What risk of ruin means
Risk of ruin is the probability that you run out of a resource before reaching a goal or finishing the game. The resource might be hit points, gold, ammunition, lives, tokens, or money. It is a different question from “what is my average result?”, because ruin is a one-way door: once you are out, you cannot benefit from the favourable long run you were relying on.
That is why an average can be reassuring and still mislead. A game where your expected change is zero or even positive can end in ruin for some players, simply because bad luck arrived before good luck had time to balance it.
A simple model
Imagine a fair coin game. You start with a stack of units, bet one unit per flip, and win or lose one unit with equal probability. You stop if you reach zero or hit a target.
For this fair, even-money game there is a clean result. The chance of reaching the target before reaching zero equals your starting stack divided by the target.
- Start with 10, aim for 20: chance of reaching the target is 10/20 = 50%, so the chance of ruin is 50%.
- Start with 10, aim for 100: chance of reaching the target is 10/100 = 10%, so ruin has probability 90%.
Nothing in the coin is unfair. The more ambitious the goal relative to your reserves, the more likely you are to hit zero first. And if there is no target at all and you keep playing indefinitely, a fair game eventually reaches ruin with probability 1, simply because a random walk with a floor tends to wander into it eventually. That statement is about unlimited play, and it says nothing about any particular finite session.
Stake size matters
Keep the starting stack at 100 units and change only the bet.
- Betting 1 unit per flip, you cannot be ruined in four flips. Your stack would still be 96 at worst.
- Betting 25 units per flip, you are ruined after four losses in a row. The probability is (1/2)^4 = 1/16 = 6.25%, and that is just in the first four flips.
Larger bets relative to the stack make a ruinous streak both shorter and more likely. Splitting the same stack into more, smaller stakes moves it closer to the average, which is the law of large numbers helping you. This is the sense in which stake size and spread jointly set your risk.
Spread matters as much as the average
Two games can have the same expected value per turn and very different ruin risk. A low-spread game, with small changes per turn, takes many turns to cause trouble. A high-spread game, with occasional large losses, can drain a reserve quickly. Measuring that spread is the job of variance and standard deviation: the larger the standard deviation compared with your reserve, the more of the risk you carry.
The expected value still plays a part. A negative one tilts the walk downward, so ruin becomes more likely, faster. A positive one tilts it upward and reduces ruin risk without eliminating it, particularly when the reserve is small.
Where this shows up in games
- Roguelikes and survival games. Health, food or consumables are the stack. A single high-damage event in a low-health state is the equivalent of a large stake. Good roguelike balance often means limiting how often a bad draw can end a run.
- Deck and resource management. Spending most of your reserve on a single risky play leaves little room if it fails.
- Tabletop campaigns. A character with few hit points relative to enemy damage faces higher ruin risk per fight, even when the average fight looks easy.
- Real-money games. The stack is money, and ruin is not a mathematical abstraction. In any game with a negative expected value, such as most casino games, extending play raises the chance of loss, and no staking plan changes the average. Betting systems cannot overcome a house edge, and results remain uncertain. If gambling is straining your finances or wellbeing, help is available, as covered in material on responsible play.
Ways players reduce ruin risk
These are descriptive habits, not guarantees.
- Keep stakes small relative to your reserve. Surviving more turns gives the average more time to work.
- Avoid single events that can wipe you out. If one outcome can end the game, treat it as a separate, larger risk.
- Set a stopping point in advance. A pre-set limit on how much you are willing to lose, or a goal to quit on, avoids decisions made after a bad streak.
- Do not chase losses. Raising stakes after a loss increases ruin risk; it does not make the previous result more likely to reverse. The gamblers fallacy and sunk cost thinking can both push toward this mistake.
A limit of the simple model
The coin game above is a clean simplification. Real games have uneven payouts, rule changes, opponents and choices. The principles carry over: a large goal relative to your reserve, large stakes and high spread all raise risk of ruin. Exact figures need a model of your specific game, and any such figure is an estimate rather than a promise.