Paid chance, with a built-in edge
Gacha games sell repeated random draws, much like a slot-style game. The structure is designed so the seller benefits over many purchases, and no timing, ritual, or counting method changes the odds. Outcomes are uncertain, and many players spend more than they planned. Please spend only what you can afford to lose, and treat this as entertainment with a price, not an investment.
What gacha means
The word comes from Japanese capsule-toy machines, where a coin buys a random toy. In video games, a gacha is a draw system: the player spends currency, receives a random character or item, and often does so in a repeated series of “pulls.” A pull is similar to opening a loot box, except that gacha systems often feature a rotating list of featured rewards and a set of tiers.
Each pull normally uses a weighted table and random number generators. In the simplest form, every pull is independent. The odds for the next pull do not change because of earlier results, and a long run of misses does not make a hit more likely. This is the lesson of independent events.
The problem pity systems address
If every pull is independent, some players will go a very long time without a top reward. Here is a hypothetical example. Suppose a top reward has a 2% chance per pull. The chance of failing one pull is 98%, or 0.98. After 100 pulls, the chance that every one has failed is 0.98 raised to the 100th power, which is about 13%. So roughly 1 in 8 players would spend 100 pulls and still have nothing. Such bad runs are unavoidable under pure chance, and they can be frustrating enough to make players leave.
A pity system is a rule that limits how bad the run can get. A counter tracks the number of pulls since the last top reward. When it reaches a threshold, the system guarantees one.
Hard pity
Hard pity is a fixed guarantee. If the threshold is 100 pulls, then the 100th pull is certain to give the reward if none came before. The longest possible drought is therefore 99 failed pulls followed by a guaranteed result.
What does this do to the average? Using the same hypothetical 2% rate and a hard pity at 100, the expected number of pulls is the sum of the chances of continuing: 1 plus 0.98 plus 0.98 squared, and so on up to 0.98 to the 99th power. This works out to (1 minus 0.98 to the 100th) divided by 0.02, which is about (1 minus 0.1326) divided by 0.02, or roughly 43 pulls. Without pity, the average at 2% is 50 pulls. So in this example, the guarantee reduces the average cost modestly, from about 50 to about 43, but the typical player still spends a similar amount.
Soft pity
Soft pity is a gradual version. After a certain number of misses, the per-pull chance begins to rise. Perhaps it climbs from 2% to 10% and beyond over the next several pulls. This tends to make the reward arrive somewhere in a predictable window, with the final guarantee as a backstop. The details differ from game to game, and the rate may only be partly disclosed. This is why published drop rates matter: a single percentage may not tell the whole story.
Carry-over and “50/50” rules
Some games add a second layer. When the top reward arrives, there might be a fixed chance it is the featured item and otherwise it is a different item from the same tier. A following “guarantee” may then promise that the next top reward will be the featured one. This makes the full cost of a specific item a combination of two probabilities. Because each stage has its own pity counter, the worst case can be double the apparent threshold. This is an instance of compound probability.
Why the sums can mislead
Several features make it difficult to judge value:
- Counters can be reset or carried over. The rules depend on the banner and the time. Check them before you pull.
- A guarantee reassures, but it also encourages continuing. Players who have already spent a lot may feel they are “close” to the guarantee, a form of sunk cost thinking.
- Average is not the same as your result. Expected cost is an average across many players; yours may be much higher or lower.
- Currency layers hide cost. If pulls are bought with a premium token, you need to convert the cost to real money, as discussed under virtual currency.
What pity does and does not do
Pity makes the worst case more predictable. It does not make the draw free, fair, or good value. A limited cap on bad luck is still a cap on a price the seller set. Whether the item is worth hundreds of pulls is your decision, and for many people the sensible answer is to skip it.
Staying in control
- Decide a fixed budget before you start pulling and stop when you reach it.
- Do not treat a pity counter as a reason to continue past your limit.
- Be careful with offers that appear only for a short time, which are discussed under dark patterns.
- Use account spending controls if you tend to buy on impulse.
If gaming purchases feel difficult to stop, or if you are borrowing to fund them, a national problem-gambling helpline or local health service can help. Rules about randomized purchases and age limits differ by country.