BFCBrilliance

How Drop Rates Actually Work

A 1-in-100 drop is not guaranteed in 100 tries — it's about 63%. The maths behind grinding, and why a long dry streak proves nothing.

By BFCBrilliance··4 min read

The 1-in-100 problem

Everyone's first instinct: a 1% drop rate means 100 kills and you've got it.

It doesn't. In 100 attempts at 1%, your chance of at least one drop is about 63%. More than a third of players will still be empty-handed, and nothing has gone wrong.

That gap between what feels obvious and what's true is why grinding feels so much worse than it looks on paper.

Your details

The rate per attempt, from the wiki.

Kills, runs, chests opened.

Result

Chance of at least one by now
92.31%

The odds you would have seen it at least once.

Chance of still having noneNot bad luck until this number is small.
7.69%
Drops expected on averageAn average across many players, not a promise.
2.5
Attempts for a 50% chanceThe coin-flip point. Half of players have it by here.
14
Attempts for a 90% chanceWhere most people have it. One in ten still will not.
45

Open the Drop Chance Calculator on its own page to bookmark or share it.

Why it's 63% and not 100%

You can't add probabilities together. What you can do is work out the chance of the drop not happening, which is much easier, and take it away from everything.

Each attempt is independent. So:

  • Chance of missing once: 99%, or 0.99
  • Chance of missing twice in a row: 0.99 × 0.99
  • Chance of missing 100 times: 0.99¹⁰⁰ = 0.366

So there's a 36.6% chance of getting nothing, which means a 63.4% chance of at least one.

The general form is the only formula you need:

chance of at least one = 1 − (1 − rate)ⁿ

That's it. It's the same in every game with a fixed drop rate. The only game-specific number is the rate, and you get that from the wiki.

Expected drops isn't a promise

Rate × attempts gives the average across everyone doing the same grind. At 5% over 50 kills that's 2.5 drops.

Nobody gets 2.5 of anything. Some players get none, some get five, and the average of all of them is 2.5. It's a description of the crowd, not of your afternoon.

The 50% and 90% points

More useful than "expected drops" is knowing where you sit.

Run the same equation backwards with a logarithm and you get the number of attempts before the chance of still having nothing falls to a half, or to a tenth:

  • 50% point — half of all players have it by here. It's usually much sooner than people expect. At 5%, it's 14 kills.
  • 90% point — most people have it. At 5%, it's 45 kills.

Those two numbers tell you the shape of the grind before you start, which is worth knowing when deciding whether to bother.

A dry streak is not evidence

This is the part worth internalising.

The game does not remember your last 200 attempts. Every roll is fresh. Being 200 kills deep changes nothing about the next kill — it's the same rate it was on kill one.

So when you're past the 90% point with nothing:

  • 90% means one player in ten is still empty-handed there
  • Someone has to be that player
  • It isn't bad luck in any meaningful sense, it's what 90% means

Judge it by the chance of still having none. At 7% it's routine. At 0.5% it's genuinely unusual, and worth double-checking that the rate is right, that the patch didn't change it, or that you're actually eligible for the drop in the mode you're playing.

Where this doesn't apply

The maths assumes fixed, independent rates. Plenty of modern games deliberately break that:

  • Pity systems — a guaranteed drop after N failures
  • Bad-luck protection — your rate quietly rises with each miss
  • Escalating rates — common in gacha games
  • Loot tables that shift with progression, difficulty or luck stats

Once any of those are in play, these numbers become a floor rather than the real odds — your true chance is better than the plain maths says. That's the point of pity systems.

The way to tell is the patch notes and the wiki. If a game has bad-luck protection, players will have documented it.

Where people go wrong

  • Adding percentages. Ten 10% attempts isn't 100%, it's about 65%.
  • Believing you're "due". The most expensive gambler's fallacy in gaming. Nothing accumulates.
  • Trusting a forum-guessed rate. Wiki or patch notes. A rate that's off by a factor of two makes every number here wrong.
  • Not checking for a rate change. Drop rates get patched quietly and old guides go stale.
  • Grinding the wrong source. Some drops are mode-specific, difficulty-specific, or need a condition met.
  • Forgetting the rate might be per-item, not per-kill. A chest with three rolls is not the same as one.

The short version

Chance of at least one = 1 − (1 − rate)ⁿ. A 1% drop over 100 tries is 63%, not certain. Expected drops is an average across everyone, not a schedule. Work out your 90% point before you start, and remember a dry streak is exactly what a fixed rate looks like from the inside.

Free tool

Drop Chance Calculator

Your real odds of getting a drop over many attempts, and how many tries it takes to be reasonably sure. Works for any game.

Open the tool →
#drop-rate#loot#probability#grinding#rng

Enjoyed this? Get the next one.

New articles straight to your inbox. No spam, ever.

Keep reading