SorceryRec
About · Methodology

How we classify archetypes

A short field guide to the labels you see on every deck — and what to do when you disagree with one.

§1Four labels, one per deck

Each deck on SorceryRec gets one of four archetype labels:

  • Aggro. The deck wants to put pressure on early and end the game before the opponent stabilises. Low-curve threats, single- or two-element atlas, very few defensive minions.
  • Control. The deck answers what the opponent does and grinds them out with card advantage. Heavy removal, atlas that does work (draw/recur/lock sites), late-game finishers that double as answers.
  • Midrange. Efficient threats and efficient answers, in roughly equal measure. The deck holds equilibrium rather than tipping fast or grindy. Not a default fallback — a deck has to earn the label.
  • Combo. The deck wants to assemble a specific 2–3 card interaction that wins the game. Often runs heavy draw/disruption, with 4-of "key cards" the rest of the deck visibly enables.

If the heuristic can't make a confident pick, the label is simply — (Unknown) instead of laundering uncertain decks into Midrange.

§2Overlays — second-axis flavour

On top of the primary archetype, a deck can carry up to three overlays — Sorcery-native flavours that explain how the deck actually contests the realm without replacing the macro label:

  • Burn — the deck closes through direct face damage (typical for Aggro mono-Fire shells).
  • Tokens — the deck threatens through generated bodies (Aggro or Midrange).
  • Ramp — the deck accelerates atlas / cost reduction.
  • Site-centric — the atlas is doing more work than the spellbook.
  • Recursion — the deck wins by replaying threats from the graveyard.
  • Aura-stack — the deck wins by buffing a single minion through a stack of auras.
  • Air Pressure — evasive Air threats are a major damage axis.
  • Ground Flood — cheap minions try to occupy and pressure the board early.
  • Big Unit — the deck pivots around 5+ cost finishers or large bodies.
  • Defensive Sites — sites are used to stall, lock, recur, or otherwise protect a game plan.
  • Threshold Attack — the deck attacks opposing sites or elemental thresholds.
  • Cemetery Engine — the cemetery is a repeatable resource, not just a discard pile.
  • Spell-dense — magics make up a large share of the spellbook.
  • Spatial Tempo — movement, bounce, projectiles, or positioning effects are central.

That's why a deck reads as Aggro · Burn or Control · Site-centric: the primary tells you the plan, the overlay tells you the flavour.

§3What the classifier reads

The label isn't guessed — it's computed from your deck's shape:

  • Spellbook composition — cost curve, type mix, role of each card (threat / answer / engine / enabler / finisher / protection / bridge).
  • Atlas composition — size, element coverage, threshold density, mobility, recursion. These are context signals, not enough to classify the primary archetype by themselves.
  • Card diversity — combo decks lean low (4-ofs of key cards) but pad with 1-of utility.

Each archetype gets a score between 0 and 1; the highest-scoring archetype wins, but only if it clears a confidence floor. Otherwise the label is Unknown. Confidence is shown in the score breakdown on each deck's detail page.

§4Override priority

The label you see comes from the first source available, in this order:

  1. Author-set — when a builder uses the Archetype dropdown on their own deck.
  2. Curator-set — when SorceryRec features a deck and the curator picks an archetype.
  3. Admin-set after dispute — when a community dispute is accepted (more on this below).
  4. Heuristic — the score-based default for everything else.

Manual overrides display 100% confidence and a small badge indicating the source.

§5If you disagree — the "I disagree" link

Every classified deck shows an I disagree link next to its label. Click it and a small modal lets you propose:

  • The primary archetype you think the deck actually is
  • Up to three overlays
  • An optional one-line reason

We don't flip the label automatically. Disputes go into a queue; an admin reviews each one and either accepts (which sets the deck's admin-archetype slot, overriding the heuristic going forward) or rejects (the heuristic stays). One person can't spam-vote a deck into a wrong label, and we don't aggregate proposals into a majority — the admin decides each time.

Anti-griefing: each user is rate-limited to 5 disputes in any 24-hour window, and a previously-rejected dispute on a deck doesn't auto-reopen if the same user files again.

§6Card-level tag disputes

The classifier reads from per-card tags grouped into three families: effect (what the card does), role (what it's for in a deck plan), and site effect (only for atlas Sites). Every card detail page shows its current tags — and an I disagree link to propose a different set, exactly the same flow as archetype disputes. Accepted card-tag changes update the tags everywhere they're used, which means the heuristic re-classifies every deck containing that card on its next render.

Same anti-griefing rules apply: 5 disputes per user per 24h, one pending dispute per user×card pair, no public dispute log.

§7What this isn't (yet)

We're honest about the limits:

  • Until cards are individually tagged (one-time data work, in progress), the heuristic returns Unknown for most decks. The label you see is mostly from the manual override chain. The tag-dispute flow above is precisely how the corpus gets built — propose tags on a card you know well and they may get accepted.
  • Limited (Draft / Sealed) decks are out of scope for now. 40-card decks need their own signal set; they'll come in a later version.
  • Thresholds and weights are provisional. They'll be tuned against the dispute corpus once we have ~50 accepted records per archetype.

§8Read the full spec

Everything above is a summary. The full specification — every tag, every score component, every threshold, with worked examples — lives in our open repo:

docs/archetype-classification.md — versioned alongside the code, updated whenever weights move.

Found a tag-level or deck-level mistake? Get in touch, or just file a dispute on the deck in question.

More

Explore

Language preferences

Site language

Changes navigation and interface copy only.

Game-content language

Changes card rules and game terms only.

Your account

How we classify archetypes — SorceryRec