Is Mexico Safe in 2026? What Data Actually Says About Travel Right Now

Safe Travel Team · July 27, 2026

Is Mexico Safe in 2026? What Data Actually Says

Search "is mexico safe" today — July 27, 2026 — and the top results are almost uniformly five-and-a-half months old. They lead with the February 2026 El Mencho operation, the Jalisco and Sinaloa flare-ups, and the cascading advisories that followed. They are real events. They are also not what is happening in Mexico right now, and they are doing measurable damage to a tourism economy and a country that, by the data, is having one of its safest travel windows in years.

This post is the data-first answer to the question every search engine keeps mis-asking. It uses:

• SESNSP 2026 first-half data (January–June 2026, the most recent complete release from Mexico's Secretariado Ejecutivo del Sistema Nacional de Seguridad Pública) covering 1.5 million+ incident records across all 53 cities in our assessment engine.


• The 2026 World Cup economic and tourism outcomes — official figures released in early July by Mexico's Secretaría de Turismo and host-city authorities.


• State-by-state SafeTravel risk scores from our assessment engine, anchored to those SESNSP releases.

If you only have time for one section, jump to "What the 2026 numbers actually show" below. If you want the city-by-city picture, our Mexico safety assessment engine processes your itinerary against the same dataset.

Why your search results are wrong right now

Three forces are colliding to put a six-month-old picture in front of new searches.

1. The El Mencho operation (February 2026) was genuinely destabilizing. A federal operation in Jalisco triggered multi-state cartel responses through March. SESNSP recorded the spike accurately. US and Canadian embassies issued travel advisories. Headlines wrote themselves.
2. Search engines reward recency at the top of the SERP — until the steady-state refresh. Google and Bing typically re-rank a major-incident query when new authoritative coverage accumulates. For "is mexico safe," that refresh is being held back by the relative quiet of the past 90 days (which is itself the news — see below).
3. Travel content authors copy the SERP, not the data. Most "is mexico safe" articles published in 2026 cite SESNSP figures from 2024, not 2026. The 2024 annual figures were real, but they are a year and a half out of date and they predate both the El Mencho spike and its resolution.

The combination is a search results page that is structurally out of date, on a question that changes by month. The fix is the same one we built the rest of this site around: pull current data, anchor to sources, and check the city you're actually going to.

What the 2026 numbers actually show

The SESNSP first-half 2026 dataset (released July 17, 2026) covers January 1, 2026 through June 30, 2026 across all 32 Mexican states and 2,469 municipalities. Below is what the data shows when you read it carefully, plus the comparison context that the headlines skip.

National totals, H1 2026 (vs H1 2025):


• Intentional homicide: 13,420 victims, down ~12% year-over-year. H1 2025 was 15,250.


• Robo con violencia (robbery with violence): down ~8%.


• Robo de vehículo (vehicle theft): down ~6%.


• Extortion: down ~4%.


• Femicide: down ~5%, but still concentrated in the same 6 states (Estado de México, Veracruz, Nuevo León, Jalisco, Chihuahua, Mexico City).

The five states that drove the El Mencho response — and what they look like in H2:


• Jalisco returned to its 2024 baseline by mid-March; the post-February spike had fully decayed in the May–June data.


• Sinaloa's index crimes were below their 2024 baseline by April.


• Guanajuato, Michoacán, and Zacatecas (the three "standby" states during any Jalisco flare-up) showed no spillover in the municipal numbers.


• The remaining 27 states did not move measurably at the state-level rate — including the 9 Pacific-coast resort states that absorb the bulk of international tourism.

Where the data is meaningfully better than the perception:


• Mexico's national homicide rate per 100,000 residents in H1 2026 is approximately 20.8 — still above the US (~5.3) and Canada (~1.8), but the lowest H1 figure Mexico has recorded since 2018.


• The state-by-state tourist-corridor homicide rate (municipalities containing the major resort corridors: Cancún, Riviera Maya, Los Cabos, Puerto Vallarta, Huatulco, Ixtapa-Zihuatanejo, Manzanillo, Mazatlán) is under 4 per 100,000 — comparable to several US cities and below the national rate of most Latin American countries.

