The Wander List2026

Homicide

Intentional homicides per 100,000 people per year, as reported to the UN — lower is better.

• What counts — deliberate killings, and nothing else: deaths in war, on the roads or by suicide are not in this number.
• Who counts it — national police and statistics offices, reported to the UN’s crime office, with the WHO’s model filling the gaps.

0 – 1	Lethal violence is uncommon and murders are rare enough to be news.
1 – 2.9	Safer than most of the world. Killings are uncommon, strangers are almost never involved.
2.9 – 5.8	Under the global average. Violence clusters in specific places and circles.
5.8 – 10	Higher than the global average. Killings are concentrated in particular cities and districts rather than spread evenly.
10 – 20	Past the WHO's epidemic line. Violence at this level reaches into everyday life rather than staying in a few areas.
over 20	Double the WHO's epidemic line. Killing is a daily event.

Sources: WHO estimates of rates of homicides · UN victims of intentional homicide · UN SDG indicator 16.1.1

Estimates: 129 of our 180 countries report to the UN directly (UNODC, SDG indicator 16.1.1); 49 — mostly in Africa and Asia — carry the WHO’s modelled estimate for 2021 instead, and Taiwan (national police statistics) and Monaco (no recorded homicides) sit outside both systems. Where a country has published a newer count than the UN’s copy of it, we use the newer count — it’s the figure the UN will print next; where a country’s numbers simply disagree with the UN’s, the UN wins. Each page’s source line says which figure it carries.

Road Fatalities

WHO estimated road-traffic deaths per 100,000 people per year — lower is better.

• What counts — everyone who dies on the roads: drivers, riders, passengers and people on foot.
• Who counts it — the WHO, modelling each country’s true toll from registered deaths and national reports, because official counts miss many deaths in much of the world.

under 4.5	Under the EU average. A road death is rare enough to make the news.
4.5 – 7.5	Under half the global average. Crashes are uncommon and mostly happen on fast roads outside towns.
7.5 – 15	Under the global average. Cars share the road with scooters, bikes and people on foot, and crossings are not always respected.
15 – 22.5	Above the global average. Speeding and overloaded vehicles are common, and enforcement is thin.
22.5 – 30	Well past the global average. The roads are a leading cause of death.
over 30	Double the global average. Every journey carries a real chance of a crash.

Source: WHO Global Health Observatory, estimated road-traffic death rate 2021 (SDG indicator 3.6.1), as published in the Global Status Report on Road Safety 2023

Estimates: the WHO covers 175 of our 180 countries. The five it misses use official national or regional figures, marked as estimates on their pages: Taiwan (transport ministry), Kosovo (Transport Community), Liechtenstein (Eurostat), Vatican City (no recorded road deaths) and North Korea (the WHO's last Global Health Estimate, 2019).

Conflict

Deaths in armed conflict inside the country, from the Uppsala Conflict Data Program’s event records, plus any UN sanctions or court order over a war — lower is better.

• Deaths at home — people killed in fighting inside the country where the state is one side. That means its army or police fighting an armed group, or fighting another state.
• Civilians killed on purpose — civilians deliberately attacked by a government, or by an armed group that is at war with the state in that country.
• UN sanctions or a court order — live UN Security Council sanctions over the country's war or its nuclear weapons, a UN General Assembly finding of aggression that names the country, whether as the aggressor or for its involvement, or an International Court of Justice order against it over the conduct of a war. The last two count for five years.

Left out on purpose: fighting between gangs, militias or communities with no state involved, and wars a country fights abroad. Where a government fights a gang, the deaths count here and in Homicide.

1 – 1.3	Twenty-five years without a single conflict death. Nobody has been targeted, and neither the UN nor the World Court has anything against the country.
1.3 – 1.8	Five years without a conflict death. Whatever happened before that has stopped.
1.8 – 2.3	6 months without a conflict-related death at home. Whatever happened has stopped recently.
2.3 – 3	Conflict-related deaths at home in the last 6 months.
3 – 4	Either of two things puts a country here. More than 1000 people have died this year in fighting inside the country. More than 100 civilians have been killed on purpose.
over 4	More than 2,500 people have died this year in fighting inside the country, or more than 250 civilians have been killed on purpose by the state, or the UN or the World Court has acted against the country over the conduct of a war.

Inside a band, countries are ordered by deaths at home plus twice the civilians killed on purpose, or, where there were none, by how long ago the last death was.

Sources: the deaths come from the Uppsala Conflict Data Program. We use its Georeferenced Event Dataset 26.1, which runs to December 2025, and its monthly Candidate Events Dataset, which runs to July 2026. Both are free downloads. The codebook defines the three kinds of violence and the best estimate that we count. The UN row rests on the Security Council's list of live sanctions regimes, on General Assembly resolution ES-11/1 of 2 March 2022, and on the International Court of Justice's orders on provisional measures. Each affected country's page links the resolution or case.

Estimates: nothing on this metric is estimated by us. Every country is counted from the same file. The window is August 2025 to July 2026. The monthly files run about a month behind, and they are provisional until the next annual release. Gaza and the West Bank are filed under Israel by UCDP, and we count them as Palestine.

Freedom

Six freedoms, each rated 0 to 4 by Freedom House in its yearly Freedom in the World report — higher is better. The weakest of the six sets the band, and the total places the country inside it.

