The Scoreboard Nobody Mentions at Dinner Parties

Ask how astronomers find planets around other stars and you will usually get a list: the transit method, radial velocity, direct imaging, microlensing, astrometry. All real, all in use. What the list leaves out is the part that actually matters if you want to understand the field, which is that these methods are nowhere near equal. They do not split the discoveries evenly between them. One technique has done most of the work, a second has done most of the rest, and the others fill in the corners. If you sorted every confirmed exoplanet by how it was first spotted, the result would look less like a balanced pie chart and more like a landslide.

That landslide belongs to the transit method. Roughly three-quarters of all known exoplanets were found by watching a star dim slightly as a planet crossed in front of it. Radial velocity, the technique that opened the whole field in the 1990s, sits in second place with a sizable but much smaller share. Everything else combined, the direct images and the microlensing events and the painstaking astrometry, accounts for a thin slice at the bottom. Knowing the ranking is the difference between treating these methods as a menu and understanding which one you would actually bet on.

None of that means the small-share methods are failures. It means they are specialists. Each one reaches a kind of planet the others cannot, and the value of a technique is not only the raw count it produces but the gaps it fills. Still, if the question is which methods are most effective at the plain task of finding worlds, the numbers give a clear answer, and it is worth sitting with that answer before getting into the caveats.

Why the Transit Method Runs Away With It

The transit method wins for an almost unfair reason: it scales. You do not have to study planets one at a time. Point a sensitive camera at a dense field of stars, watch all of them at once, and wait. Any star with a planet crossing its face on your line of sight will betray itself with a tiny, repeating dip in brightness. A telescope monitoring tens of thousands of stars in a single frame can harvest candidates by the hundred, and that is exactly what happened once the right instruments went up. The method turns planet-hunting from a craft into an assembly line.

It also has a second advantage that matters enormously for the kind of planets people care about most. The depth of the dip tells you the size of the planet relative to the star. A small rocky world blocks very little light, a fraction of a percent or less, but a stable enough detector can catch even that. This is why the transit method became the workhorse for finding Earth-sized and super-Earth planets rather than just the bloated giants. The other big technique, radial velocity, struggles with small worlds because their gravitational tug is faint. Transit photometry only needs the geometry to line up and the instrument to hold steady.

The catch lives in that phrase, “on your line of sight.” A transit only happens if the planet’s orbit is edge-on from where we sit, and most orbits are not. For a planet in an Earth-like orbit around a Sun-like star, the odds of the alignment are well under one percent. So the method misses the overwhelming majority of planets that are genuinely there; it only counts the lucky few whose orbits happen to point at us. The reason it still dominates the tally is that there are so many stars that even a sub-one-percent hit rate, multiplied across hundreds of thousands of targets, produces an avalanche.

What Kepler Did to the Numbers

You cannot talk about the transit method’s dominance without talking about Kepler. NASA’s Kepler spacecraft, launched in 2009, stared at a single patch of sky in the constellation Cygnus for years, measuring the brightness of around 150,000 stars over and over with a precision no ground-based telescope could match. The result rewrote the scoreboard. Kepler alone is responsible for thousands of confirmed planets, more than every other method and mission combined up to that point, plus thousands more candidates that kept teams busy long after the spacecraft itself was retired.

Kepler’s real gift was not just the count but the statistics. Because it watched so many stars so carefully, it let astronomers estimate how common planets are in general, including the small ones that are hard to find. The headline conclusion, that planets outnumber stars and that small rocky worlds are abundant, came largely out of Kepler’s haul. That single mission is most of the reason we can say with a straight face that the galaxy is full of planets. It took the transit method, which was already promising, and turned it into the defining tool of the era.

Even Kepler’s troubles ended up productive. When two of its reaction wheels failed and it could no longer hold a steady stare, the team repurposed it into a second mission, K2, that swept along the plane of the sky and added still more planets to the list. The lesson buried in all this is about method effectiveness: a technique that can be industrialized and flown above the atmosphere will out-produce more delicate approaches by orders of magnitude, simply because it can keep counting.

