# Interstellar destination priorities, revised 11 October 2026 These scores compare destinations under reference travel scenarios. They are not prices for complete spacecraft, or proof that a mission can reach or stop at any particular star. No score requires solar sails or starlight braking. ## Shared assumptions * 20 years from the beginning of development to launch. * Flyby cruise speed: 0.2c. Settlement cruise speed: 0.05c. * Acceleration and deceleration time, propellant, payload, engine efficiency, beam infrastructure, reliability and construction costs are not modelled. * Baseline mission cost is equal across destinations. This removes cost as a ranking factor; it does not claim real missions will cost the same. * Scores are normalized separately: 100 times a raw score divided by the best whole-system raw score. There is no combined score. * Map and CSV ranks compare 2,686 systems. The article splits Alpha Centauri into Proxima and A+B, producing 2,687 destinations on the same score scale. The lineup's ranks compare its nine shortlisted planets. Let d be the distance in light-years, d0 = 4.24 light-years, and D(t) the discount remaining after t years. c denotes light speed. ## Knowledge ``` time to first data = 20 + d/0.2 + d relative communications factor = (d0/d)^2 raw knowledge = science points * relative communications factor * D_standard(time) ``` This link factor assumes the same transmitter, receiver, available transmission energy and signal efficiency at all destinations. It expresses inverse-square spreading. Larger antennas, greater power, different coding or relays could alter it. It is not an absolute data-rate prediction and does not select a sail, a rocket or a particle-beam craft. Science points: 2 per unevolved F/G/K star; 1 per other star; 0.5 per white or brown dwarf. Add 1 per reported planet and a further 4 for a rocky habitable-zone planet. Multiply planet contributions by confidence (confirmed 1, weak 0.5, disputed 0.25, refuted 0), zone and size weights from the input census. A large super-Earth gets half the rocky-world credit. These are editorial weights for potential scientific interest, not a guaranteed itinerary or data yield. Standard annual discount rates: 3% for years 0–30, 2.5% to year 75, 2.25% to year 125; the last rate is held beyond that point as a model assumption. ## Settlement ``` reference arrival time = 20 + d/0.05 raw settlement = estimated twin probability * D_social(arrival time) ``` The twin probability retains the research model's planet-confidence, size, habitable-zone, water, occurrence, binary and stellar-evolution assumptions. It is conditional on those assumptions, not a measured frequency of habitable planets. Multiple contributions are combined as 1 minus the product of their failure probabilities. The water fractions are from Tian and Ida (2015), and the unknown-planet occurrence inputs are 0.24 per F/G/K dwarf and 0.16 per M dwarf. Stellar flares have no quantitative penalty. The article explains the major water and binary assumptions. The source code records the remaining factors. Social-insurance annual discount rates: 1.5% for years 0–30, 1% to year 75, 0.75% to year 125; 0.75% is held beyond that point as a model assumption. There is no distance-squared baseline cost penalty. At fixed mass and speed, kinetic energy does not depend on travel distance. A fixed-duration comparison would instead require speed to increase with distance. Actual mission costs could vary for many other reasons and require a specific spacecraft design. The separate twice-cost threshold is `2/(twin probability * D_social)`, in units of each destination's own mission cost. It is an illustrative investment hurdle, not a market price for a planet or a verified optimum launch policy. ## Scouting ``` raw scouting = sum(testable twin contribution * D_social * observability / test cost) ``` Facilities, resolving-angle gates, test completeness and estimated test costs are retained from the research model: TOLIMAN/METIS for A+B, JWST transit work, ELT/ANDES reflected-light work and the planned Habitable Worlds Observatory. HWO is included as a scenario, not as an operating instrument. One sensitivity case excludes it. The baseline uses $5 million for a small test and $440 million per HWO exo-Earth target as an average programme allocation, not a quoted marginal observation price. ANDES observability uses the source estimates where available and a distance/resolution proxy otherwise. This is a measurement-priority heuristic, not a full decision-theoretic value-of-information calculation. ## Strategy ``` body points = all stellar components + confidence-weighted reported planets debris points = 2 for a documented debris belt, otherwise 0 raw strategy = (body points + debris points) * d0/d ``` Stellar luminosity has no effect on this score. Nearness to Earth and the inventory of potential destinations/materials are explicit preferences. Body count does not establish extractable resources, fuel, available power or the cost of reaching those bodies. The model does not calculate onward routes between stars. It does not equate a planet, a star and a debris belt physically; the points are a coarse comparison of opportunities. ## Sensitivity and reproduction The [machine-readable report](scoring-report.json) contains the top ten under 22 alternative scenarios: slower/faster travel, linear/squared mission-cost growth, no communications-distance penalty, different discount and water assumptions, planet occurrence, confidence/size weights, HWO availability, development time, debris credit, proximity preference and stellar evolution. Cost-growth cases are hypothetical target penalties, not engineering laws. Every [CSV score](scored-systems.csv), the map and the lineup use this baseline. The model and source snapshots are committed under `scripts/interstellar/`. The report records snapshot hashes. Census distances used in the ranking can slightly differ from updated distances in individual planet profiles. Sources: Kirkpatrick et al. 2024 ([doi:10.3847/1538-4365/ad24e2](https://doi.org/10.3847/1538-4365/ad24e2)); NASA Exoplanet Archive, October 2026 research snapshot; Tian and Ida 2015 ([doi:10.1038/ngeo2372](https://doi.org/10.1038/ngeo2372)); Dressing and Charbonneau 2015 ([doi:10.1088/0004-637X/807/1/45](https://doi.org/10.1088/0004-637X/807/1/45)); Kraus et al. 2016 ([doi:10.3847/0004-6256/152/1/8](https://doi.org/10.3847/0004-6256/152/1/8)); Freeman and Groom's June 2026 [discount-rate review](https://assets.publishing.service.gov.uk/media/6a43915e7ac6fd9c6a94abbe/Findings_and_Recommendations_-_Green_Book_Discount_Rate_Review.pdf).