Event Choices: Success Chances and Rewards

A strong event stat improves your chances; it does not generally turn an event into a guaranteed success. Start by checking which cat participates, what the option tests, and what each possible outcome actually gives.

Ordinary Stat Checks Are Probability Checks

SciresM’s analysis of the event code describes ordinary non-wheel event rolls. The February 2026 model uses the tested stat, difficulty, whether that stat is the participating cat’s highest or lowest, an optional bonus from an earlier event, and Luck.

In that historical model:

  1. Calculate base = clamp((5 + stat - difficulty + 2 × relativeBonus + previousBonus) × 0.1, 0, 1).
  2. Calculate target = 0.05 + 0.9 × base.
  3. Make the Luck-adjusted roll against that target.

The difficulty values are 5, 6, 7 and 8 for Normal, Hard, Crazy and Impossible. The relative bonus is +1 for a highest stat, -1 for a lowest stat, and 0 if all stats are tied. Before Luck or special overrides, the target stays between 5% and 95%.

Worked historical example: on Normal, a tested stat of 7 that is the cat’s highest, with no previous bonus, gives base = 0.9 and target = 86%. If all its stats are tied at 7, the relative bonus is 0 and the target is 68%. Both values are targets before the Luck-adjusted roll, not final success rates. They are calculations from an older code analysis, not current-build measurements.

Luck also affects critical success and critical failure in that analysis. A high STR cat with low Luck and a high STR cat with high Luck are therefore not equivalent event participants.

Some Options Follow Different Rules

Not every choice has both a success and a failure branch. Leaving may have one outcome, and some events use fixed chances or other overrides. Do not apply the ordinary-stat formula to every wheel, quest or payment option.

Coin checks are a separate case: the historical code chooses the good or bad branch based on whether you have enough coins. That does not make the purchase free or its reward fixed.

For example, the extracted A Beggar event gives its 5-coin donation’s good branch a 95% common reward pool and a 5% rare reward pool. The 10-coin option changes those weights to 90% and 10%. Affording an option and receiving its rarer reward are different questions. These values are from the site’s older data snapshot.

Choose for the Outcome You Need

Before selecting an option:

  1. Check the participating cat and its relevant stat and Luck.
  2. Read both available branches, including common and rare rewards inside each.
  3. Compare the cost and possible harm with the benefit: an item pool is not a promise of a particular set piece.
  4. Choose based on the current run’s needs. A useful failure reward can be preferable, but you cannot normally guarantee that failure.

The snapshot has 68 STR options, 62 INT and 48 LCK; CON and DEX have 46 each, CHA 23 and SPD 18. These counts describe recorded options, not how frequently you encounter them or a measured team-building advantage. Luck matters beyond options explicitly labelled LCK.

Use the Events reference to inspect a particular choice, Items to evaluate rewards, and Combat when comparing the risk of an extra fight.

For quest events, follow the hidden-area unlock steps: a stat roll is not a substitute for equipping the required item or completing an earlier quest action.

Changes and Sources

The official 1.1 patch notes removed bad outcomes from Baphomet’s sacrifice and lose-10-HP options, and from Sacrificial Altar sacrifices. Itchies now also grants a lasting flea familiar; Wheezies also grants +2 LCK. Old outcome tables can differ from these changes.

Reviewed on October 9, 2026. The probability model comes from the linked original reverse-engineering notes. Choice examples and counts come from this site’s extracted event data, whose game build was not recorded. The examples are useful for explaining the distinction between a check and a reward pool; compare changed events with your installed version before relying on their exact outcomes.

Last updated: October 9, 2026