Estimación con fuentes, cobertura local parcial y referencias geográficas más amplias.
Estimación urbana: 80/100. Más alto significa más seguro en la escala del modelo. Evidencia: · Confianza: D
¿Qué cubre este mapa?
Estimación de investigación: solo una dimensión varía con evidencia local; las demás usan la base de ciudad o país. Las zonas sin color carecen de evidencia local; no necesariamente son seguras.
Localiza el alojamiento y comprueba si su zona tiene evidencia.
Lee qué explica las diferencias locales. Un valor urbano y uno local responden preguntas distintas.
Consulta por separado los consejos actuales y el transporte. Estas estimaciones no evalúan un hotel, calle u hora concretos.
Evidencia de toda la ciudad
Violencia personal
86/100
Patrimonio y estafas
52/100
Movilidad
86/100
Salud y rescate
92/100
Conflicto y estabilidad
76/100
Desastres y entorno
86/100
Respuesta de infraestructura
91/100
Una estimación urbana no evalúa una calle, hotel ni hora concreta.
Fuentes y límites (idioma original)
Compare Madrid districts with an official traffic-safety layer and source-linked Spain safety context.
The local layer varies only the mobility component using injury involvement among police-recorded crashes in 2025. It is not a local crime layer. An injury case has at least one person with source severity code 1–7. Each case ID is counted once. All recorded crashes, including cases with missing severity, remain in the denominator. This conditional percentage is not risk per trip or per resident; reporting and road mix affect comparisons. The city score uses the existing Spain research priors. None of the seven citywide inputs is represented as a Madrid-specific rate. Official district codes join all 21 districts; no neighbourhood or time-of-day model is inferred.
The local layer varies only the mobility component using injury involvement among police-recorded crashes in 2025. It is not a local crime layer. An injury case has at least one person with source severity code 1–7. Each case ID is counted once. All recorded crashes, including cases with missing severity, remain in the denominator. This conditional percentage is not risk per trip or per resident; reporting and road mix affect comparisons. The city score uses the existing Spain research priors. None of the seven citywide inputs is represented as a Madrid-specific rate. Official district codes join all 21 districts; no neighbourhood or time-of-day model is inferred.
The local layer varies only the mobility component using injury involvement among police-recorded crashes in 2025. It is not a local crime layer.
An injury case has at least one person with source severity code 1–7. Each case ID is counted once. All recorded crashes, including cases with missing severity, remain in the denominator.
This conditional percentage is not risk per trip or per resident; reporting and road mix affect comparisons.
The city score uses the existing Spain research priors. None of the seven citywide inputs is represented as a Madrid-specific rate.
Official district codes join all 21 districts; no neighbourhood or time-of-day model is inferred.
does not provide a comparable Barcelona incident rate; U.S.-traveler perspective; advisory level is not an absolute safety score; countrywide terrorism and unrest signals do not imply uniform city exposure
country-level only; modeled perceptions rather than traveler incidents; cannot represent neighborhood variation or current events; confidence intervals must not be discarded
The local layer varies only the mobility component using injury involvement among police-recorded crashes in 2025. It is not a local crime layer.; An injury case has at least one person with source severity code 1–7. Each case ID is counted once. All recorded crashes, including cases with missing severity, remain in the denominator.; This conditional percentage is not risk per trip or per resident; reporting and road mix affect comparisons.; CC BY 4.0; Ayuntamiento de Madrid. Aggregation and research transformation by Safrom. https://datos.madrid.es/pages/condiciones-de-uso