World · Portugal · Lisbon · Areas
Best areas to live in Lisbon: map by score
indicative score new includes estimatesAn indicative 0–10 score for each of 73 parishes (freguesias) of the municipality of Lisbon and seven surrounding municipalities (Oeiras, Cascais, Sintra, Amadora, Odivelas, Loures and Almada), 8 of the 18 in the Lisbon Metropolitan Area. The score is relative to the region itself: for almost every item, 10 is one of the best parishes and 0 one of the worst (air uses a fixed scale, see method). The page looks at the environment close to home: beach or nature, restaurants and bars within walking distance, safety, rent, green space and air. Transport and schools are not included yet.
New map: numbers will change when newer data come out. Data collected on 12 November 2026.
Map of the parishes
Colour the map by
Top group (statistical tie)
Read this as a group, not a podium: #1 to #5 are only 0.16 points apart, and each of them reaches 1st, 2nd or 3rd place in some test variant (weights, crime denominator, anchors, air, nature, with or without Sintra). The range next to each score shows the highest and lowest position.
All parishes
| 1 | 6.05 | 1–5 | 5.0 | 2.6 | 9.7 | 6.0 | 10.0 | 5.7 | 11.40 | |
| 2 | 6.03 | 1–6 | 5.0 | 4.0 | 8.3 | 5.3 | 10.0 | 6.2 | 12.21 | |
| 3 | 5.96 | 1–5 | 7.0 | 5.5 | 8.4 | 2.4 | 5.7 | 4.3 | 15.95 | |
| 4 | 5.92 | 3–8 | 4.1 | 5.1 | 9.5 | 3.7 | 8.8 | 5.9 | 14.20 | |
| 5 | 5.89 | 2–7 | 7.1 | 5.4 | 7.1 | 5.2 | 4.3 | 2.7 | 12.34 | |
| 6 | 5.83 | 2–9 | 7.0 | 4.5 | 8.0 | 3.9 | 5.8 | 3.5 | 13.84 | |
| 7 | 5.74 | 3–11 | 7.0 | 5.5 | 8.7 | 1.8 | 3.0 | 5.6 | 16.92 | |
| 8 | 5.58 | 5–23 | 5.0 | 0.1 | 9.4 | 7.7 | 10.0 | 5.5 | 9.77 | |
| 9 | 5.54 | 7–19 | 7.0 | 4.2 | 7.7 | 4.5 | 3.1 | 4.4 | 13.12 | |
| 10 | 5.51 | 6–19 | 7.1 | 5.4 | 5.9 | 5.5 | 2.5 | 3.2 | 11.96 | |
| 11 | 5.50 | 8–22 | 4.1 | 5.8 | 9.8 | 0.9 | 6.3 | 6.1 | 18.42 | |
| 12 | 5.45 | 9–21 | 5.0 | 4.5 | 7.3 | 5.5 | 5.0 | 5.7 | 12.00 | |
| 13 | 5.45 | 8–23 | 7.1 | 4.7 | 6.2 | 5.2 | 3.0 | 3.5 | 12.25 | |
| 14 | 5.42 | 10–26 | 5.0 | 4.2 | 7.2 | 6.0 | 4.6 | 6.2 | 11.47 | |
| 15 | 5.42 | 10–25 | 7.1 | 5.8 | 6.8 | 3.8 | 1.4 | 2.7 | 13.99 | |
| 16 | 5.38 | 11–27 | 7.1 | 5.6 | 6.8 | 3.5 | 2.0 | 3.0 | 14.47 | |
| 17 | 5.35 | 11–28 | 5.0 | 4.9 | 6.9 | 5.1 | 5.3 | 3.7 | 12.39 | |
| 18 | 5.34 | 12–29 | 5.0 | 4.3 | 6.8 | 5.5 | 6.1 | 4.4 | 11.92 | |
| 19 | 5.32 | 14–31 | 5.0 | 4.6 | 7.1 | 5.9 | 3.4 | 5.4 | 11.54 | |
