About the game observations

Game reference and current rules

The reference describes Balance 26. It includes stats, abilities, variants, and evolution links.

The run, opponent search, lives, stars, MMR, and season reset rules use the current source review.

The review covers build 26078162, Balance 26 / 1.4.0.

Read How to Play and the rank guide for these current rules. The source review date is 2026-10-22.

Merge modifiers and the Berroon shop exception received a separate source correction review on 2026-10-17.

This review inspects the current game source. It does not measure live opponent pools or prove every runtime path.

Tactical notes give conditional advice. They do not claim a measured best strategy.

Public observations

Leaderboard ranks describe public observations, not a live result for every player.

Player profiles show public rank statistics. A missing field does not mean a zero result.

Entity and pair counts describe repeated observations. An appearance on a team is not a shop offer or gift pick rate.

Trainer counts describe observed runs, not offers or picks. Private Companion matches are not part of these samples.

How Comps summarizes teams

Comps uses public ranked final-team observations. The sample selects up to 5,000 recent candidate runs by date and rank band.

The last observed round must match the reported round reached, with three to six recognized units.

Recurring three-unit cores define team groups. Each eligible board belongs to at most one group.

The team grid shows representative units, not verified positions or a recommended exact placement.

The 10-win finish rate divides qualifying 10-win runs by terminal runs in the group.

The calculation treats at least 10 wins or 5 recorded losses as terminal. This observation rule is not the game's life-loss rule.

The tier score averages results within each player first, then gives players equal weight.

The score pulls each group's finish rate and capped-wins fraction toward the sample baseline with 20 player-equivalents.

The score uses 70% finish-rate difference and 30% capped-wins difference. Wins above 10 do not add score.

A qualified group needs 40 runs from 25 players, including 25 terminal runs from 15 players.

Letters appear only with at least 8 qualified groups. A 30-day view can widen to 90 days when the comparison is too small.

Read the displayed window and sample context. Item ideas describe effect compatibility, not measured purchase success.

Team Builder makes an editable plan. It does not calculate your win probability.

What the sample cannot establish

Public histories cover only some players and runs. Missing rounds, older profiles, selection bias, and small samples can change results.

The raw finish rate counts runs. The tier score balances players. Neither predicts your next battle.

A team or trinket can accompany success without causing it. Player skill, opponents, and earlier choices also matter.

Some runs lack a balance label. Date-based patch attribution is an inference, not proof that every board used the current balance.

Use the window, sample size, observation date, and balance together.

Public statistics and private matches

Public pages need no account. A public player ID is not a Steam ID or a Companion account ID.

Companion explains the download and connection steps. Sign in with a six-digit email code to see your own uploaded runs. Other people cannot see them.

The separate example uses synthetic records, not your personal matches. It is not evidence of a live private account flow.

Do not put private credentials in a shared plan. Read the privacy policy for complete details.

Observation dates and corrections

An observation date does not mean that every player changed at that time. It is not a new game release.

A recent observation does not establish complete coverage. Interruptions can delay updates.

Check observation dates and the release notes before you compare results.

Editorial responsibility

Batomon.com maintains this independent community reference and its methods. The game developer supplies the game, not this site's conclusions.

We distinguish printed catalog values, checked rules, and public observations. A new build does not automatically renew an older tactical review. Dated notes stay gated to their reviewed payloads.

Report a correction with evidence

Email the correction to batomon@di.gl.

Include the page URL, the visible claim, the build or observation date, and what differs. Do not send codes, cookies, private run files, or unrelated personal information.

For a privacy or deletion request, use the existing contact in the privacy policy.

For a practical next step, read the statistics guide or use the planning checklist.