How to recognise cities from street plans

Find an anchor before following the streets

Do not start with a city name. Describe what you can actually see: water on one side, a dense centre, a large park, bridges or an undeveloped strip. An anchor gives the other details a position. A river through the centre and a sea along the edge can look similar in a small crop, so check whether the opposite bank is visible.

Next, consider scale. If individual blocks are clear, compare their shapes and proportions. If the whole city is visible, major roads and the arrangement of districts become more useful. A ring road missing from a small crop is not evidence against a city: it may simply lie outside the image. [2]

Recognise three patterns that can overlap

Grids: look for the rhythm of the blocks

A grid produces repeating rectangles, but “straight streets” is a broad clue. Are the blocks elongated? Do their sizes repeat? Are diagonal avenues cutting across them? Do squares create regular breaks? Rotating a grid does not change its structure; its direction on your screen is not distinctive by itself.

Repeated squares within a block pattern can tell you more than rectangular streets alone. The National Park Service’s study of Savannah shows streets, wards and open squares forming a repeatable system. The useful observation is the recurring unit, rather than a single attractive square. [1]

Rings and spokes: check for a shared centre

Radial roads spread from a central area, while ring roads curve around it. Check whether several directions genuinely converge. One roundabout surrounded by straight streets does not make a radial-and-ring plan. You need to examine the structure of the visible area.

Irregular networks: compare neighbouring districts

Winding streets and uneven blocks may sit beside a regular grid. That boundary is more useful than simply calling an area an “old town”: it marks a change in pattern. A bend alone does not tell you a street’s age. The road may follow a slope, water or another obstacle.

Build a combination, rather than chasing a symbol

An ordinary grid fits many cities. Elongated blocks, two diagonal routes and water along one edge form a narrower hypothesis. Clues are more useful together when they describe different things: the street pattern, shoreline and position of the centre.

Avoid counting one feature twice. “A long avenue” and “a wide straight road” may describe the same line. Those observations do not provide two independent confirmations. Add a different question: where does that avenue sit in relation to the water or the densest buildings?

A comparison: which hypothesis fits the map?

Imagine a fictional map of a whole city with elongated rectangular blocks, two diagonal routes and water along one edge. Candidate A is described as a city around a circular centre with several rings. Candidate B has a grid crossed by diagonal roads beside the coast. Candidate C has an irregular inland street network.

B fits best. A would need visible rings and a shared centre, which are absent. C conflicts with the regularity and the water, assuming its description covers the entire area shown. We have not named a city, but we have checked which account explains the pattern without extra assumptions.

An actual game may offer less certainty. The image can be cropped, details simplified and several choices structurally similar. Keep two hypotheses when necessary, then look for a difference between them. That is more useful than finding reasons to defend your first impression.

Four checks before you answer

A map becomes easier to read when you can explain your choice briefly: “This candidate fits both the blocks and the shore; the other only fits the straight streets.” That explanation is easier to test than a feeling that you have seen the place before.

1. Establish the area and scale: a centre, one district or the whole city.
2. Describe the streets without naming a place: grid, spokes, rings or a mixture.
3. Find a second anchor: water, a park, bridges or a district boundary.
4. Compare the remaining candidates using a feature that actually distinguishes them.