More Stars Than Grains of Sand? The Math of Big Estimates

Quick answer: Probably yes for the observable universe and Earth’s beaches, but the answer depends on the assumptions. A widely cited 2003 estimate by Simon Driver’s team puts the stars at about 7 × 10^22 (70 sextillion), and astronomers commonly give a range of about 10^22 to 10^24. A widely circulated estimate of Earth’s sand grains is 7.5 × 10^18, and a beach-only Scientific American model gives about 4 × 10^20. With those numbers stars win by roughly 25 to 10,000 times. But the same Scientific American calculation, assuming a low average of 10 million stars per galaxy, gets only 2 × 10^19 stars, so sand wins. These are models, not counts.

Facts checked November 22, 2026.

“There are more stars in the universe than grains of sand on all the beaches of Earth.” Carl Sagan popularized the comparison in his 1980 book and TV series Cosmos, and it shows up everywhere. Is it true? Nobody has counted either one, so the honest answer comes from estimation: break a huge number into a few factors you can roughly measure, then multiply. Here is how to do it for both, why careful people get different answers, and a calculator where you choose the assumptions.

The Short Answer

EstimateStarsSand grainsWinner
Driver (2003) stars; widely circulated sand estimate reported by NPR (2012)about \(7 \times 10^{22}\)about \(7.5 \times 10^{18}\)Stars, by about 9,300×
Low end of the usual star range; Scientific American beach model\(10^{22}\)about \(4 \times 10^{20}\)Stars, by about 25×
High end of the usual star range; same beach model\(10^{24}\)about \(4 \times 10^{20}\)Stars, by about 2,500×
Phil Plait’s simplified Scientific American model (2024)about \(2 \times 10^{19}\)about \(4 \times 10^{20}\)Sand, by about 20×

These are independently published models, not measurements, and they do not all count the same thing: the NPR figure is for Earth’s sand in general, while the Scientific American figure is beaches only. The Plait model does not prove sand wins; it shows how sensitive the answer is to one assumption, the average size of a galaxy. Note also that “the universe” here means the observable universe, the part whose light can reach us.

Estimating the Stars

Count galaxies, then multiply by the average number of stars per galaxy:

\[ \text{stars} \approx (\text{number of galaxies}) \times (\text{average stars per galaxy}). \]
  • Galaxies: a 2016 study led by Christopher Conselice estimated about 2 trillion galaxies in the observable universe, most of them too small and faint to see individually. Hubble directly detects only a fraction of these; estimates range from hundreds of billions to about 2 trillion.
  • Stars per galaxy: our Milky Way has a few hundred billion stars, but most galaxies are small dwarfs with far fewer. The average is very uncertain, anywhere from about 10 million to 100 billion depending on which galaxies you count.

For example, \(2 \times 10^{11}\) galaxies \(\times\ 10^{11}\) stars each \(= 2 \times 10^{22}\) stars. With 2 trillion galaxies but only 10 million stars on average, you get \(2 \times 10^{19}\). That one factor, the average galaxy size, moves the answer by a factor of a thousand.

Estimating the Sand

Estimate the volume of beach sand, then divide by the volume of one grain:

\[ \text{grains} \approx \frac{\text{beach length} \times \text{width} \times \text{depth} \times \text{packing fraction}}{\text{volume of one grain}}. \]

Phil Plait’s Scientific American calculation used about 2.5 million km of coastline, 30% of it sandy (750,000 km), a beach cross-section of about 500 square meters (width times depth), giving 375 billion cubic meters, and about 1 billion grains of 1 mm sand per cubic meter: roughly \(4 \times 10^{20}\) grains. Grain size matters enormously: grains 0.1 mm across are 1,000 times smaller in volume, so the count rises 1,000-fold. In general the number of grains scales as \(1/d^3\): halve the grain diameter and you get 8 times as many grains. About 30% sandy shoreline matches satellite studies (Luijendijk and colleagues, 2018, found about 31% of ice-free coastlines sandy), but the beach width and depth are illustrative guesses.

And “beaches” is a choice. Add deserts, riverbeds and the sea floor and the sand count rises by orders of magnitude, possibly past many star estimates.

