Loot Table Design Checklist

This checklist covers the four things every loot table needs before launch: a documented drop rate structure, a tuned pity timer, duplicate protection for repeat drops, and playtested fairness under real random rolls. For the theory behind why raw randomness reads as unfair to players, see the loot table design guide; to have a specific loot table reviewed, book a free consultation.

Loot tables are easy to get mathematically "correct" and still have players call them broken, because the math and the felt experience of randomness are two different problems. This checklist is meant to be worked through item by item before a loot system ships, and it pairs well with the wider set of checklists and worksheets if you're auditing more than just drops.

Drop Rate Structure

Every drop is classified as guaranteed, weighted-random, or tiered-rarity — nothing is left ambiguous
Guaranteed drops are documented separately from random ones, with a clear reason each exists
Weighted-random pools have explicit probabilities assigned to every item, summing to 100%
Tiered pools have both a tier-selection probability and a within-tier item probability documented
Drop rates are recorded in a single source of truth, not scattered across code comments or spreadsheets

Pity Timer / Bad-Luck Protection

Every rare or high-stakes drop has an explicit decision on whether it needs a pity timer
The pity threshold is calculated from the actual drop rate, not picked arbitrarily
The threshold has been tested to confirm it isn't low enough for players to reliably grind toward it
A decision has been made on whether the pity counter is shown to players or kept hidden — and why
If hidden, the reasoning for hiding it is documented somewhere the design team can revisit later

Duplicate Protection

Every finite, collectible reward pool (characters, cosmetics, skins) has a duplicate-handling rule
Repeat drops convert into currency, crafting material, or credit toward a guaranteed unlock
The conversion value feels meaningful, not so small that duplicates still read as a wasted drop
Duplicate protection logic is tested against a player who owns most or all of the pool already
Edge case handled: what happens when a player owns literally everything in a pool

Perceived Fairness Testing

The drop table has been tested with real random rolls, not just verified against theoretical math
Worst-case streaks (extended runs of bad luck) have been reviewed, not just the average outcome
Actual probabilities have been modeled with a tool like the loot drop calculator to confirm the table behaves the way the design intends across many attempts
Playtesters have been asked directly whether a drop system felt fair, not just whether it felt rare
Any loot behind a paid mechanic has been reviewed separately for disclosure and fairness perception

Illustrative example A rare weapon drop is set at 4%. Simulating ten thousand players attempting it fifty times each shows most succeed well before the fiftieth try — but a predictable slice are still empty-handed. Reviewing that slice, rather than only the 4% headline number, is what perceived fairness testing is meant to catch.

FAQ: Loot Table Design Checklist

When should I run this checklist — during design or right before launch?

Both. Use it during initial design to make sure drop rate structure and pity logic are actually decided rather than left implicit, then run it again right before launch as a final pass, since playtesting often surfaces edge cases the original math missed.

Do I need a pity timer for every drop in the game?

No. Pity timers matter most for rare, high-stakes drops that players consciously track — signature weapons, character unlocks, key progression items. Common, low-value drops that players barely notice are usually fine left as pure random rolls.

How do I actually test perceived fairness before launch?

Run real random rolls against your drop table hundreds or thousands of times — either through playtesting sessions or a simple simulation — and look at the worst-case streaks, not just the average. The average is rarely what players complain about; the unlucky tail is.

Should pity timers always be visible to players?

Not necessarily. Some games show a visible pity counter to build anticipation, while others hide it so the guarantee reads as a pleasant surprise rather than a countdown players grind toward. Either can work — the mistake is not deciding, and ending up with a threshold nobody chose on purpose.