diff --git a/docs/prd-evolution/05-rating-boosts.md b/docs/prd-evolution/05-rating-boosts.md
index 2af0e1d..d655302 100644
--- a/docs/prd-evolution/05-rating-boosts.md
+++ b/docs/prd-evolution/05-rating-boosts.md
@@ -122,7 +122,7 @@ proportionally so the 108-sum invariant is always preserved. Specifically:
 
 1. Negative deltas are applied first, each capped at the column's current value (0 floor).
 2. The total amount actually reduced is computed.
-3. Positive deltas are scaled by `actually_reduced / total_requested_reduction` so that
+3. Positive deltas are scaled by `actually_reduced / total_requested_addition` so that
    additions always equal reductions.
 4. A warning is logged when truncation occurs.
 
@@ -230,18 +230,19 @@ are currently always 0. The evolution system uses variant to store evolved versi
 variant number derived from a **deterministic hash** of all inputs that affect the card:
 
 ```python
-import hashlib
+import hashlib, json
 
-def compute_variant_hash(player_id: int, evolution_tier: int,
+def compute_variant_hash(player_id: int, refractor_tier: int,
                          cosmetics: list[str] | None) -> int:
-    """Compute a stable variant number from evolution + cosmetic state."""
+    """Compute a stable variant number from refractor + cosmetic state."""
     inputs = {
         "player_id": player_id,
-        "evolution_tier": evolution_tier,
+        "refractor_tier": refractor_tier,
         "cosmetics": sorted(cosmetics or []),
     }
-    raw = hashlib.sha256(str(inputs).encode()).hexdigest()
-    return int(raw[:8], 16)  # 32-bit unsigned integer from first 8 hex chars
+    raw = hashlib.sha256(json.dumps(inputs, sort_keys=True).encode()).hexdigest()
+    result = int(raw[:8], 16)  # 32-bit unsigned integer from first 8 hex chars
+    return result if result != 0 else 1  # variant=0 is reserved for base cards
 ```
 
 - `variant = 0`: Base card (standard, shared across all teams)