What makes a good tactics map

Two failures bracket the whole subject. A map with exactly one solution is a puzzle wearing a tactics game's clothes: you are not making decisions, you are guessing what the designer did. A map with no shape at all, an open field with some crates on it, is a shooting gallery where the best play is whatever your strongest unit does.

Everything worth saying about map design lives between those two, and it comes down to one question: does the ground offer the player more than one reasonable plan, and can they tell what the plans are before committing to one.

None of these are decorative. Each one changes what a squad can do on it.

The cheapest way to double the value of a map is to change what you are asked to do on it. Clear every enemy and the map is a sequence of fights you can take in any order you like, slowly. Reach a van at the far end and the same ground becomes a route with threats beside it. Hold a position for five minutes and it becomes a question about angles and ammunition. Get somebody out alive and every open stretch you crossed on the way in is a problem on the way back.

This is why the best mission designers spend their time on objectives rather than on terrain. Terrain is expensive and objectives are nearly free.

The two traditions in this genre answer the question differently and both are right.

X-COM: UFO Defense (Mythos Games and MicroProse, 1994) assembles a fresh map for every mission out of prefabricated blocks. You never learn a map, so you learn the game instead, and a farmhouse at night is frightening because you genuinely do not know what is in it. The cost is intention: no sequence of rooms was ever composed by anyone, so the maps are interesting rather than crafted.

Commandos: Behind Enemy Lines (Pyro Studios, 1998) is the opposite. Every guard stands exactly where a designer put him, every patrol is timed, and the level is meant to be solved rather than survived. The cost is that once you have solved it, it is solved forever, and the game lives or dies on how good the composition was.

Modern games mostly do both: hand-made set pieces in a campaign, generated or shuffled ground for everything else. The honest version of that split is the one that tells the player which kind of map they are on, because the two reward completely different behaviour. You scout a generated map. You plan a built one.

The golden path. There is a correct route and the designer knows it. Everything else is punished, usually invisibly, and the player's reward for creativity is a loaded save. This is the most common fault in hand-built levels and it is a failure of generosity rather than of skill.

Unreadable ground. The map is interesting and the camera will not show it to you. An isometric view hides the inside of buildings and the far side of walls, which is why this genre invented so many cutaway tricks. A threat you could not have seen is not difficulty, it is a mistake with the player's time.

Ground that does not matter. Maps where the rules are so strong that terrain is noise. If the best play is the same on a street, in a warehouse and in a forest, the systems have eaten the level design.

Too much of it. A map large enough to take two minutes to cross is a map where most of your attention is spent on walking. The most respected levels in this genre are small and dense, not big.

RAD Tactics is a squad tactics game for Windows, real time with pause, set after a collapse, with zombies and with factions fighting over what is left. It is in development, and we are not going to describe anything in it that is not finished, including any particular map.

The standard we are working to is the one at the top: more than one reasonable plan, and enough visible to choose between them. A fight the player lost because of a decision is worth ten they lost because of something off screen.