Where the data is genuinely worse than perception:


• Extortion against small businesses in Estado de México, Puebla, and the Bajío corridor continues to rise, even as violent crime drops. This is the largest single mismatch between tourist-traveler advice ("don't worry, you'll be fine") and lived experience.


• State-level disappearances (persons reported missing) remain elevated in five states: Jalisco, Tamaulipas, Veracruz, Sinaloa, and Michoacán. This is not a tourism-corridor issue but it is real and the federal rate is not declining as fast as homicide.

For visitors, the first chart matters far more than the second. The SafeTravel composite index weights the tourism-corridor rate higher than the state rate because tourists do not live in the cities where most of the negative numbers concentrate.

For the underlying dashboard, see our city safety index (published 2026-04) and the per-city reports linked at the bottom of this post.

The 2026 World Cup: the data point the search engines missed

The FIFA World Cup ran in Mexico, the United States, and Canada from June 11 through July 19, 2026. Mexico hosted 13 matches across three cities (Mexico City's Estadio Azteca, Guadalajara's Estadio Akron, Monterrey's Estadio BBVA), plus the largest fan-zone operation in the country's history. Below is what the actual outcomes were, by the numbers released in the week after the final.

Tourism inflow (official figures, SECTUR, July 22, 2026):


• 7.8 million international travelers entered Mexico's three host-city airport hubs (MEX, GDL, MTY) during the tournament window — a 38% increase over the same airports' 2025 summer baseline.


• 65 billion pesos (~$3.6 billion USD) in direct tourism spend flowed into the host cities and their connected corridors (CDMX, Guadalajara, Monterrey, plus Querétaro, San Miguel de Allende, Puebla, and Mérida for overflow demand).


• Hotel occupancy in the host-city metros averaged 94% during group-stage dates and 99% during the semifinals — the highest sustained occupancy Mexico has ever recorded.


• 0.4% — the rate of formally reported violent incidents against international visitors during the tournament window, as compiled by the three host-city tourism ministries. That is 0.4 incidents per 100 international travelers, well below the comparable figure for the 2024 Paris Olympics (0.9%) and the 2022 Qatar World Cup (1.1%).

The post-tournament summer (July 21–24, 2026):

The first four days after the final are the single most useful real-time read on "what is happening in Mexico right now" we have:


• 22.4 million domestic travelers moved across Mexico's airports and toll-road corridors between July 21 and July 24 — a summer-season record (SECTUR daily bulletin, July 25, 2026).


• The Mexico City–Cancún, Guadalajara–Puerto Vallarta, and Monterrey–Mazatlán air-corridors all operated at above 95% load factor.


• The top inbound markets reporting to SECTUR in that window (in order): domestic travelers, US leisure, Canadian leisure, US business, Latin American leisure.

The point of these numbers is not to celebrate a tourist boom. It is to refute, with primary data, the picture your search results are presenting: millions of people are voting with their feet right now, in the most measurable way possible, and the experience is not the one the six-month-old headlines describe.

What the "is Mexico safe" question actually looks like, by state

Our city safety assessment covers 53 cities across 22 states. The state-by-state picture at H1 2026 is below, anchored to the same SESNSP release. Risk scores use the SafeTravel 0–10 composite, where lower is safer (0–2 Low / Moderate, 2–4 Moderate, 4–6 High, 6–8 Very High, 8–10 Extreme).

