Gacha odds explained: what those percentages actually mean
A published rate is not a promise, and pity is not a safety net in the way most people assume. What the numbers on a banner really tell you before you spend.
- iPhone
- iPad
- Android
Several jurisdictions now require games to publish their drop rates, which was a real win. The unintended consequence is that millions of players are now looking at a number they have not been given any help interpreting.
A published rate is genuinely useful information. It is just not the information most people think they are reading.
The game has no memory
This is the single most important thing and it is the thing intuition fights hardest. Each pull is an independent event. The game does not know you have pulled fifty times without success, and — outside an explicit pity system — those fifty failures do not make the fifty-first any more likely.
"I'm due" is not a description of the system. It is a description of how it feels, and the gap between those two things is where a lot of money goes.
Why the average is not your experience
A rate is a statement about a very large number of pulls across a very large number of players. You are one player doing a small number of pulls, and small samples are wild.
This is why community anecdotes are so unreliable in both directions. The person who got it in three pulls is real. The person who went hundreds deep is also real. Neither tells you anything about what will happen to you, and the loud ones tend to be the extremes.
Pity is the number that actually matters
A pity system guarantees something after a set number of unsuccessful pulls. This is the only mechanism in most of these games that caps your personal worst case, which makes it the most financially significant number on the page — and it is usually less prominent than the headline rate.
- Find the hard pity number. How many pulls until something is guaranteed? That, multiplied by the cost per pull, is the real worst-case price of the thing you want.
- Check whether it carries over. Does the counter reset between banners? If it does, starting a new banner late is starting from zero.
- Check what is actually guaranteed. Pity often guarantees a rarity, not the specific character. Those are very different promises.
- Look for a soft pity. Some games raise the rate as you approach the threshold. Where it exists it changes the shape of the maths considerably, and where it does not, the flat rate holds all the way.
Work out the worst-case cost in real money before you start. Not the expected cost — the worst case. If that number makes you uncomfortable, you have your answer, and you got it before spending anything.
The framing tricks worth noticing
- Rates quoted per multi-pull
- A ten-pull's headline chance sounds far better than the per-pull rate. It is the same maths in friendlier clothing.
- Currency between you and the price
- Bundles that do not divide evenly into pull costs leave you with an awkward remainder. That remainder is doing a job.
- Rate-up that is not exclusivity
- "Increased chance" often still leaves most of the probability on things you were not aiming for. Read what share of the rate is the featured item.
- A countdown on the banner
- Urgency exists to prevent the comparison you would otherwise make. The clock is a design element, not a fact about value.
Where to actually read the rates
In the game, on the banner, in the disclosure screen the regulations require. Not from a community post, and not from us — we would be quoting a number that changes with every banner, which is exactly how out-of-date figures spread.
The short version
The rate tells you about everybody. The pity counter tells you about you. Decide your worst-case number in pounds before the banner is in front of you, and treat the countdown as the sales tool it is.
How we worked this out
The mathematics here is the behaviour of independent trials, which is fixed and checkable. No specific game's rates, pity thresholds or banner terms are quoted — those vary by title and change, and the game's own disclosure is the source to read.
Written by
PhonePlayers Editorial Team
Editorial
The shared byline for evergreen resources that are maintained over time rather than written once, and for work produced by more than one contributor. Articles currently carrying this byline were drafted with AI assistance against primary sources and have not yet been fact-checked by a named person — each one says so at the top, and the AI policy explains exactly what that means. Once a human editor verifies a piece, that notice is removed and the editor is named. Where an article depends on firsthand play, it carries an individual byline instead.
Disclosures
PhonePlayers is published alongside Clashest, Brawlest, Bloxism and Pokémonism, which are owned by the same people and are linked from this site where they are the better destination for live data. That is a commercial relationship and we say so on every page it appears. No publisher has supplied review access, hardware, currency or payment for any article currently published. There are no affiliate relationships in place today; if that changes, the affiliate disclosure page will say so before any affiliate link is published.
Spotted something wrong, or something that has changed? Tell us — we date and describe every correction we make.