TESS and the All-Sky Strategy

Kepler went deep on one small region. Its successor, TESS, launched in 2018, made the opposite bet and went wide. Instead of staring at faraway stars in one patch, TESS scans almost the entire sky in segments, focusing on the brightest, closest stars. The trade is deliberate. TESS finds fewer planets per star than Kepler did, and many of its discoveries sit on shorter orbits, but the planets it does find orbit nearby stars bright enough to study in detail afterward.

That detail is the whole point. A transit tells you a planet’s size and its orbit, but not much else on its own. To learn a planet’s mass you need radial velocity follow-up, and to sniff its atmosphere you need a big telescope catching starlight filtered through the planet’s air. Both jobs are far easier when the host star is close and bright. By prioritizing nearby targets, TESS feeds a pipeline rather than just padding a number. It is the difference between a method that produces statistics and one that produces planets you can actually interrogate.

So within the single most effective technique there are two distinct strategies, and they are effective at different things. Kepler maximized the census; TESS maximizes the follow-up. Neither is “better” in the abstract. If you want to know how common Earth-sized planets are, you want a Kepler. If you want a short list of worlds whose atmospheres a future telescope could read, you want a TESS. The transit method earns its crown partly because it is flexible enough to serve both goals.

Radial Velocity: The Method That Gives You Mass

Before any spacecraft watched a star blink, radial velocity found the first planets around Sun-like stars. The technique reads the spectrum of a star and watches it shift, ever so slightly, toward the blue as the star is tugged toward us and toward the red as it is tugged away. That wobble is the star’s response to the gravity of an unseen planet. The first such detection around a normal star, 51 Pegasi b in 1995, earned a Nobel Prize and kicked off the modern hunt. For years afterward, radial velocity was the dominant method, until transit surveys overtook it on sheer volume.

What keeps radial velocity essential is the one thing transits cannot do: it measures mass. A transit gives you a planet’s diameter; a radial velocity measurement gives you, in effect, its weight. Put the two together and you get density, the figure that reveals whether a planet is a heavy rocky world or a light, gas-dominated one. Neither number alone settles what kind of planet you are looking at. Together they nail it. This is why the two leading methods are most powerful not as rivals but as a pair, each supplying what the other lacks.

Radial velocity has its own ceiling, though. The wobble caused by a small planet is tiny, sometimes a stellar motion of barely a meter per second, slower than a walking pace, read across light-years. The natural churning of a star’s own surface can drown out a signal that faint. So radial velocity has historically favored bigger planets and those orbiting close to their stars, where the tug is strongest. Pushing it down to genuinely Earth-like worlds is one of the hardest instrument-building problems in the field, and progress there is measured in slow, hard-won increments rather than landslides.

The Specialists: Microlensing, Imaging, Astrometry

Below the two heavyweights sit the methods that contribute smaller shares but reach places the others cannot. Gravitational microlensing uses the fact that a star’s gravity bends and briefly magnifies the light of a more distant star lined up behind it. If the foreground star has a planet, that planet adds a short extra spike to the brightening. Microlensing is strange and powerful because it does not depend on light from the planet or even a long stable orbit; it can catch planets far from their stars, and even rogue planets drifting through space with no star at all. The drawback is that each event is a one-off. The alignment never repeats, so you cannot go back and confirm by looking again.

Direct imaging is the method everyone wishes worked better, because it does the obvious thing: it takes a picture. Block the overwhelming glare of the star with a coronagraph or a starshade and you can, in rare cases, see the faint dot of the planet itself. The trouble is brutal contrast. A planet can be billions of times fainter than its star and sit right next to it. So direct imaging today works mainly on young, hot, massive planets orbiting far from their stars, where the planet still glows from its own heat and the separation is wide enough to resolve. It produces few planets, but the ones it gets are the only ones whose light we can study directly rather than by inference.

Astrometry is the patient one. Instead of measuring a star’s motion toward and away from us, it measures the star’s tiny side-to-side shuffle across the sky as a planet swings it around. The wobble is minuscule and demands extraordinary positional precision over long stretches of time, which is why astrometry has historically confirmed almost nothing on its own. That is about to change, and the reason is a single spacecraft worth its own paragraph.

