The paradox
The galaxy has billions of worlds that could host life, and billions of years of head start. In 1975 Michael Hart argued that one civilization expanding at a tenth of light speed could cross the galaxy in about two million years. So it should only take one.
That's why this calculator has two ways to notice someone. We can catch a signal from a neighbor alive right now, or we can see the traces of a spreader from any point in history. Run the standard numbers and the second one makes silence look very strange.
The Ore Filter
My theory: the usual chain jumps from intelligence to technology. It skips a step. Every element exists somewhere on a planet, but a species needs them in a form it can reach with stone tools and fire.
Life → Intelligence → Mineable ore → Industry → Spaceflight
Asteroids could fix scarcity later, but you need industry to get there. Rhenium is the good test case: it's rare, most of it goes into turbine alloys, and rockets don't strictly need it. No single element is the gate. The gate is whether easy iron, copper and fuel exist early enough to bootstrap. The Drake term fc has been quietly hiding that assumption.
The math
Every slider is a guess, not a measurement. The ore gates are illustrative: 0.5 × 0.15 × 0.03 × 0.10 is about 1 in 4,444. With the defaults, that takes 800,000 civilizations ever down to about 180.
The chance of detection is 1 − e−λ, the Poisson chance of at least one hit, where λ is the expected number of civilizations we'd notice. Signal λ is alive-now civilizations × share of the galaxy in range × search coverage × share not hiding. Spreader λ is civilizations ever × share that spread × share not hiding.
Two things to keep in mind. Filters can overlap: Ore's geology gate and Rare Earth both lean on plate tectonics, so running both may count it twice. And Sandberg, Drexler and Ord showed that single-number Drake estimates mislead, because the real uncertainty spans many orders of magnitude. Using full distributions, they put the chance we're alone in the galaxy at 53% to 99.6%. Treat this as a toy for seeing which assumptions matter. Press DICE to feel how wide that range is.
The cartridges
Left out on purpose
Grabby aliens (Hanson et al. 2021) work across many galaxies, so they don't fit a Milky Way dial. Planetarium and simulation ideas have no number to slide. "They're already here" is a different website.
Is the ore idea out there already?
In pieces. Stern and Gerya put geology in the Drake equation. Balbi and Frank made smelting a requirement. Robert Hazen's mineral evolution work shows the loop that makes this interesting: most of Earth's roughly 4,300 mineral species depend directly or indirectly on oxygen, and the banded iron formations hold trillions of tons of iron ore. Life helped build the ore that intelligence later needed. I found fossil-fuel versions of the argument online, but I didn't find a paper treating accessible ore as its own filter with its own number.
It also isn't a full resolution. With the defaults, the ore filter drops the odds from near certain to about 2%, because one spreader anywhere in history would still be enough. Move the spreader slider up and the paradox comes back.
Sources
- Drake equation: term definitions, Drake's 1961 ranges and current R★ values.
- Sandberg, Drexler & Ord 2018, Dissolving the Fermi Paradox.
- Ćirković 2009, Fermi's Paradox: the Fermi-Hart timescale of 10⁶ to 10⁸ years.
- Wright, Kanodia & Lubar 2018: SETI has searched about a hot tub's worth of the ocean.
- Sheikh et al. 2025, Earth Detecting Earth: radio beats other technosignatures by about 1,000× in range.
- Balbi & Frank 2023, The oxygen bottleneck for technospheres.
- Stern & Gerya 2024, continents, oceans and plate tectonics.
- Hazen et al. 2008, Mineral evolution.
- NOVA on Hazen: two-thirds of minerals and banded iron ore.
- Snyder-Beattie et al. 2021, timing of evolutionary transitions.
- Sandberg, Armstrong & Ćirković 2017, the aestivation hypothesis.
- Hanson et al. 2021, grabby aliens.