start here

A beginner’s guide to real-time strategy

Real-time strategy has a reputation for being hard. It is not — it is just unforgiving about one specific thing, and almost everybody gets that thing wrong for their first ten games.

Updated 26 August 2026 · 5 min read

Here is the whole genre in one sentence, and it is worth reading twice:

You are not trying to build an army. You are trying to build the thing that builds the army, faster than the other person builds theirs.

Everything below is a consequence of that sentence. If you take nothing else, take that.

The four things every RTS has

The specifics change between games; the four parts do not.

1. An income. Something collects a resource and brings it home. In Command & Conquer, Dune II and TANKS it is a vehicle that drives out to a field and drives back. In StarCraft and Age of Empires it is worker units near your base.
2. Production buildings. These convert income into army. You cannot build faster than you have buildings, which is why one barracks eventually becomes a bottleneck no matter how rich you are.
3. A tech tree. Each building unlocks the next tier of things. Rushing up it costs you army now for better army later.
4. An army. The thing you actually fight with, and the only one of the four that does not pay for itself.

The mistake everybody makes

You start the game, you feel exposed, and you build soldiers. It feels responsible. It is the single most reliable way to lose.

Work through the arithmetic once. In TANKS a harvester costs 700 credits and brings home roughly 1,400 credits a minute. It pays for itself in thirty seconds and then prints money for the rest of the match. A light tank costs 500 credits and earns nothing, ever. Two minutes in, the player who bought the harvester has around 2,800 more credits than the player who bought the tank — which is five more tanks.

What to build first, in almost any game

1. Whatever unlocks income. In TANKS that is a power plant and then a refinery — the plant first, because everything else draws power and a building you cannot run is a building that runs at quarter speed.
2. A second source of income. A second harvester, or more workers. This is the decision that wins most games, and it is boring, which is why people skip it.
3. Cheap defensive units, but only a few. Enough that a single scout cannot walk into your base and kill your harvester. In TANKS, three riflemen on guard at the refinery. They cost 300 credits and they buy you the two minutes you need.
4. Your main production building. The tank factory, the war factory, the stables — whatever turns money into the units you will actually fight with.
5. Now army, and keep building it. From this point your production buildings should never be idle. An idle factory is money you already earned and are choosing not to use.

When to attack

New players attack whenever they have units, which trickles them into the enemy base one at a time to be killed one at a time. This is called feeding, and it is the second-most-common way to lose.

The rule is simple: attack with a group, or do not attack. Gather everything at a rally point out of sight, count it, and then move as one. Twelve tanks arriving together beat twelve tanks arriving over ninety seconds by an enormous margin — the group fights twelve-on-four three times, instead of one-on-four twelve times.

Good moments to commit:

• You have just finished a tech tier the opponent has not. There is a window before they catch up, and it closes.
• You scouted their base and it is undefended.
• Your income is worse than theirs and getting worse. If you are losing the economic game, waiting makes it worse. Attack while you can still afford to.
• You are about to hit a resource cap or your money is piling up unspent.

The five habits that fix most losses

1. Never let your money sit still. Credits in the bank do nothing. If you are above a couple of thousand and not saving for something specific, you have under-built. Add a production building.
2. Keep production running. Every time you look at your base, check whether the factory is making something. It usually is not.
3. Scout. One cheap unit sent out at the three-minute mark tells you whether they went economy or army, which tells you whether to attack or defend. A rifleman costs 100 credits. Losing to a surprise costs the game.
4. Use control groups. Select your army, press Ctrl and a number. Now you can call it back with one key from anywhere on the map. Almost all the apparent speed of good players is this, not reflexes.
5. Put defenders on the right stance. Units set to chase will follow one enemy scout off the map and leave your harvester alone with the raid that was actually coming. Anything defending anything goes on guard.

Micro and macro, briefly

You will see these words everywhere. Macro is everything to do with your base — income, production, build order, spending. Micro is controlling individual units well in a fight — pulling a damaged tank out, focusing fire, keeping artillery behind the line.

Beginners assume the genre is about micro because that is the part that looks impressive. It is not. Macro decides matches between players below a very high level, and macro is mostly a checklist you can learn in an afternoon. Learn macro first. The full list of terms is in the RTS glossary.

Now go and lose four games

This is not a pep talk, it is the method. Play four skirmishes with the explicit goal of practising the build order rather than winning, and watch what actually kills you each time. The failure will be the same one three times out of four, and it will be economic.

Next: the build order guide gives you an actual opening to follow, and the economy guide covers harvester counts and how to protect them.

Questions people actually ask

Read next