Chance is built into most games
Almost every competitive game contains some randomness, even those that look purely skill-based. A card game shuffles. A shooter may add bullet spread. A strategy game may roll for damage or generate a different map each match. Even a game with no dice at all has chance at its edges: who you are matched with, which side of the map you start on, or which opponent a bracket draw hands you.
The useful question is not whether chance is present. It is where it enters, how large its effect is, and whether the player can do anything about it.
Where randomness enters a match
It helps to sort random elements by the moment they act.
Before the match. Matchmaking, team assignment, map selection, starting hands and tournament draws all decide the conditions you play under. You have no influence over these once they happen, though over many matches they tend to even out.
During the match. Critical hits, damage ranges, proc effects, loot and card draws change the state of play as it unfolds. These are the events players notice most, because they happen in front of you and can swing a moment.
In the measurement. The way results are recorded also adds noise. A short win-loss record, a single tournament placement or a ranking after a few games all reflect chance as well as ability.
Input and output randomness
Designers often distinguish randomness that happens before a decision from randomness that happens after it.
Input randomness gives the player new information and asks them to respond. Drawing a card, seeing the map or finding an item are examples. Skill shows in how well you adapt, and a strong player can often turn a mediocre draw into a decent plan.
Output randomness happens after the player has committed. A shot that has a chance to miss, or an attack that rolls for damage, are examples. Here the player chose well or badly, and then a die decides how it turned out. This kind of randomness hides the quality of a decision, which is one reason players find it frustrating and also one reason it keeps weaker players in the game.
Neither type is automatically better or worse. A game with mostly input randomness usually feels more skill-driven; a game with heavy output randomness feels more volatile. The balance is a design choice, and the way a studio approaches random mechanic design shapes how much of each a game contains.
How much does chance decide?
There is no single number, because the answer depends on how large the random swings are compared with the skill gap between the sides.
A simple model makes this concrete. Suppose a player wins each game against an opponent with probability 0.6. That is a real edge, but any single game is far from settled. Over a best-of-three series, that player wins the series with probability p squared times (3 minus 2p), which for p = 0.6 is 0.36 times 1.8, or about 0.648. Over a best-of-five, the figure is about 0.683. The stronger side wins more often as the series lengthens, but the weaker side still wins roughly one series in three. This is the core reason series length matters in tournament design.
The same logic applies in a bracket. In a single elimination event, one unlucky game ends a run, so even a clearly best team often fails to win the whole thing. Estimates of tournament win probability tend to look surprisingly flat for this reason, even when the favourite is clearly strong.
Why one result says little
A single outcome is a very noisy measurement. If a player’s true win rate is 55 percent, then over ten games they would expect 5.5 wins, with a standard deviation of about 1.6 wins. A 7-3 record and a 4-6 record are both entirely plausible for that same player.
This is why sample size matters so much when you judge a change, a build or a player. Small samples are dominated by chance, and conclusions drawn from them are fragile. As the number of games grows, the spread of the observed win rate shrinks in proportion to one over the square root of the number of games. Quadrupling the number of games roughly halves the noise.
Averages are stable, moments are not
Chance affects individual moments strongly and long-run averages weakly. A 1 percent item that drops on the first try and a 1 percent item that takes 300 attempts are both ordinary results from the same underlying odds. What stays reasonably stable is the average: the expected value of a single attempt, or of a single attack, can be calculated and relied on over large numbers.
Game systems sometimes intervene in this behaviour on purpose. Published drop rates describe the underlying chance, but some games layer on mechanics that make streaks of bad or good luck less extreme than pure randomness would give, which changes how a run of results feels without changing the long-run average much.
Chance and skill work together
Treating luck and skill as opposites is a common mistake. In most games both contribute to a result, and the useful question is the proportion. Skill shows up in the choices you control and in the average outcome over many games. Chance shows up in the gap between that average and what happened this time.
Seen this way, a loss does not automatically mean a mistake, and a win does not automatically mean good play. Players who track results over a meaningful number of games, and who practise separating luck from skill when they review, tend to draw more reliable conclusions than players who react to each result as it comes.
What to take from this
- Chance enters before, during and after a match, and each stage works differently.
- Randomness that comes after a decision hides decision quality; randomness that comes before one tests adaptation.
- A modest skill edge produces a clear advantage over many games and only a modest one in any single game.
- Short records, single tournaments and individual moments are poor evidence on their own.
- Calculating the average effect of a random event is usually more informative than remembering the most memorable roll.
None of this removes the unpredictability that makes games exciting. It simply gives a way to describe how much of a result belongs to chance, so that wins and losses can be read with a little more care.