Earth’s total topographic relief, from the bottom of the Mariana Trench to the peak of Mount Everest, spans roughly 20 kilometers. Scaled down to the diameter of a regulation billiard ball, that entire range of mountains and ocean trenches would compress to a bump barely 0.04 millimeters tall. That number sits comfortably within the manufacturing tolerances allowed for a pool ball, which is where the popular claim “the Earth is smoother than a billiard ball” comes from. The claim is not wrong, exactly, but it conflates two different properties of a sphere and skips over some details that make the real comparison far more interesting.
Where the Claim Comes From
The comparison traces back to at least the 1970s and was popularized by Neil deGrasse Tyson and others in public science outreach. The arithmetic is simple. A regulation billiard ball, per World Pool-Billiard Association standards, has a diameter of 2.25 inches (57.15 mm) with an allowed tolerance of ±0.005 inches (±0.127 mm). Earth’s mean diameter is about 12,742 km. The ratio of the billiard ball’s permitted imperfection to its diameter is roughly 1 in 450. The ratio of Earth’s maximum peak-to-trough relief (about 20 km) to its diameter is roughly 1 in 637. Since 1 in 637 is a smaller ratio than 1 in 450, Earth’s biggest bumps, when scaled down, would be smaller than what a billiard ball manufacturer is allowed to leave on the product.
On that basis, the claim holds. But “smoother” is doing a lot of work in that sentence, because the billiard ball tolerance spec is about how round the ball is, not about how its surface feels under your fingertip. Those are fundamentally different things.
Roundness and Surface Smoothness Are Different Measurements
When engineers describe a sphere, they care about at least two separate qualities. The first is sphericity, or how closely the overall shape matches a perfect sphere. The diameter tolerance on a billiard ball is a sphericity spec: the ball should not be lopsided, egg-shaped, or lumpy at the scale of tenths of a millimeter. The second quality is surface roughness, sometimes called surface finish. That is the texture of the surface at microscopic scales: tiny pits, scratches, and grain left behind by the manufacturing and polishing process. A ball can be very round but rough, like a basketball. Or it can be locally smooth but not quite round, like a slightly squeezed rubber ball.
The popular Earth-versus-billiard-ball comparison only addresses the first quality. Earth’s biggest features, scaled down, fit inside the billiard ball’s diameter tolerance. But nobody has ever claimed a billiard ball’s surface roughness tolerance is ±0.005 inches. The actual polished surface of a high-quality phenolic resin billiard ball has a roughness measured in single-digit micrometers or less. Scaled to Earth’s size, that kind of finish would correspond to surface irregularities of a few meters at most. Earth’s surface, even in its flattest regions, is nowhere near that smooth. The Great Plains have local relief of tens of meters. The abyssal ocean floor, often described as flat, is broken by ridges, seamounts, and fracture zones with hundreds of meters of vertical variation. So while Earth passes the billiard ball’s roundness test for peak-to-trough relief, it would feel like sandpaper compared to the ball’s actual polished finish.
Earth Is Not Even Round Enough
There is actually a deeper problem with the comparison that the popular version glosses over entirely. Earth is not a sphere. It is an oblate spheroid, wider at the equator than from pole to pole because of its rotation. The equatorial diameter is about 12,756 km, while the polar diameter is about 12,714 km, a difference of roughly 42.8 km. That equatorial bulge, expressed as a ratio of the mean diameter, works out to about 1 in 298. The billiard ball’s permitted deviation is 1 in 450. Earth’s oblateness alone, ignoring every mountain and trench, exceeds the billiard ball spec by a wide margin.
So if you shrank Earth to billiard ball size while preserving its true shape, the result would be slightly but measurably egg-shaped. You could probably not see it with the naked eye, but a precision gauge would reject it. The ball would be about 0.04 mm wider at the equator than at the poles, which doesn’t sound like much until you realize the total tolerance budget is only 0.127 mm. The bulge eats up roughly a third of that budget before you even add a single mountain.
The popular claim survives only if you assume “smooth” refers strictly to local surface relief and you ignore the overall shape. That is a generous reading.