| 20 | 5.31 | 15–32 | 5.0 | 4.9 | 7.2 | 5.5 | 3.5 | 4.3 | 12.01 | |
| 21 | 5.31 | 12–33 | 7.1 | 5.8 | 6.3 | 4.1 | 0.8 | 2.9 | 13.64 | |
| 22 | 5.20 | 10–36 | 1.0 | 5.6 | 10.0 | 2.0 | 8.9 | 7.4 | 16.51 | |
| 23 | 5.17 | 15–37 | 7.0 | 3.6 | 8.1 | 3.7 | 0.8 | 4.8 | 14.14 | |
| 24 | 5.17 | 18–38 | 5.0 | 4.5 | 7.1 | 6.3 | 1.2 | 6.3 | 11.11 | |
| 25 | 5.14 | 13–40 | 1.0 | 6.7 | 7.9 | 5.1 | 3.8 | 10.0 | 12.44 | |
| 26 | 5.12 | 17–44 | 4.1 | 6.2 | 9.1 | 1.4 | 2.1 | 5.9 | 17.50 | |
| 27 | 5.12 | 15–43 | 7.4 | 4.3 | 5.2 | 5.3 | 1.8 | 3.5 | 12.20 | |
| 28 | 5.07 | 18–46 | 7.4 | 5.0 | 4.4 | 4.0 | 2.7 | 3.8 | 13.73 | |
| 29 | 5.06 | 20–47 | 4.1 | 5.4 | 8.7 | 4.0 | 0.6 | 5.7 | 13.81 | |
| 30 | 5.05 | 19–48 | 7.4 | 3.8 | 4.5 | 6.0 | 2.5 | 3.8 | 11.39 | |
| 31 | 5.02 | 17–49 | 7.4 | 5.0 | 3.8 | 4.7 | 2.4 | 4.0 | 12.84 | |
| 32 | 5.02 | 20–50 | 1.0 | 7.1 | 7.5 | 4.0 | 4.0 | 9.9 | 13.72 | |
| 33 | 4.95 | 24–51 | 1.0 | 4.6 | 9.4 | 3.9 | 6.8 | 8.0 | 13.84 | |
| 34 | 4.92 | 23–53 | 1.0 | 5.0 | 9.1 | 5.3 | 4.1 | 8.4 | 12.14 | |
| 35 | 4.82 | 1–39 | 0.3 | 7.1 | 7.2 | 1.6 | 10.0 | 5.5 | 17.26 | |
| 36 | 4.70 | 5–44 | 0.3 | 8.9 | 6.4 | 0.4 | 8.1 | 4.8 | 19.23 | |
| 37 | 4.62 | 8–52 | 0.3 | 6.7 | 7.5 | 0.8 | 10.0 | 4.6 | 18.47 | |
| 38 | 4.58 | 10–48 | 0.3 | 8.3 | 6.1 | 0.8 | 8.7 | 4.1 | 18.48 | |
| 39 | 4.53 | 30–57 | 5.0 | 2.1 | 5.3 | 8.2 | 2.3 | 4.1 | 9.26 | |
| 40 | 4.52 | 32–58 | 5.0 | 0.9 | 8.2 | 7.4 | 0.1 | 5.8 | 10.00 | |
| 41 | 4.46 | 12–51 | 0.3 | 6.5 | 8.0 | 1.5 | 7.0 | 4.8 | 17.42 | |
| 42 | 4.41 | 35–59 | 5.0 | 0.1 | 5.2 | 10.0 | 3.3 | 5.0 | 7.39 | |
| 43 | 4.39 | 30–60 | 5.5 | 6.2 | 3.1 | 3.1 | 1.9 | 3.9 | 14.95 | |
| 44 | 4.38 | 36–61 | 5.0 | 0.0 | 6.7 | 10.0 | 0.0 | 5.6 | 7.85 | |
| 45 | 4.37 | 13–53 | 0.3 | 10.0 | 6.0 | 0.0 | 3.6 | 4.7 | 21.55 | |
| 46 | 4.32 | 15–55 | 0.3 | 9.2 | 6.5 | 0.0 | 3.4 | 5.7 | 20.00 | |
| 47 | 4.31 | 16–56 | 0.3 | 10.0 | 5.5 | 1.9 | 1.3 | 4.3 | 16.67 | |
| 48 | 4.29 | 18–57 | 0.3 | 8.7 | 5.0 | 2.7 | 4.2 | 4.5 | 15.57 | |
| 49 | 4.26 | 34–63 | 5.5 | 4.2 | 2.8 | 7.1 | 0.1 | 4.0 | 10.31 | |
| 50 | 4.22 | 36–64 | 5.5 | 4.0 | 3.2 | 5.4 | 2.0 | 3.8 | 12.10 | |
| 51 | 4.21 | 22–59 | 0.3 | 7.0 | 5.9 | 2.1 | 6.8 | 3.9 | 16.36 | |
| 52 | 4.19 | 24–60 | 0.3 | 10.0 | 6.0 | 0.0 | 1.2 | 5.8 | 19.95 | |
| 53 | 4.18 | 39–65 | 5.5 | 4.6 | 3.4 | 4.5 | 1.1 | 3.9 | 13.19 | |
| 54 | 4.15 | 41–66 | 5.0 | 0.0 | 6.8 | 8.1 | 0.5 | 5.4 | 9.38 | |
| 55 | 4.10 | 30–61 | 0.3 | 8.9 | 5.5 | 1.0 | 3.1 | 4.6 | 18.26 | |
