the boring part that wins
The RTS economy guide
Nobody watches a replay for the harvesting. It decides the match anyway. Here is the arithmetic, and what to do with it.
The economy is the least glamorous system in a real-time strategy game and it is the one that decides who wins. This is uncomfortable, because it means the skill that matters most is essentially bookkeeping.
The compounding problem
Income in an RTS is not linear, it is exponential, because income buys income. Work an example through in TANKS numbers:
| Minute | Player A (buys a 2nd harvester at 0:45) | Player B (buys a tank instead) |
|---|---|---|
| 1 | ~1,400/min income, no army | ~1,400/min income, one tank |
| 3 | ~2,800/min, 3rd harvester paid for | ~1,400/min, three tanks |
| 6 | ~4,200/min, and a factory running non-stop | ~1,400/min, seven tanks |
| 9 | Producing tanks faster than B can kill them | Cannot replace losses |
Player B is ahead for the first four minutes and then loses, permanently and without a fight ever having gone badly for them. If B does not use that early window to attack, the window was worthless.
How many harvesters
The answer in a C&C-style game with vehicle harvesters is smaller than people expect, because the refinery can only unload one at a time.
- Two per refinery is the working number on a normal map.
- Three if the nearest field is far, or if you have taken a second field in a different direction.
- More than three per refinery is waste — they will queue at the door. Build a second refinery instead, closer to the ore.
The test that removes all guesswork: watch the refinery door. A queue means you have too many harvesters, or your refinery is in the wrong place. An empty refinery means you have too few. Adjust and stop thinking about it.
Where the refinery goes
Toward the ore, not in the middle of your base. This looks reckless and it is correct. Every second of driving is money that does not exist, and a round trip you shorten by four seconds is roughly a 20% raise for every harvester you will ever build.
Expanding
Ore fields are finite. This is the mechanic that stops a base-building game from becoming a stalemate, and it means the map has a clock on it. Whoever runs dry first has to move first, into ground the other player has had time to prepare.
So: expand while you are comfortable, not when you are forced. The signs it is time are
- your home field is visibly thinning,
- your harvesters are driving further each trip,
- you have more than about two thousand credits banked with nothing urgent to buy.
An expansion needs three things to survive: a refinery, a power plant to run it, and something that shoots. A gun turret and two units on guard is usually enough, because a raid that costs the raider anything is a raid that stops coming.
Protecting income
Your harvesters are the softest valuable thing on the map, and every good player knows it. Three habits:
- Two guards, permanently, on guard stance. Not aggro — aggro chases, and chasing is exactly what a scout is sent to provoke.
- Pull, do not fight. When a raid lands on a harvester, drive it home. You are not trying to win the engagement, you are trying to keep 700 credits.
- Watch your income number, not your harvesters. If credits stop rising at the rate you expect, something is wrong that you have not noticed yet. This is the single most useful piece of information on the screen and almost nobody looks at it.
Power is part of the economy
In games with a power grid — Command & Conquer, Red Alert, TANKS — power is not a separate system. It is a tax on your production speed.
In TANKS the rule is exact: if demand exceeds supply, everything is produced at (supply ÷ demand) speed, to a floor of 25%, and the radar goes dark. So a base at 70% power is not "slightly inconvenienced" — it is a base whose entire income is being converted into units 30% less efficiently than the other player’s.
It also means the enemy power plants are an economic target, not a military one. Destroy enough of them and you have not killed their army — you have reduced the rate at which they can build a new one, which is better. Two of the twelve campaign missions are built entirely on this idea.
The one-line summary
Keep income rising, keep production busy, keep your bank near zero, and look at the refinery door occasionally. That is macro. It is not exciting and it will beat people with better reflexes than you for years.
Reading about it is slower than playing it.
TANKS runs in this browser tab. No download, no account, no launcher.
Questions people actually ask
How many harvesters should I have?
Two per refinery is the working answer in a Command & Conquer-style game, and a third if the ore field is far away. The practical test: if a harvester ever queues to unload you have enough, and if the refinery sits idle you do not.
When should I expand to a second ore field?
When your first field is visibly thinning, or when you have spare money and nothing urgent to spend it on. Expanding early is strong and risky; expanding after your field runs dry is too late, because you will be building the expansion while broke.
Why does my income keep dropping?
Usually one of three things: a harvester died and you did not notice, the nearest ore field ran out and your harvesters are now driving much further, or you are in a power deficit and everything is running slowly.
Read next
RTS build orders explained
What a build order is, why the first four minutes decide the match, and an opening you can copy.
A beginner’s guide to real-time strategy
Starting from nothing: economy first, what to build, when to attack, and the five mistakes everybody makes.
Games like Dune II
The 1992 game that invented the genre, and everything that inherited its harvester-and-sidebar loop.