Judging a position you can only partly see
In a real-time strategy game you almost never see the whole board. Fog of war hides most of the map, and everything keeps changing while you decide. Reading the game state means turning scattered, incomplete observations into a working estimate of who is ahead and what is likely to happen next. Good players do this constantly and without much fuss. The skill is less about secret knowledge and more about knowing which few numbers matter.
The examples below use made-up figures to illustrate ideas. Real games differ in rules and balance.
Income matters more than the bank
The resources a player currently holds are the least informative number. A large unspent bank may mean a player is saving for something, or that they cannot spend efficiently. Income, the rate at which resources arrive, says much more about the future.
Imagine one player earns 900 resources per minute and another earns 700. The gap is 200 per minute. After three minutes, the first player has collected 600 more in total, which is enough to buy a noticeable chunk of extra army or tech. Small income edges compound, because extra resources become extra workers, buildings or units, which can raise income further. This is why many strategy players focus on economic growth early and why a stalled economy is often a bigger warning sign than a smaller army.
The reverse also holds. A player who is behind on income but ahead in army has a window of opportunity, and that window typically closes as the economy gap widens. Knowing which side of that trade you are on helps decide whether to attack or to expand.
Map control as information and income
Map control is the share of the map where you can move, see and gather resources without being contested. It matters for two reasons.
First, it supports income. Extra resource sites need to be defended, and a player pinned in one corner often cannot take them. Second, it creates information. Vision of the central routes shows when an enemy army is moving, which gives time to respond. A player who sees more of the map makes decisions with fewer unknowns, so their estimates are tighter.
Control is not only about territory. A player with a few well-placed scouts, towers or fast units can hold vision cheaply, while another controlling more ground but seeing less of it may still be surprised.
Counting units and estimating what you cannot see
You rarely know the exact size of an opponent’s army, but you can build an estimate from what you do know. A basic approach looks like this:
- Note what you saw last and when you saw it.
- Estimate how much income the opponent had since then.
- Subtract rough costs for known buildings, upgrades and expansions.
- Assume most of what remains went into units.
This gives a range, not a figure. The useful habit is to hold a range in mind and update it whenever new information arrives, rather than treating the last scouting report as permanent truth. Information goes stale quickly in a real-time game.
Counts are not the whole story either. Unit types, upgrades and positioning change the value of an army. A smaller army with a good matchup against the opponent’s composition can beat a larger one.
Army size and the square-law idea
Many strategy players use a rule of thumb called Lanchester’s square law. In a simplified model where all units fight at once and can shoot at each other, the fighting strength of an army grows with the square of its numbers, not in direct proportion.
Take 10 identical units against 6. By the model, the strengths are 10² = 100 and 6² = 36. The larger side wins and ends with roughly the square root of the difference: √(100 − 36) = √64 = 8 units left. Losing only two units from a 10-unit force while destroying all 6 sounds generous, and it is, because the model assumes ideal conditions.
Real games break that assumption constantly through limited range, narrow chokepoints, area damage, unit abilities and uneven reinforcements. The model is best used as a reminder that an extra few units can matter more than they appear to, and that fighting in a spot where only part of the larger force can engage reduces their advantage. It is not a prediction tool.
Supply, timing and tempo
Other measurable signals tell you about the state of the game:
- Worker count. A drop in workers often means losses or a switch to an army focus.
- Production buildings. More production buildings indicate the capacity to spend a high income on units.
- Tech and upgrades. These suggest what unit types are coming and when.
- Time. The same army size means something different at three minutes and at twelve.
None of these is conclusive alone. Together they offer a rough picture of tempo, meaning who has the initiative and who is reacting.
Confidence, and the risk of reading too much
Reading the game state is closely related to making informed decisions with imperfect information. Professional matches are often discussed in terms of esports probability, but the same habit applies at any level. Two mistakes are common. One is overconfidence: committing to a plan based on a single scout that may be out of date. The other is paralysis, where a player waits for certainty that never comes and loses tempo.
A sensible middle path is to estimate, act on the most likely read, and keep a cheap way to check. If your estimate says the opponent has a small army, a quick scout confirms or corrects it before you commit. If you are wrong, the cost is small. Treat each read as a probability, not a fact, and the occasional bad guess hurts much less.
A short checklist
During a game, it can help to ask the same few questions every minute or so:
- Whose income is higher, and by roughly how much?
- Who controls more of the map and sees more of it?
- What is the last time I saw the opponent’s army, and what could have changed since?
- Would a fight now happen on terms that favour me?
Answering these in a few seconds is a trainable habit, and it tends to improve with repetition, though no habit removes the uncertainty of fog of war.