Chance and skill in the same game
Action games blend two kinds of challenge. One is skill: dodging an attack by moving at the right moment, landing a combo, reading an enemy’s animation. The other is chance: a percentage that decides whether an attack is critical, an enemy drops an item, or a hit is avoided. Telling the two apart helps you understand, in the wider sense of luck and skill, which parts of your results you can control and which are subject to variation.
The values below are illustrative. Each game sets its own, and developers rebalance them over time.
Critical hits and average damage
A critical hit is an attack that has some chance of dealing extra damage. Two numbers describe it: the chance, and the multiplier.
Suppose an attack has a 20% critical chance and a critical hit deals double damage. The average multiplier across many attacks is 0.8 × 1 + 0.2 × 2 = 1.2. In the long run, that crit stat acts like a 20% damage increase. If the critical bonus is smaller, say 1.5 times damage with a 25% chance, the average multiplier is 0.75 × 1 + 0.25 × 1.5 = 1.125, about 12.5% extra damage.
This lets you compare stats directly. A 20% chance for double damage (an average of 1.2) beats a 25% chance for 1.5 times (an average of 1.125), even though the second one has a higher chance and probably triggers more often.
Average damage, however, says nothing about a short fight. A character with a 20% critical chance who attacks ten times will see no crits at all with probability 0.8¹⁰, which is about 10.7%. That is rare but far from impossible, and it will happen to players now and then. A lot of the frustration with “bad luck” comes from expecting criticals to arrive evenly spaced, when independent random events cluster and gap in ways that feel unfair. This is a feature of random events in general.
Dodge and evasion
Dodge, evasion and miss chances reduce incoming damage in a similar way. If an enemy attack is avoided 15% of the time, the long-run damage you take falls by 15%, because you take 85% of the hits.
Chances also stack across several attacks. The chance of avoiding at least one of three attacks, each with a 15% dodge chance, is 1 − 0.85³. Since 0.85³ is 0.614125, the result is about 0.386, or 38.6%. The chance of dodging all three is only 0.15³, about 0.3%. So a 15% dodge rate works as a gentle reduction in damage, not as protection you can rely on in any particular fight.
Because dodge is random, it adds variance without changing the average much. Two characters with the same total health can differ in how safe they are: one with high armour takes steady, predictable damage, while one with high dodge takes less on average but has swings in which several hits land in a row. Reliable damage reduction is easier to plan around; dodge helps on average but can fail at the worst moment.
Timing: where the dice stop
Many action game mechanics are not random at all. A parry, a perfect dodge or a combo cancel depends on pressing a button inside a window. These windows are usually measured in frames.
At 60 frames per second, one frame lasts about 16.7 milliseconds. A parry window of 6 frames is 6 ÷ 60 = 0.1 seconds, or 100 milliseconds. Wider windows are more forgiving; narrower windows reward practice. In either case, the result comes from your reaction and your timing, not a roll. Practice improves the outcome in a measurable way, and the variation you see in your own results reflects your consistency, not chance.
A game with a tight timing window and a game with a high dodge percentage can both reduce damage, but they feel different. One rewards execution and the other gives a probabilistic buffer. Developers often combine them, which is one of the core design choices in how video game design handles luck.
Loot, drops and independent trials
Item drops are another place where chance appears. If an enemy has a 5% drop chance for a rare item and each kill is independent, the chance of at least one drop in 20 kills is 1 − 0.95²⁰. Since 0.95²⁰ is about 0.358, the result is about 64%. So after 20 kills, about a third of players would still have nothing, and that is normal.
A common misunderstanding is the idea that a drop is “due” after a long dry spell. With independent rolls, each kill still has the same 5% chance, regardless of previous failures. Some games do change this deliberately: they raise the chance after each failure or guarantee an item after a set number of attempts. Those systems are intended to limit extreme bad streaks. If you want to know how a specific game behaves, look for official information, because not every title explains it.
Streaks are normal
The same pattern runs through everything above. Random outcomes come in clumps, and a run of good or bad results does not mean the game is rigged or that you did something wrong. Small samples are a poor guide, so avoid drawing conclusions from a few fights. If you want to judge whether a build with a higher crit chance really performs better, compare many fights, not three, and keep in mind that timing-based performance and chance-based performance are measured differently.
Practical takeaways
- Convert chances to average multipliers to compare stats.
- Treat dodge and crit as long-run effects, not guarantees.
- Separate the parts of a fight that depend on timing (and can be practised) from those that depend on rolls (which you can only plan around).
- Do not assume a result is “due”; ask whether the game actually adjusts its odds.
None of this removes the unpredictability that makes action games exciting. Understanding it mostly helps you stay calm when luck turns, and make steady choices about builds and tactics.