| 56 | 4.05 | 41–68 | 5.5 | 4.9 | 2.9 | 3.8 | 1.2 | 4.0 | 14.02 | |
| 57 | 4.00 | 34–62 | 0.3 | 6.3 | 6.4 | 2.2 | 5.9 | 2.9 | 16.28 | |
| 58 | 3.97 | 35–66 | 0.3 | 9.7 | 5.7 | 0.0 | 0.2 | 6.0 | 19.92 | |
| 59 | 3.96 | 42–69 | 5.5 | 0.5 | 3.2 | 6.9 | 5.9 | 4.3 | 10.53 | |
| 60 | 3.90 | 41–66 | 0.3 | 9.0 | 5.0 | 1.9 | 0.5 | 4.3 | 16.77 | |
| 61 | 3.90 | 39–67 | 0.3 | 9.5 | 5.2 | 0.9 | 0.0 | 5.5 | 18.44 | |
| 62 | 3.89 | 42–66 | 0.3 | 8.0 | 4.9 | 2.2 | 2.6 | 4.6 | 16.24 | |
| 63 | 3.86 | 45–67 | 0.3 | 7.7 | 4.7 | 2.4 | 3.4 | 4.6 | 15.99 | |
| 64 | 3.78 | 51–70 | 5.5 | 0.8 | 1.7 | 8.5 | 4.0 | 3.8 | 9.00 | |
| 65 | 3.78 | 46–71 | 5.5 | 2.9 | 1.3 | 6.9 | 2.0 | 3.9 | 10.45 | |
| 66 | 3.76 | 49–72 | 5.5 | 2.2 | 0.4 | 8.5 | 2.7 | 3.8 | 9.00 | |
| 67 | 3.69 | 52–68 | 0.3 | 6.2 | 4.9 | 3.0 | 4.2 | 4.2 | 15.04 | |
| 68 | 3.54 | 54–70 | 0.3 | 6.5 | 5.5 | 1.8 | 2.8 | 3.6 | 16.86 | |
| 69 | 3.51 | 53–73 | 5.5 | 1.8 | 1.6 | 7.7 | 0.2 | 3.9 | 9.72 | |
| 70 | 3.39 | 55–71 | 0.3 | 5.0 | 4.2 | 4.3 | 3.6 | 4.1 | 13.39 | |
| 71 | 3.33 | 56–71 | 0.3 | 6.3 | 4.3 | 0.9 | 4.6 | 4.2 | 18.33 | |
| 72 | 3.09 | 57–72 | 0.3 | 6.1 | 3.7 | 2.0 | 2.5 | 4.0 | 16.58 | |
| 73 | 2.82 | 58–73 | 0.3 | 6.4 | 3.1 | 0.6 | 2.2 | 3.9 | 18.88 |
*Range = highest and lowest position across the 30 variants of the sensitivity test (see method).
How the score works
- Unit. Parish (freguesia, CAOP 2025, DGT): the 24 in the municipality of Lisbon, all those in the municipalities bordering it (Oeiras, Amadora, Odivelas, Loures) and those in Cascais, Sintra and Almada. 73 parishes in 8 of the 18 municipalities of the Lisbon Metropolitan Area. It is the smallest unit with official rent (INE) and Census data; crime is only published by municipality.
- Population weighting. Distances, counts, green space and air are measured in every populated ~1 km cell (GHSL 2025) and then averaged by population.
- Safety (25%). Robbery (50%) and theft (50%) recorded by police (DGPJ, published by INE), 2023 to 2025, per 100,000 people present per year, by municipality. *Robbery = street robbery and bag snatching; *theft = theft of and from vehicles (the only robbery and theft categories INE publishes by municipality; residential burglary is not available). Homicide (2021–2025) is shown in the panel but left out of the base score: with few cases and many years hidden by INE, the differences between municipalities are small and uncertain; the sensitivity test includes it with a 20% weight.
