super-Earths and sub-Neptunes, between Earth and Neptune in size. Our own system, oddly, has nothing in that range.

This is a good discipline to carry into any survey result. The first thing to ask about a catalogue is what it was capable of finding, and the answer usually explains a large part of what it contains.