Absy Labs ABSY LABSKingshot Theorycraft Launch the simulator

Questions

Frequently asked questions

The short answers. Each links to the longer version where there is one.

Is the simulator accurate?

It's anchored to a known State-of-Survival round-zero example that produces exactly 189 deaths, kept green as a permanent test. The confirmed math comes from the open-source SoS engine and is cross-checked against in-game reports. Anything only probable or speculative is labelled and usually made a toggle. Full methodology →

Does KingShot have a fatigue factor?

Speculative. The ~0.01%-per-round fatigue is confirmed in the SoS engine but not independently verified for KingShot. It's assumed from shared lineage, so it's a toggle you can switch off. See the term →

What's the cavalry bypass rate?

About 20% per round per cavalry-attacking fighter — on success the cavalry skips the frontline and hits archers. Confirmed ability, community-quoted rate. More →

What's the archer volley rate?

A 10% chance to attack twice in a round. Probable — in the description, not independently confirmed for KingShot.

Can epics lead a rally?

No — leaders are mythic only, one Infantry, one Cavalry, one Archer, no duplicates. Epics join as joiners. Rally rules →

Do joiner skills stack?

Non-chance skills stack additively in their op family; chance skills don't stack across copies (four Jabels = one roll). Only the first four maxed first-skills count. Why →

Why do two Chenkos and two Amanes beat four Chenkos?

Same op adds, different ops multiply. 4× op 101 = 2.00; 2× op 101 and 2× op 102 = 1.50 × 1.50 = 2.25 — a 12.5% edge for the same total. Worked examples →

Does joiner gear matter?

No. Joiner gear and stats are ignored entirely — only the first skill and the troops count. Confirmed.

How are combat losses split?

Into permanent dead, severely wounded and lightly wounded, by battle type — a city attack is harsher than open field; a Bear rally costs nothing permanent. Full table →

What does it cost?

Nothing. It's free, runs on Streamlit, and is supported through Ko-fi if you want to chip in.

Which battle types are modeled?

Expedition rally vs garrison, with open-field and outpost casualty variants. Not Conquest duels or the Bear Hunt damage model — those are different engines.

I think a value is wrong. What do I do?

Bring it. The reliability grades exist to be upgraded with evidence — the Discord is where that happens.

Still have a question?

Ask in Discord — theorycraft welcome, screenshots even more so.