Conquest Lab methodology

Known Wins, Alternatives and Experiments Explained

Understand how Conquest Lab separates verified stage wins, evidence-based alternatives, utility experiments, and trust labels.

Intermediate Website HelpConquestData & Testing Updated July 16, 2026

Known winning teams

A known winning team is an exact five-hero formation with trustworthy winning evidence on the selected stage. It is the strongest type of evidence the public tool returns, but it is still not a guaranteed win for another account.

Differences in stars, Conquest skills, hero gear, widgets, and total development can prevent a recorded team from working at the same displayed power.

Tested and evidence-based alternatives

Alternatives may come from a trustworthy exact-stage lineup that needed roster replacements, a team tested against the same enemy on another stage, a nearby-stage pattern, or a broad utility fallback when no complete recorded team fits.

Always read the source and replacement note. A team with no substitutions is stronger evidence than a five-hero fallback assembled from roles and tags.

Experimental recommendations

An experimental recommendation tests a utility hypothesis. It normally starts from the strongest available evidence and replaces one hero with another that improves position fit, counters an enemy mechanic, or adds missing team utility.

Experimental means unverified

The exact lineup is intentionally separated from known wins. Its value is the clarity of the test, not a claim that it already works.

Trust labels

  • High: strong direct evidence with few or no substitutions.
  • Medium: useful evidence with some uncertainty.
  • Low: broad or indirect evidence that needs testing.
  • Experimental: a deliberate hypothesis rather than a recorded team.

What the labels do not mean

A trust label is not a win percentage. It does not know your exact skill levels or gear. It also does not mean that every row behind the recommendation was equally useful; overpowered, instant, and near-equal-power wins are excluded from trustworthy formation evidence.

Which result should you test first?

  • Use the known win first when your roster and development are close enough.
  • Use a tested alternative when the known team contains an unavailable hero.
  • Use the experiment to test one specific counter or utility idea.
  • Keep the exact positions shown.
  • Submit the result so the evidence can improve after manual review.

Conquest conclusions use the current raw Conquest Lab dataset. Arena conclusions are preliminary projections unless a section explicitly says otherwise. Exact private rows, lineups, powers, and ratios are not published.