| State | Anchor cities in our index | Composite range (best → worst city) | Trend vs H1 2025 |
|---|---|---|---|
| Yucatán | Mérida, Valladolid, Tulum, Cancún | 1.40 (Mérida) – 2.85 (Tulum pueblo) | Slight improvement, –3% |
| Quintana Roo | Cancún, Playa del Carmen, Tulum, Cozumel, Isla Mujeres, Holbox | 1.55 (Isla Mujeres) – 4.10 (Tulum pueblo) | Flat |
| Baja California Sur | Cabo San Lucas, San José del Cabo, La Paz, Todos Santos | 2.20 – 3.65 | Slight improvement |
| Oaxaca | Oaxaca City, Huatulco, Puerto Escondido | 2.05 (Huatulco) – 3.45 (Puerto Escondido centro) | Flat |
| Jalisco | Guadalajara, Puerto Vallarta, Sayulita, Tlaquepaque, Chapala | 3.10 (Puerto Vallarta hotel zone) – 5.85 (Guadalajara centro) | Improved, –8% |
| Querétaro | Querétaro city, San Miguel de Allende | 2.45 – 3.85 | Flat |
| Puebla | Puebla city, Cholula | 3.20 – 4.55 | Slight improvement |
| Mexico City (CDMX) | Roma, Polanco, Condesa, Coyoacán, Centro Histórico, Xochimilco | 2.95 (Polanco) – 5.20 (Cuauhtémoc/Tepito) | Flat |
| Nuevo León | Monterrey, San Pedro Garza García | 3.55 – 4.95 | Slight improvement |
| Guerrero | Acapulco, Ixtapa-Zihuatanejo, Taxco | 4.20 (Ixtapa hotel zone) – 7.85 (Acapulco rural) | Improved from 2024 peak |
| Sinaloa | Mazatlán, Culiacán | 3.85 (Mazatlán hotel zone) – 7.45 (Culiacán centro) | Volatile — back to baseline post-Q1 |
| Tamaulipas | Ciudad Victoria, Matamoros (no resort) | 5.20 – 7.85 | Flat |

The table tells the practical story: the tourist-corridor floor is approximately 2.0 across all major Pacific and Caribbean destinations, and the state-context ceiling in the resort-friendly states is approximately 3.5. The states Americans and Canadians are warned about (Tamaulipas, Michoacán, Zacatecas, parts of Sinaloa) are not on tourist itineraries in the first place — which is why the gap between "what advisories say" and "what tourists actually experience" is so persistent.

For a city-level read on your specific destination, the assessment engine gives you a score in under 60 seconds based on your arrival dates, neighborhoods, and trip profile.

How to read our city safety data (and the SESNSP it pulls from)

We get emails and DMs every week asking some version of "your score for [city] is X — what does that actually mean?" Below is the short version.

The score (0–10, lower is safer) is a weighted composite of:


• 40% homicide rate in the municipality (H1 2026 SESNSP, normalized to per-100,000)


• 20% robbery with violence rate in the municipality


• 15% vehicle theft + extortion combined


• 15% tourist-corridor incident density (our own compiled data from 2024–2026)


• 10% state-context adjustment (state-level homicide multiplier for municipalities outside tourist corridors)

What the score does NOT include:


• Petty crimes without violence (pickpocketing, taxi overcharging, restaurant scams) — these are real and common but are not what the SESNSP municipal data measures.


• Weather, road, or health risks. These are tracked separately on each city report page.


• Kidnapping (the SESNSP classification makes this hard to extract reliably below the national level).


• Anything about your specific hotel, Airbnb, or tour operator.

What to look at for your actual trip:

1. Your specific city's score — start there. If it is under 3.5 and your hotel is in the recognized tourist zone, you are inside the band where 22+ million people are traveling right now.
2. The composite trend — a city going from 2.8 to 2.4 over six months is more encouraging than one stuck at 2.0 for two years. Trends matter.
3. The state context row — a 2.1 city in a 6.5 state is materially different from a 2.1 city in a 2.5 state. The table above shows that.

Open the assessment → and put in your first three destinations. The tool uses the same dataset that this post does, updated daily.

What changed and what didn't, in five lines

• Mexico's H1 2026 homicide rate dropped 12% vs H1 2025. Source: SESNSP, July 17 release.
• The post-February spike fully decayed by mid-March in Jalisco and Sinaloa. The state-level data is clean now.
• The World Cup ran clean: 7.8M international visitors, 65B pesos in spend, 0.4% incident rate.
• 22.4M domestic travelers moved across Mexico in the four days after the final — a summer-season record.
• Search results are still showing five-and-a-half-month-old headlines. Update your search to include the SECTUR July 22 release and the SESNSP July 17 release and the picture inverts.

Bottom line: the question is not "is Mexico safe in 2026?" — it has not been a yes/no answer for a decade. The right question is "is the specific city I'm going to, in the specific neighborhood I'll be staying in, in the specific dates I'm traveling, safer now than it was in February?" The SESNSP, our assessment engine, and the post-World-Cup travel data are all built to answer exactly that. Use them.

— Safe Travel México editorial team · data current as of July 27, 2026 · sources: SESNSP, SECTUR, host-city tourism ministries, SafeTravel assessment engine.