Broodscale Unholy Heat Matchup Winrates Six Weeks

Let me gather the necessary data first — I'll look up Broodscale and Unholy Heat to get their IDs.

Got both IDs. Now let me query the database for matchup winrates, filtering only Broodscale entries that include Unholy Heat in the mainboard, with at least 10 matches per matchup.

The results show all wins/losses/draws as 0, which indicates the match results aren't being captured correctly. Let me investigate the matches table structure to fix the query.

Results are WIN/LOSS/DRAW — let me fix the query accordingly.

The Unholy Heat filter is too restrictive — very few Broodscale entries in the DB have it, yielding almost no matchups. Let me check how many qualifying entries there actually are and whether Heat appears in SIDE too.

124 entries have Unholy Heat in the main — that's solid. The issue was the matches JOIN. Let me check the matches table more carefully to understand the linkage.

571 matches found — that's plenty. The issue was duplicate matches inflating from the deck_cards join. Let me fix with DISTINCT in the CTE and re-run the full matchup query.

Something's off — 571 matches exist but opponents aren't resolving correctly. Let me debug the opponent_entry_id join.

All 571 matches have valid opponent archetypes. The issue must be duplicate rows inflating counts from the deck_cards join. Let me use DISTINCT on match IDs properly.

Very strange — only 4 archetypes coming through. Let me see what opponents ARE actually matching without the HAVING filter.

The archetype IDs are there but aren't joining to names. The archetypes table might need the format_id check. Let me join properly.