Calculator Editorial
What This Deck Health Read Is Measuring
This page is trying to answer one blunt question: is the deck structurally sound, or is it hiding the kind of flow and coverage problems that only show up once a hallway fight goes bad. The numbers are simple on purpose, because decks usually break for simple reasons.
Most bad decks are slow, top-heavy, or missing real jobs
Players often lose with decks that look powerful on paper because the shell is lying: it draws too slowly, spends too much Energy on the first pass, or has no clean answer to wide fights.
This tool turns that vague feeling into a structural read. It tells you whether the list is compact or bloated, how quickly it cycles, what jobs it covers, and which warnings matter first.
- Use it to audit the deck you are actually carrying, not the deck you wish you had.
- Use it before taking more payoff cards, because deck health usually dies from support gaps first.
The card list is the data structure, everything else is noise
Character choice limits the card pool. Search, count, and upgrade state build the exact deck list that gets analyzed. Include Colorless matters because those cards can add draw, energy relief, or coverage the main color is missing.
Every row in the builder is a real deck entry with a count and an upgraded flag. If the list is wrong, the result is wrong. This page is intentionally unforgiving about that.
- Count changes density, not just raw copies.
- Upgrade state matters because the upgrade scorer compares base and upgraded card analysis directly.
- Clear Deck is not cosmetic. It resets the shell when you want to audit a fresh run state.
The thresholds are simple and explicit
Deck size bands are fixed: Lean at 14 or less, Steady at 20 or less, Wide at 26 or less, and Bloated above that. Cycle speed is estimated from deck size and total draw, then labeled Fast at 3.0 turns or less, Steady at 4.4 or less, and Slow above that.
Role coverage is scored against density thresholds instead of raw counts. Block targets 22%, Draw 12%, AOE 8%, Energy 8%, and Scaling 12%. Curve risk fires when 2-cost and 3+ cost cards push above 30% density without stable energy help.
- Role rows are built from card analysis tags like block, draw, all-enemy damage, energy, scaling, and frontload.
- Warnings are not random; they come from size, cycle speed, role density, and curve density checks.
- Upgrade priority rewards cost reduction, extra draw, extra energy, stronger block or damage, and engine payoff changes.
The page reads structure, not every combat edge case
This analyzer does not know your relics, route, potion stash, exhaust loops, or generated cards unless the deck list itself already captures that effect. It is a shell audit, not a full run simulator.
That is also why the page stays valuable. It catches the boring structural mistakes that survive regardless of seed: too much cost, not enough draw, missing AOE, or no real defensive floor.
- If a relic completely changes the deck tempo, treat the output as a baseline rather than a complete verdict.
- If the deck wins through a very specific infinite or external combo, this page will understate that plan unless the card list visibly supports it.
24-card Ironclad shell with weak flow
Suppose the list has already grown to 24 cards, the expensive half of the curve has swollen past the 30% risk line, draw support is still thin, and there is no clean all-enemy answer. That is the kind of deck that looks fine in isolated hands and then collapses in actual hallway fights.
The analyzer will not romanticize it. It will call the deck wide, push cycle speed toward the slow band, flag the curve as top-heavy if energy support is still thin, and warn that wide fights remain awkward because the AOE row is missing.
- 24 cards already sits above the steady size band.
- One or two draw tools are rarely enough to keep a 24-card shell out of the slow cycle read.
- When expensive cards exceed 30% density without energy support, the curve warning is expected, not surprising.
- No Whirlwind-style all-enemy coverage means the AOE row stays missing and hallway risk stays live.