- Present population estimate. Crime happens where people are, not only where they live. Denominator = residents (INE annual estimates) + 2021 Census commuting balance (people coming in minus people going out of the municipality to work or study) × 8/24 (one 8-hour working day) + nights in tourist accommodation (INE) ÷ 365. Lisbon ends up 25% above its residents (commuting balance of 354,658 people and 47,373 guests per night); the other municipalities sit between 6% below and 2% above. Low confidence: commuting is from 2021 and tourist nights leave out small short-term lets and day visitors. The test uses the full balance (factor 1) and residents only.
- Safety scale. Logarithmic, anchored at the 5th and 95th percentiles of the 18 municipalities of the Lisbon Metropolitan Area (Grande Lisboa and Península de Setúbal): 10 at 36 robberies and 171 thefts per 100,000, 0 at 162 and 349. This way the score does not depend on which municipality is 2nd worst among the 8 on the map.
- Walkable leisure (25%). Restaurants, cafés, bars, theatres, cinemas, museums and galleries from OpenStreetMap within 2 km of home, equal weight. Log scale.
- Beach or nature (20%). 60% beach and 40% nature. Score 10 up to 1 km, falling linearly to 0 at 25 km. Beach = bathing water identified by APA on the ocean and the Tagus mouth (from Caxias westwards and on the Costa da Caparica), rated Excellent, Good or Sufficient in the latest classification (2024 season, APA data at the European Environment Agency; none of the region's 65 was Poor); the Albufeira lagoon and the Lizandro river stretch are left out. Nature = national-level Natural Park or Protected Landscape in the RNAP (ICNF) of 1,000 ha or more: Sintra-Cascais and Arrábida Natural Parks and the Costa da Caparica Fossil Cliff Protected Landscape. The Tagus and Sado Estuary Nature Reserves (wetlands with limited access) only enter the sensitivity test.
- Cost (15%). Median rent per m² of new leases in the 12 months to June 2026, by parish (INE, local housing rent statistics). Cheaper means a higher score (log scale). Purchase prices are not included.
- Green space (10%). Share of the land within ~1 km of home covered by parks, gardens, reserves and woods in OpenStreetMap. River and sea are left out (land mask = CAOP parishes). 30% or more scores 10.
- Air (5%). Mean PM2.5 for 2023–2025 (validated data from APA's QualAr network): for each station, the mean of the years with 75% or more of hours in every quarter. In each population cell, the 3 nearest stations weighted by 1/distance²; then the population-weighted mean. Fixed scale: 10 at 5 µg/m³ (WHO 2021 guideline) and 0 at 10 µg/m³, the annual limit value the EU set for 2030. The legal limit in force today is 25 µg/m³, and every parish is below it.
- Scale of the other items. For leisure, cost and green space, 0 and 10 are the 2nd worst and 2nd best parish; beach, nature and air use fixed scales.
- Population in the panel. It is the 2021 Census figure by parish. Crime rates use INE's annual estimates by municipality, which in 2026 moved to the Resident Population Base (administrative records) and were revised up, mostly for foreign residents: that is why the 2023–2025 average sits 9% to 25% above the 2021 Census in the 8 municipalities. For parishes created in the 2025 reorganisation, the Census population was split using GHSL (estimate).
- Sensitivity test. The score was recomputed in 30 variants: each weight 5 percentage points up and down; crime with the full commuting balance and per resident only; the old anchors (2nd worst and 2nd best of the 8 municipalities); homicide with a 20% weight (with INE's hidden values set to 0, 1, 1.5 or 2 and with 0.5 and 2 million person-years of smoothing); beach without Caxias and Paço de Arcos; nature with the estuary reserves or natural parks only; PM2.5 for 2025 only, without Laranjeiro, without Mem Martins or background stations only; without Sintra; and Lisbon municipality's safety at the median of the 8 municipalities (5.27). #1 to #5: 0.16 points; #1 to #10: 0.54.
- Top group. Parishes within 0.25 points of #1 that reach the top 5 in some variant form a statistical tie: Colares, Sintra (Santa Maria e São Miguel, São Martinho e São Pedro de Penaferrim), Algés, Linda-a-Velha e Cruz Quebrada-Dafundo, Alcabideche, Alfragide, Carnaxide e Queijas. The order within the group is not robust.