Earth’s Shape Gets Even More Complicated
The oblate spheroid model is itself an approximation. Earth’s actual shape, called the geoid, is lumpy in ways that reflect the uneven distribution of mass inside the planet. Dense rock in the mantle pulls the ocean surface slightly toward it; lighter material lets it sag. The result is a surface that deviates from a perfect ellipsoid by up to about ±100 meters. That does not sound like much against a 12,742 km diameter, and it isn’t, but it adds another layer of imperfection on top of the bulge and the topography.
Mantle convection is the main driver of these geoid undulations. Plumes of hot, buoyant rock push the surface up in some places; dense, sinking slabs pull it down in others. Recent modeling work has shown that including chemical density variations in the continental lithosphere, not just thermal effects, substantially improves the match between predicted and observed surface topography, suggesting that the lumps are partly a record of how the continents were assembled over billions of years. 1Journal of Geophysical Research: Solid Earth. Modeling Geoid and Dynamic Topography From Tomography‐Based Thermo‐Chemical Mantle Convection These geoid undulations are invisible to the naked eye, and they would vanish into noise at billiard ball scale, but they are a reminder that “the shape of Earth” is not a single number you can plug into a ratio.
What a Miniature Earth Would Actually Feel Like
Suppose you could hold the shrunken Earth in your hand. What would you notice? The oblateness would be too subtle to detect by touch. The geoid undulations would be far below your threshold of perception. But the mountains and ocean trenches, scaled down, would produce features in the range of tens of micrometers, roughly the height of a fingerprint ridge or a fine scratch on a glass lens. Would you feel those?
Probably yes. Human fingertips are remarkably sensitive to surface texture. Research on tactile perception has shown that people can distinguish surface features with amplitudes as small as about 10 nanometers when the features are presented as regular wrinkle patterns on a smooth substrate.2PubMed Central. Feeling small: exploring the tactile perception limits The scaled-down Himalayas, at roughly 40 micrometers, would be about 4,000 times larger than that threshold. You would absolutely feel the mountain ranges. The mid-ocean ridges, the continental shelves, and probably even large individual features like the Hawaiian island chain would register as texture under your fingertip. It would feel less like a polished billiard ball and more like a ball that had been rolled across fine sandpaper, with a few noticeable ridges and scratches.
That said, the oceans complicate things. If the shrunken Earth retained its water, the oceans would form a film roughly 0.01 mm deep on average. Surface tension at that scale would likely smooth the water into a nearly uniform coating, filling in ocean trenches and possibly softening the feel of coastal mountain ranges. The continents would still protrude, but the overall texture might be less dramatic than a dry-Earth model suggests. No one has actually manufactured such a globe, so this remains a thought experiment rather than an empirical finding.
The Most Perfect Spheres Ever Made
If you want a real comparison for Earth’s roundness, billiard balls are the wrong benchmark. The most spherical objects humans have ever manufactured are the silicon-28 spheres created for metrology, specifically for redefining the kilogram through exact atom-counting. These spheres are about 93.6 mm in diameter and are polished to astonishing tolerances. Their surface deviations from a perfect sphere are measured in tens of nanometers, and the relative uncertainty of their volume measurements has been pushed below two parts in a hundred million.3Elsevier / Comptes Rendus Physique. Silicon spheres for the future realization of the kilogram and the mole
To put that in perspective: if you scaled one of these silicon spheres up to the size of Earth, its surface imperfections would be the equivalent of bumps a few centimeters tall. The difference between a billiard ball and a silicon metrology sphere is like the difference between a gravel road and a mirror. And compared to either of these manufactured objects, Earth is a rugged, lumpy, wildly imperfect ball. The popular comparison flatters our planet by choosing the least demanding benchmark available.
The silicon spheres also highlight how different “smooth” and “round” really are. Making them involved not just grinding them into a nearly perfect spherical shape but also polishing the surface to near-atomic flatness and then characterizing the thin oxide layer and any adsorbed contaminants on top. Sphericity and surface finish had to be controlled independently, through completely different processes, because they are fundamentally different physical properties.
