The balance every designer has to pick
Every board game that includes chance has made a choice about how much. At one end, results are decided mostly by dice or cards and decisions matter little. At the other end, results follow almost entirely from planning, and the better player wins nearly every time. Most games sit somewhere in between, and the right point depends on the audience. A family game can lean toward luck so that children and adults compete evenly. A game for experienced players may lean toward skill so that good play is reliably rewarded.
The luck versus skill balance is not a single dial. The same amount of randomness can feel very different depending on where in a turn it appears.
Input randomness and output randomness
Input randomness happens before a decision. A card is drawn, a tile is revealed or dice are rolled, and then you decide what to do with the result. Because players respond to what has appeared, skill shows in how well they use it.
Output randomness happens after a decision. You choose a move, and then a die determines whether it succeeds. A good plan can fail for no fault of the player.
Neither is inherently better. Input randomness tends to keep skill visible and make a game feel fair. Output randomness tends to create tension and memorable swings, but a long chain of dependent output rolls can make planning feel pointless. Many designers use output randomness for small, frequent events, so that bad luck averages out, and input randomness for important decisions.
Why more rolls can make a game fairer
Randomness matters less when it is spread across many small events. A single roll that decides the whole game is mostly luck. A game with fifty small rolls tends to see the results average out, so the player with better decisions has more chances to pull ahead. This is the law of large numbers working inside a game.
An illustration: suppose Player A has a 55% chance of winning each small contest, based on better positioning, against Player B’s 45%. In one contest, A wins only slightly more often than not. Across many contests, A is much more likely to come out ahead overall, though a streak for B remains possible. Designing for many small contests rather than one large one makes a modest skill edge show up more clearly.
Catch-up mechanics
A trailing player in a long game can feel the result is already decided. Designers use catch-up mechanics to keep the game open. Common ideas include:
- Bonuses for trailing players. Extra resources, better draws or stronger moves for players who are behind.
- Costs for leaders. Leaders may face extra penalties or become targets.
- Diminishing returns. Each additional unit of lead is worth less than the last.
- Hidden or delayed scoring. Final points are revealed late, so no one can be sure who leads.
These tools have trade-offs. Too strong a catch-up effect removes the reward for playing well, since a lead is not worth building. Too weak, and players who fall behind early simply wait for the game to end. A common aim is to keep outcomes uncertain without making the early game meaningless.
Measuring balance with results
Designers and playtesters often look at data. If a game is intended to reward skill, experienced players should beat new ones consistently. If a first player wins much more often than the second across many plays, there is probably a structural advantage. Evidence like this needs enough games to be reliable; a handful of results can mislead, as discussed under small sample sizes. Win rates can be summarised with confidence intervals to show how uncertain they are, and rating systems such as Elo ratings are sometimes used to track how strongly results reflect skill.
A simple test follows. Have a strong player and a beginner play a set of games. If the beginner wins a third of the time, luck is doing a lot of work. If the beginner almost never wins, skill dominates. Neither is a flaw on its own; it only matters whether it matches what the game is meant to be.
Hidden information and bluffing
Not all uncertainty comes from dice. Hidden cards or secret roles introduce uncertainty that players can partly reduce through inference. Reading hidden information is a skill, and it makes luck feel less arbitrary because players can reason about what is probably true. Games built on hidden information often feel more skill-based than games with the same amount of dice luck.
Where fairness and fun can diverge
A game can be balanced and still unsatisfying, or unbalanced and still loved. Players tolerate substantial luck when the game is short, when the luck produces funny or dramatic moments, or when decisions between rolls feel meaningful. They tolerate less when a long investment can be wiped out by one roll. Because feelings about luck are partly emotional, numbers and playtests need to be used together.
Streaks also colour perception. Truly random results include runs of good or bad outcomes, and a player on a bad run may suspect the dice or the design. Understanding that streaks are normal can reduce frustration, though designers may still add mechanics to soften extremes.
A short checklist for evaluating a game
- Where does the randomness occur: before or after decisions?
- How many separate random events does a game include, and how big is each?
- Does a skilled player win more often over repeated plays?
- Is there a catch-up effect, and does it still let a lead matter?
- Does hidden information give players something to reason about?
Games with real-money gambling work differently. Casino games are set up so that the operator has a long-run advantage, and skill has little or no effect in most of them. If you play such games, treat it as entertainment you pay for, set limits in advance and seek local support if it becomes a problem.