The turn is a sequence of decisions
In a turn-based strategy game you usually control several units or actions each turn, and the game waits while you think. That removes the pressure of speed, but it makes planning the core skill. Many such games also include random elements: percentage-based hit chances, damage ranges, critical hits or random events. Making the most of your moves means working out what order to act in, which risks to take, and how to leave yourself a sound position if luck goes against you.
The figures in this article are examples. Actual rules differ between games.
Expected value is a starting point
When an action has a chance to fail, expected value gives a quick comparison. Multiply the result by its probability.
Say an attack has a 70% chance to hit for 30 damage. The expected damage is 0.7 × 30 = 21. A different attack hits 90% of the time for 20 damage, for an expected 18. On average, the first attack does more, but it fails three times in ten, while the second fails one time in ten. Which is better depends on the situation. If you need to finish a target at exactly 30 health, only the first can do it in one go. If you just want steady chip damage, the second may be the more reliable choice.
Averages are handy, and they are the same tool used for board game probability, but they hide the spread of outcomes. Look at what each result means for your position, not only the average.
Move order: use information as you go
One of the most useful ideas is that you often see the result of one action before choosing the next. Order is how you take advantage of that.
Consider a target with 40 health, and three units that can attack it:
- Unit A: 60% chance to hit for 40 damage (a kill on its own)
- Unit B: 90% chance to hit for 20 damage
- Unit C: 90% chance to hit for 20 damage
B and C together need both hits to kill, which happens with probability 0.9 × 0.9 = 0.81.
If A goes first, it kills 60% of the time. When it misses (40% of the time), B and C still have an 81% chance of finishing the job. The overall kill chance is 0.6 + 0.4 × 0.81 = 0.924, about 92%.
If B and C go first, they kill 81% of the time and A is still free. When they fail (19%), A attacks with a 60% chance. The overall chance is 0.81 + 0.19 × 0.6 = 0.924. It is the same total, as it should be if every unit is used when needed.
What differs is which units are free afterwards. If A goes first and succeeds, B and C are free about 60% of the time. If B and C go first and succeed, A is free about 81% of the time. So the better order depends on what the spare units could do instead. If A is a valuable unit you would rather use elsewhere, send B and C first. If A is in an exposed spot and you would like to retreat it, sending it first and moving it is another idea. The point is that the same kill probability can leave you with different resources, and the order is where you make that choice.
Do the uncertain thing first
A general rule of thumb follows from the example: when you can, resolve the uncertain action first, then adapt. If a risky attack misses, you still have the rest of the turn to respond, perhaps by retreating or by switching targets. If you did the safe actions first and then the risky one, you would have committed your other units without knowing the outcome.
This works only if the actions are independent. Sometimes an earlier move changes the odds of a later one, such as repositioning a unit to gain a hit bonus or to open a line of attack. In that case the order is driven by those dependencies, not by uncertainty.
Think in worst cases as well as averages
Some positions are fragile: a single bad roll can lose a key unit or the game. In those cases it is worth checking what happens if the unlucky outcome occurs.
A sensible check is to ask, “If this fails, can I still recover?” If so, the gamble might be acceptable as a calculated risk. If not, a safer line that gives up some expected value may be the better practical choice. This is risk management applied to a single turn. When you are well ahead, reducing variance is often sensible, because you only need to avoid disasters. When you are behind, a higher-variance play may be reasonable, because the safe line is unlikely to win anyway. Any such gamble remains a gamble, and the odds still apply.
Probability of at least one success
Some plans let you try again, and then the useful number is the chance that at least one attempt succeeds. If each try succeeds with probability p, the chance of at least one success in n tries is 1 − (1 − p)ⁿ.
With a 60% chance per try and two tries, the chance of at least one success is 1 − 0.4 × 0.4 = 0.84. With three tries it is 1 − 0.064 = 0.936. Extra attempts help, but with diminishing returns, so it is worth asking whether the second or third attempt is the best use of that unit.
Positioning and exposure
Probability does not only apply to your attacks. Where you stand determines the number of enemies that can target you, and each of their attacks has its own chance to hit. A unit exposed to three attackers at 50% each is far more likely to be hit at least once than one exposed to a single attacker. Using cover, terrain or blocking lines reduces the number of rolls against you, which is often more reliable than hoping for low rolls.
Planning ahead
Beyond the current turn, think about what the opponent can do in reply. Check which of your units could be attacked, what their best response might be, and whether your position still works if two or three random results go against you. Strong turn-based players tend to plan a turn or two ahead and to keep a reserve, whether health, spare actions or an escape route, so that one bad outcome does not decide the game.
A small habit helps: before ending a turn, scan once for the most damaging thing the opponent could do, and decide whether you are comfortable with it.