A head-to-head page puts two cities side by side and answers, in its first sentence, which one is ahead. This page explains where every number on it comes from, and how the set of published comparisons is chosen.
What is on a head-to-head
Eleven rows, and every one of them is a fact rather than an opinion:
- Population, for the city proper, from the GeoNames open geographic database. Not the metropolitan area — the same basis for both cities, which is what makes the comparison meaningful.
- Size within its own country, by population, among the cities in our catalogue. Only stated for cities at or above the catalogue’s inclusion floor; below that the catalogue is not a complete list of the country’s cities and the claim would not be derivable.
- Country, region and capital status, also from GeoNames.
- Local time, resolved from each city’s IANA time zone at the moment the page was generated. It is a snapshot, not a ticking clock, and the page says so.
- CityWar rank, score, backers and score added today, read from the live board. A city nobody has backed shows “unclaimed” rather than a position.
The distance between the two is a great-circle calculation from the two sets of coordinates, centre to centre. It is computed, not looked up, which is why it is stated to the kilometre and not rounded to something vaguer.
One matchup, one URL
“Mumbai vs Delhi” and “Delhi vs Mumbai” are the same question. If both had their own page they would be two pages competing to answer it, splitting whatever authority either one earned.
So a pair has exactly one address. The city with the larger population leads the URL, ties broken alphabetically, and the reverse order redirects permanently to it. Population order rather than alphabetical because it is also the order a person writes: nobody types “Delhi vs Mumbai” first.
Which comparisons are published
5,908 cities make 17,449,278 possible pairs. We publish 1,636 of them, chosen by five rules:
- Cities adjacent in size among the 120 largest cities on earth — the closely matched ones.
- Every pair among the 12 largest, which is where the “biggest city in the world” questions land.
- Cities from the same country, top 5 by population. This is where the overwhelming majority of real “X vs Y” searching happens.
- A country’s capital against its largest city, where those are different places. A genuine question about roughly ninety countries and a trivial one about the rest, so the rest are not published.
- Large cities against the city physically nearest them, which catches the metro-area and cross-border pairs that population order misses.
Plus, at any moment, any two cities sitting next to each other on the board. That is the one comparison the product itself creates, and it is the only part of the set that changes.
Why not all of them
Because 17,449,278 near-identical pages would be a doorway farm, and a site that publishes one does not rank for “mumbai vs delhi” — it gets its whole domain discounted. The test a comparison has to pass is whether somebody would ever ask it. Almost none of the seventeen million would.
Any pair can still be compared. Type or link any two city slugs and the page renders with all the same data — those pages simply carry a directive asking search engines not to index them, while still following their links through to the two city pages. Useful to a person who follows a link, invisible to a search index, which is exactly the split that situation calls for.
Finding a comparison
The comparison hub lists the featured matchups by group. Every city page also carries the comparisons that city takes part in, which is usually the faster route: start from the city you care about and the relevant rivals are already there.