- Final score. Weighted average of the six items. The score compares parishes with each other, not with other cities in the ranking.
Gaps and estimates
- Crime by municipality only. PSP and DGPJ do not publish crime by parish; every parish in a municipality gets the same safety score. Within Lisbon, the difference between, say, Santa Maria Maior and Carnide does not show.
- Present population estimate, low confidence (method above). Fallback, had the estimate been rejected: treat Lisbon's safety as not comparable; that case is in the test (Lisbon safety at the median).
- Almada. Theft per person is the highest of the 18 municipalities (440 per 100,000). Day visitors to the Costa da Caparica may inflate the rate, but there is no official count of visitors by municipality; the present population does not fix this (more people leave Almada for work than come in).
- Hidden homicide counts estimate: when the number is low, INE publishes "x" (not available). No 0, 1 or 2 appears anywhere in the homicide table, so "x" covers 0 to 2 and 1 (the midpoint) was used in: Almada, Amadora, Cascais, Loures, Odivelas, Oeiras, Sintra. Homicide is out of the base score; the test uses 0, 1.5 and 2.
- Air estimated for 68 of 73 parishes estimate: it only counts as measured when a valid station is inside the parish (Alfragide, Algueirão-Mem Martins, Laranjeiro e Feijó, Alvalade, Olivais); the panel shows the station value. Low confidence: Colares, Alcabideche, Oeiras e São Julião da Barra, Paço de Arcos e Caxias, São João das Lampas, Cascais e Estoril, Costa da Caparica, Carcavelos e Parede, Almada, Cova da Piedade, Pragal e Cacilhas, Charneca de Caparica e Sobreda, Caparica e Trafaria, Almargem do Bispo, Loures, Montelavar, Lousa, Fanhões, Bucelas, Santo Antão e São Julião do Tojal. On the south bank only Laranjeiro has all three years valid; its series fell from 7.3 µg/m³ in 2022 to 5.0–4.7 since 2023, with no published explanation, and Paio Pires (7.9 to 8.9) only has 2023 and 2024 valid. The test runs without Laranjeiro and without Mem Martins. Years left out for quarterly coverage under 75%: Paio Pires, Alfragide/Amadora, Camarinha, Fernando Pó, Lourinhã.
- Census population split estimate for the Sintra parishes created in 2025: Almargem do Bispo, Belas, Montelavar, Pêro Pinheiro, Queluz, São João das Lampas, Terrugem.
- Rent with a small sample (under 100 leases in 12 months): Terrugem, Pêro Pinheiro, Montelavar, Lousa, Fanhões, Bucelas, Santo Antão e São Julião do Tojal.
- Partial robbery and theft. By municipality, INE only publishes street robbery and bag snatching and theft of and from vehicles; residential burglary and other robberies are not included.
- OpenStreetMap. Leisure and green counts depend on what is mapped.
- Not included yet: public transport, schools, noise, earthquake or flood risk and purchase prices.
Sources
- DGT — Official Administrative Map of Portugal (CAOP 2025), parishes
- INE — 2021 Census, resident population by parish
- INE — Annual resident population estimates by municipality (2021–2025)
- INE / DGPJ — Crimes recorded by police by municipality and category (2021–2025)
- INE — Median rent per m² of new leases, 12 months to June 2026, by parish
- INE — New leases, last 12 months, by parish
- INE — 2021 Census, population commuting into and out of the municipality
- INE — 2021 Census, share of residents commuting out of the municipality
- INE — Nights in tourist accommodation per 100 inhabitants, by municipality (2022–2025)
- APA — QualAr, validated 2023–2025 PM2.5 data and stations
- APA — Identified bathing waters (INSPIRE AM_BathingWaters)
- APA / European Environment Agency — bathing water quality classification (WISE, Directive 2006/7/EC)
- ICNF — National Network of Protected Areas (RNAP)
- OpenStreetMap (Geofabrik Portugal extract, 12/11/2026), ODbL — leisure, parks and neighbourhood names
- JRC — GHS-POP R2023A, epoch 2025, 30″
- Map: OpenFreeMap, OpenMapTiles and OpenStreetMap; MapLibre GL JS
Updated
Share this page
Card for Instagram and TikTok
Download the image and post it to your feed, story or video. To send people to the site, use the link sticker in your story or the link in your bio (links in captions are not clickable).
Download card (PNG 1080×1350)