Matching the Method to the Question

The honest way to rank these methods is to first ask what you are trying to do, because “most effective” depends entirely on the goal. If you want the largest possible catalog of planets, the transit method from space wins and it is not close. If you want a planet’s mass and density, you need radial velocity, ideally paired with a transit. If you want to study a planet’s atmosphere, you want transits of nearby bright stars, the TESS strategy, handed off to a spectrograph. If you want to see a planet as an actual point of light, only direct imaging will do, and only for a narrow class of big, young, far-out worlds.

If you care about planets on wide orbits or worlds wandering without a star, microlensing is the only tool that reaches them. And if you want a complete census of a nearby star’s full planetary family, including the long-period giants that take years to reveal themselves, astrometry is the method built for that job. Seen this way, the small-share techniques are not weak. They are answers to questions the dominant method simply cannot address, and a field that only ran transit surveys would be blind to whole categories of planet.

This is also why serious discoveries usually lean on more than one method. A transit candidate becomes a confirmed planet with a known mass only after radial velocity follow-up rules out the impostors, things like a faint companion star mimicking a planetary dip. The combination is what produces trustworthy results. No single technique, taken alone, gives the full picture, and the most effective programs are the ones that chain methods together rather than betting everything on one.

The Quiet Engine: Software That Sorts the Flood

There is a part of effectiveness that has nothing to do with telescopes. Kepler and TESS produce so many light curves that no team could ever eyeball them all, and the real planets hide among countless false alarms, instrumental glitches, eclipsing binary stars, and ordinary stellar jitter. Sorting that flood is where modern software earns its keep. Trained algorithms now scan the brightness curves, flag the dips that behave like genuine transits, and toss the ones that do not, surfacing the handful of candidates worth a human astronomer’s time.

This matters for the ranking because the transit method’s dominance is partly a software story. A technique only counts the planets you actually dig out of the data, and for a while the bottleneck was not collecting transits but vetting them. Better automated classification has pulled real planets out of archives years after the observations were taken, including some that earlier passes had missed. The method did not change. Our ability to mine it did, and the count climbed as a result.

The same shift is creeping into radial velocity and the others, where the challenge is teasing a faint planetary signal out of the noise of a living, churning star. None of this replaces the physics. It just means that, increasingly, the most effective method is whichever one we can process well enough to trust, and the gap between a detection sitting in raw data and a confirmed planet on the list is now as much a computing problem as an observing one.

What Will Shift the Rankings Next

The current standings are not permanent. A few missions on the horizon are built to change them. PLATO, a European transit mission, is designed to hunt specifically for Earth-sized planets in temperate orbits around Sun-like stars, the exact corner of the map that even Kepler only sampled thinly. The Roman Space Telescope will run a large microlensing survey toward the galactic center, which could finally give that specialist method a bulk harvest and a much clearer view of planets on wide orbits and free-floating worlds.

Then there is Gaia. The European astrometry spacecraft has spent years measuring the positions of more than a billion stars with staggering precision, and as its long baseline of data matures it is expected to reveal thousands of planets through their host stars’ tiny side-to-side wobbles, mostly massive planets on wide orbits. If that pans out, astrometry leaps from a footnote to a major contributor almost overnight, filling in exactly the long-period giants that transit and radial velocity surveys tend to miss. On the ground, the next generation of giant telescopes, with mirrors tens of meters across, will push direct imaging toward smaller and cooler planets than it can reach today.

Put it together and the future looks less like one method winning even bigger and more like the field balancing out. Transits will likely stay on top by raw count for a while, but microlensing, astrometry, and imaging are each lined up for the kind of leap that reshuffles a leaderboard. The most effective method ten years from now may simply be whichever one is best matched to the specific world you are chasing, which is, when you think about it, exactly how a maturing science is supposed to work.

Register on our portal to dig deeper into exoplanet discovery methods, mission breakdowns, and the search for habitable worlds beyond our Solar System.