How Other Worlds Compare
Earth is actually one of the smoother rocky bodies in the solar system, at least at large scales. Mars has Olympus Mons, which rises about 21.9 km above the surrounding plains, and Valles Marineris, a canyon system up to 7 km deep. Relative to Mars’s diameter of roughly 6,779 km, the peak-to-trough relief ratio is about 1 in 234, worse than Earth and clearly outside billiard ball tolerance. The Moon has a smaller diameter (about 3,475 km) and a maximum relief of roughly 19.8 km between its deepest craters and tallest peaks, giving a ratio of about 1 in 176. The Moon would be an even lumpier billiard ball than Mars.
At finer scales, the Moon is rougher still. Research comparing meter-scale topographic roughness across planetary surfaces has found that the lunar maria, the dark basaltic plains visible from Earth, are rougher at small scales than comparable volcanic surfaces on Earth or Mars, largely because the Moon has accumulated enormous numbers of small impact craters (diameters under a kilometer) over billions of years without erosion to smooth them out.4Journal of Geophysical Research: Planets. Meter‐Scale Topographic Roughness of the Moon: The Effect of Small Impact Craters Earth’s atmosphere, water, and plate tectonics are constantly erasing small-scale roughness. Rain, rivers, glaciers, and wind break down mountains and fill in craters. That erosive recycling is a big part of why Earth’s surface, despite its dramatic mountain ranges, is relatively smooth compared to airless bodies of similar size.
Why the Comparison Keeps Circulating
The billiard ball comparison endures because it delivers a genuine surprise: we think of Earth as a rugged place, full of towering peaks and crushing ocean depths, and learning that those features are proportionally tiny is a real shift in perspective. Everest feels enormous when you are standing at its base, but it protrudes from Earth’s surface by only about 0.07 percent of the planet’s radius. The entire depth of the Mariana Trench is barely 0.17 percent of that radius. At planetary scales, Earth is overwhelmingly a ball of rock and iron with a very thin textured skin.
The comparison also works as a gateway to thinking about scale more carefully. Most people have an intuitive sense of how smooth a billiard ball is: they have held one, felt it, watched it roll on felt. Mapping Earth’s surface onto that familiar object forces a recalibration. The mistake is treating the comparison as a precise scientific claim rather than a useful analogy. It is true in a narrow sense (peak-to-trough relief fits the diameter tolerance), misleading in a broader sense (Earth is oblate and would fail the roundness spec), and outright wrong if you interpret “smooth” as surface finish. Knowing which version of “smooth” is in play is the whole game.
The Role of Water in Hiding Roughness
About 71 percent of Earth’s surface is covered by ocean, and water is extraordinarily effective at hiding topographic roughness. The mid-Atlantic ridge is a mountain chain that runs for over 16,000 km down the center of the Atlantic Ocean, with peaks rising 2 to 3 km above the surrounding seafloor. You would never know it was there by looking at the ocean surface. The ocean floor has fracture zones, abyssal hills, volcanic seamounts, and trenches that collectively produce more topographic variation than the exposed continents, yet from space Earth looks like a smooth blue marble.
If you drained the oceans, the billiard ball comparison would get more complicated. The exposed ocean floor would reveal a surface with enormous relief at all scales, from the tens-of-kilometers depth of the trenches to the meter-scale pillow lavas on fresh mid-ocean ridge basalt. The continents, by contrast, have been worn down by erosion for hundreds of millions of years. A dry Earth would look bimodal: relatively smooth continental platforms surrounded by deeply textured ocean basins. The water fills in the low spots and presents a deceptively uniform surface to the casual observer, or to the back-of-the-envelope calculation that says Earth is smoother than a billiard ball. The comparison implicitly uses the ocean surface as “smooth,” which is convenient but hides the real roughness underneath.
Gas giant planets, by the way, sidestep the whole question. Jupiter and Saturn have no solid surface at all, just progressively denser layers of gas and liquid. Their visible “surface” is the cloud tops, which can be remarkably smooth at large scales. If you allowed gaseous surfaces into the competition, Jupiter might genuinely be smoother than a billiard ball by almost any reasonable definition, though the comparison would be even more metaphorical than the one with Earth.