Try It: Choose Your Assumptions

Stars

Beach sand

The sand model treats grains as spheres filling 60% of the beach volume. Try small dwarf-galaxy averages (10 million stars) with fine 0.2 mm sand, and the slogan fails; try Milky Way-sized galaxies, and stars win easily.

Did New Horizons Find Fewer Galaxies?

No, despite some headlines. In 2021 the New Horizons spacecraft, far beyond Pluto, measured the faint glow of the “cosmic optical background.” The original news release suggested this meant fewer galaxies, but it was corrected: the measurement constrains the total light from galaxies, not their number. A 2024 follow-up, after better removing light scattered by dust in the Milky Way, found the glow consistent with the light expected from known galaxies.

The Real Skill: Fermi Estimation

Physicist Enrico Fermi was famous for answering impossible-sounding questions by multiplying rough factors. The trick works because errors partly cancel and because, for huge numbers, getting the right power of ten is what matters. Three habits help:

  • Write the formula first. Stars = galaxies × stars per galaxy; sand = volume ÷ grain volume.
  • Use powers of ten. Round each factor to 1, 2, 5 or 10 times a power of ten, then add exponents.
  • Find the most uncertain factor. Here it is average galaxy size and grain size. That is where better data changes the answer.

For a full guide with more worked examples, see Fermi problems explained, and for even bigger quantities, the biggest numbers in math and 52 factorial, a number larger than both stars and sand combined.

Other Comparisons

  • Stars visible to the naked eye on a dark night: only a few thousand.
  • Water molecules in a few drops of water: about \(10^{21}\) to \(10^{22}\), comparable to all the stars.
  • Ways to shuffle a deck of cards: about \(8 \times 10^{67}\), far more than either.

Common Misconceptions

“Someone counted the stars.” No. Star totals are estimates built from counts in small patches of sky, scaled up with models.

“The stars win by an enormous margin, for sure.” Not for sure. Under some reasonable assumptions sand wins; the comparison depends on choices like grain size and average galaxy size.

“Estimates this rough are useless.” No. Knowing a number to within a factor of 10 is often enough to answer the question that matters.

Watch: Stars vs. Sand

If you learn better by watching, these videos are worth your time.

Are There More Grains of Sand Than Stars? (Fraser Cain). An astronomer walks through both estimates.
How Many Stars Are There? (Be Smart). How astronomers estimate the number of stars in the universe.

Check Your Understanding

Five quick questions to see how much of this has stuck. Pick an answer and press Submit; you will see a short explanation either way.

Answer each question and press Submit to see your result right away.

1. How do astronomers estimate the number of stars?
Found a bug? Let us know!
2. Which factor changes the star estimate most?
Found a bug? Let us know!
3. If sand grains are 0.1 mm instead of 1 mm, the count is about…
Found a bug? Let us know!
4. About how many stars are visible to the naked eye?
Found a bug? Let us know!
5. What is Fermi estimation?
Found a bug? Let us know!

Frequently Asked Questions

Are there more stars than grains of sand on Earth?

Most estimates say yes: about 10^22 to 10^24 stars versus about 10^18 to 10^20 grains of beach sand. But with conservative galaxy sizes and fine sand, sand can win, so the answer depends on assumptions.

How many stars are in the observable universe?

Estimates range from about 10^22 to 10^24. One widely quoted 2003 estimate is 7 × 10^22.

How many grains of sand are on Earth?

Estimates vary widely. A figure of about 7.5 × 10^18, attributed to University of Hawaii researchers, was reported by NPR in 2012; a beach-only Scientific American model gives about 4 × 10^20. Including deserts and the sea floor would raise totals.

How many galaxies are there?

Estimates range from hundreds of billions to about 2 trillion in the observable universe; a 2016 study by Conselice and colleagues estimated about 2 trillion, most too faint to see individually.

How can you estimate huge numbers?

Write the quantity as a product of factors you can estimate, round each to a power of ten, and multiply. This is called Fermi estimation.

Keep Exploring

Sources and Further Reading

Related to This Article

What people say about "More Stars Than Grains of Sand? The Math of Big Estimates - Effortless Math"?

No one replied yet.

Leave a Reply