Methodology

MapDash publishes one fresh country-guessing puzzle every UTC day. Each guess returns great-circle distance and an eight-compass-point direction cue. Targets come from a balanced rotation across continents and country-size tiers.

Target country pool

The puzzle's target pool is the set of internationally-recognized sovereign states — about 195 countries. We exclude dependencies, territories, and disputed regions from the target pool. Dependencies and territories remain valid guess inputs (so a player can guess Greenland or Puerto Rico to gain distance/direction information), but they will not be the answer.

The selection logic weights candidate countries inversely to their recent appearance frequency (so players who follow the daily over months see a balanced rotation rather than a steady diet of European countries) and inversely to their global familiarity ranking (so smaller and less-familiar countries appear with intentional regularity rather than being statistically squeezed out).

Distance calculation

MapDash reports the great-circle distance between the centroid of the guessed country and the centroid of the target country. Centroids are computed from OpenStreetMap country geometries using a standard centroid-of-polygon algorithm; for countries with multiple disjoint regions (Russia, Indonesia, Chile), the centroid is the weighted centroid of the combined geometry.

We chose centroid-to-centroid distance over border-to-border distance because border-to-border distances under-discriminate between neighboring countries (adjacent countries report near-zero distance, which conflates structurally different geographic relationships). Centroid distance is also more stable for countries whose borders are subject to ongoing dispute or revision in the OSM dataset.

Direction calculation

Direction feedback is the initial bearing from the guessed country's centroid to the target country's centroid, rounded to one of eight compass points: N, NE, E, SE, S, SW, W, NW. Initial bearing accounts for the curvature of the earth; a long-distance bearing can rotate substantially along the great-circle path, but the published cue reflects only the bearing at the start of that path.

We round to eight points rather than reporting precise degrees because the discrete cue is easier to read at a glance, produces a cleaner share-emoji grid, and is more accessible for screen-reader users. The exact bearing in degrees is available on the post-solve explainer page for any completed puzzle.

Solvability validation

Before publication, the in-house solver attempts to solve each candidate puzzle within the published guess budget using a heuristic that combines distance-narrowing and direction-pivoting moves. Candidates the solver cannot resolve are rejected and regenerated. The guess budget is set so a competent human player using common geography knowledge solves the puzzle about 80% of the time on a first try.

The validator is heuristic, not optimal-information. Expert players using rigorous binary-search-by-distance can sometimes solve in fewer guesses than the validator's heuristic would predict; novice players who guess by name recognition rather than geography will sometimes fail puzzles the validator considers solvable. The 80%-on-first-try target is a typical-player calibration, not a guarantee.

Data sources and licensing

Country geometries and centroids are derived from OpenStreetMap (OSM), licensed under the Open Database License (ODbL). We comply with the ODbL attribution requirement by including OSM credit on every page where OSM data is used, including the homepage and each country-detail page.

OSM's representation of disputed regions follows OSM editorial consensus, which may differ from any specific country's preferred treatment. We document known disputes (Western Sahara, Kashmir, Taiwan/PRC, Crimea, Northern Cyprus) on the relevant country-detail pages so players can see how the dataset represents the contested territory.

Known limits

Three classes of puzzle outcome are worth flagging:

• Centroid distance for non-compact countries (Russia, Indonesia, Chile, USA) can feel unintuitive. The centroid is a single point; the country sprawls.
• Disputed-region boundaries follow OSM consensus and may not match every player's map mental model.
• Direction rounding to eight points hides fine-grained bearing information. Two countries due-east-ish of the guess may both report "E" even when one is 5° north of due-east and the other is 5° south.

When a player reports a puzzle outcome that seems wrong, we check the underlying centroid and bearing data; in most cases the unexpected result is mathematically correct but pedagogically surprising, and we update the strategy-guide page to cover the case rather than alter the puzzle.

Frequently asked questions