The Drake equation is almost always presented with point estimates. Propagate uncertainty correctly — as a Bayesian Monte Carlo over the seven factors — and the question "how many civilisations?" stops being a number and becomes a distribution. Per Sandberg, Drexler & Ord (2018).
N = R* · fp · ne · fl · fi · fc · L
where R* is the rate of suitable star formation, fp is the fraction of stars with planets, ne is the number of habitable planets per such system, fl is the fraction on which life arises, fi the fraction where intelligence evolves, fc the fraction that communicates, and L is the lifetime of the communicating phase in years.
The biology factors (fl, fi) are uncertain across at least ten orders of magnitude — abiogenesis is observationally unconstrained beyond N=1. If you draw one factor from a distribution spanning [10⁻³⁰, 1] and three more from similarly wide distributions, the product distribution is dominated by the low tail. The expected value of N can still be high (the optimistic tail pulls the mean up), but the median and the probability mass at small N are what determine whether the Fermi observation is surprising. Sandberg et al. show that under credible wide priors, P(N<1) in the Milky Way is roughly 30–55%, and P("we're alone in the observable universe") is roughly 10–40%. That is not what most people picture when they hear "Drake equation says ten."
Sandberg 2018 (default): approximates the paper's "current scientific knowledge" prior — moderate uncertainty on biology, tighter on the astronomical factors thanks to exoplanet data. Reproduces the paper's headline result that P(N<1 in MW) is roughly 30–50% rather than the textbook "N ≈ 10". Drake 1961 params: narrow ranges around Drake's 1961 Green Bank values; note the MC gives N ≈ 3,162 with these parameters (not the historical ~10, because different sources reconstruct Drake's exact values differently — the preset uses his documented star formation rate and optimistic biology). Modern optimistic: tightens biology priors toward "life is common", keeps L generous. Modern pessimistic: opens the biology priors out to many decades — closer to the Sandberg "synthetic literature uncertainty" prior, where the answer mass piles up at log N → −∞ and P("alone in observable universe") rises substantially. Custom: whatever you set the sliders to; the URL hash preserves it.
For each of K samples (default 10,000): draw each factor xi log-uniformly from [10ai, 10bi] where [ai, bi] are the slider ranges; compute log₁₀ N = Σ log₁₀ xi; histogram the result; report median, mean (linear), and P(log₁₀ N < 0), P(log₁₀ N < -3). Log-uniform is the standard "scale-uncertain prior" — it's what you use when you don't know the order of magnitude.
The Macro Transcension Hypothesis (MTH) requires the Fermi paradox to be soluble — if N is huge and nobody's expanding outward, you need either a Great Filter or a Transcension explanation. If the Sandberg MC already gives P(N<1) ≈ 0.4 on plausible priors, then the silence is not paradoxical and the MTH is one explanation among several rather than the leading one. Tool 17 (Great Filter Localizer) and Tool 18 (Aestivation) explore the other branches.
The 1961 Drake equation factors N = R* · f_p · n_e · f_l · f_i · f_c · L have aged unevenly. R* (star formation rate): ~1.5 stars/yr Milky Way average (~2 in the past, fewer now). f_p (fraction of stars with planets): essentially 1.0 today from Kepler/TESS — every sun-like star has at least one planet. n_e (habitable planets per system): 0.1–0.5 (debated; Kepler suggests ~22% of sun-like stars have Earth-sized planets in HZ, per Bryson et al. 2021). f_l, f_i, f_c, L remain unconstrained empirically — N depends almost entirely on assumptions about these.
Sandberg, Drexler & Ord 2018 (arXiv:1806.02404, "Dissolving the Fermi Paradox") performed the first rigorous Monte Carlo over the joint Drake-factor uncertainty distribution. Headline result: when you sample factors from their published log-uniform uncertainty ranges, the posterior probability that humanity is alone in the observable universe is ~30% — high enough that the Fermi paradox dissolves as a paradox. The result depends sensitively on f_l and f_i priors (the abiogenesis and intelligence priors).
Before Kepler (launched 2009), Drake factors f_p and n_e were estimated at 0.5 and 0.01 respectively in Sagan-era treatments. Today f_p ≈ 1 and n_e ≈ 0.1–0.5 — a ~1000× upward revision in expected habitable planets. This makes the unconstrained behavioural factors (f_l, f_i, f_c, L) carry essentially all the uncertainty in N. The "rare Earth hypothesis" (Ward & Brownlee 2000) survives by attacking the difficulty of land life, complex multicellular life, and technological civilisation — not the abundance of planets.