• Protest — "Is there freedom of assembly?"
• Speech — "Are individuals free to express their personal views on political or other sensitive topics without fear of surveillance or retribution?"
• Press — "Are there free and independent media?"
• Religion — "Are individuals free to practice and express their religious faith or nonbelief in public and private?"
• Movement — "Do individuals enjoy freedom of movement, including the ability to change their place of residence, employment, or education?"
• Due process — "Does due process prevail in civil and criminal matters?"

Left out on purpose: Freedom House's question on physical force, war and insurgency, because war is the Conflict column's job, and its questions on political parties, non-governmental organisations and trade unions. Nothing is weighted — every rating is the one on the country's Freedom House page, linked from that country's Freedom section.

9 – 10	All six freedoms rated 4. Nothing you do or say is a risk.
7.5 – 9	Lowest freedom rated 3. One freedom has a limit somewhere, and you will not notice it unless you go looking.
5 – 7.5	Lowest freedom rated 1 or 2. Speaking up costs, but no freedom is absent, the courts work and you can leave.
4 – 5	Protest, speech, press or religion rated 0. Using that freedom is out, but the courts and the borders still work.
2 – 4	Due process rated 0. If anything goes wrong, there is no fair trial and the courts will not protect you.
under 2	Movement rated 0. You may not be free to leave.

Source: Freedom in the World 2026 (Freedom House, published 19 March 2026, rating calendar year 2025) — the question scores on each country's report page; the questions and the 0–4 scale are set out in the methodology. Freedom House rates the West Bank and Gaza separately from Israel: Israel's page carries Israel's own ratings and Palestine's carries the West Bank's, and both pages say so.

Estimates: Freedom House does not rate Vatican City. Its six ratings are ours, assigned against the same questions from the Fundamental Law of Vatican City State (2023), and its page says so. Every other country is Freedom House's rating, untouched.

LGBT Rights

Equaldex Equality Index, 0–100, shown out of 10 — higher is better. The law and the public mood both count, because both decide what a visitor can safely do. It's the average of two halves:

• The law — what's legal and what's protected: same-sex relationships and marriage, adoption, anti-discrimination protections, legal gender change, bans on conversion therapy, and the smaller items (blood donation, military service, equal age of consent).
• Public opinion — how accepting people say they are, from published surveys (Pew, Gallup, ILGA-RIWI and national polls). Good laws with a hostile public land mid-table; so does a tolerant public with no legal protections.
• Some colours come from the law or the state, not the score — and then the cell shows one of the words below instead of a number, with every criminal case cross-checked against the Human Dignity Trust's map.

7 – 10	Equal in law or close to it, with public acceptance to match. The remaining arguments are about details.
5 – 7	Legal and liveable, but something substantial is missing. That may be marriage, protections or public acceptance, and each country's page says which.
3 – 5	Legal but second-class. Couples have no standing in law, protections are thin and most people stay discreet.
under 3	Legal only because no law forbids it, with police, officials and the public doing the work a ban would. A country whose anti-LGBT law censors LGBT life short of a ban lands here too, and its cell says Censorship.
illegal	Homosexuality is a crime punished with prison, so the cell says Illegal instead of a score. A state that raids and arrests without such a law lands here too, and its cell says Crackdown.
life or death	Homosexuality is punished with life in prison, death or death by stoning. The cell says which.

Sources: Equaldex Equality Index · Human Dignity Trust map of criminalisation · ILGA World legal database · Human Rights Watch LGBT rights reporting · Amnesty International LGBTI rights reporting

Estimates: Equaldex covers all 180 of our countries — nothing on this metric is estimated by us.

Women's Safety

Built from Georgetown’s Women, Peace and Security Index and OECD data: four things, each scored out of one and averaged, shown out of 10 — higher is better.

• Feeling safe at night — the share of women who say they feel safe walking alone after dark where they live.
• Political violence against women — recorded attacks on women by militias, mobs, security forces or at elections.
• A fair hearing in court — whether a woman can take a case to court and be heard without putting herself at risk.
• Violence at home — the domestic violence rate and the share of women who say a husband is justified in beating his wife, averaged into one because they measure the same thing.

Feeling safe at night also sets a floor. A country cannot score top unless two in three women feel safe, green unless more than half do, or yellow unless one in three do.

8.6 – 10	2 in 3 women or more feel safe walking alone at night. Political violence against women is next to none, and the courts work for women.
8 – 8.6	More than half of women feel safe out at night. Political violence against women barely registers, and the courts mostly work for women.
7 – 8	1 in 3 women or more feel safe out at night, but taking a case to court is a gamble. In some of these countries a woman cannot do it safely at all.
6 – 7	Fewer than 1 in 3 women feel safe out alone at night, or the courts and political violence pull the score down.
3 – 6	Most women are afraid to go out alone. The courts offer little or no protection.
under 3	Political violence against women is common.

Sources: Georgetown Institute for Women, Peace and Security, WPS Index 2025/26 (data file and methodology appendix), built on the Gallup World Poll, ACLED, V-Dem and the UN SDG database · OECD Gender, Institutions and Development Database 2023 (attitudes to domestic violence, women aged 15–49)

Estimates: Vatican City isn't covered by Georgetown, so it carries Georgetown's placeholder figure for Europe's microstates. Where a country has no national survey, an input is filled with a regional average — Georgetown does this for some domestic-violence figures (the Gulf states all share one), and we do the same for the 18 countries the OECD attitudes survey misses. Every affected country page says so in its source line.