diff --git a/ai-engine/scripts/extract_training_data.py b/ai-engine/scripts/extract_training_data.py
index 4c92ca2..99819f5 100755
--- a/ai-engine/scripts/extract_training_data.py
+++ b/ai-engine/scripts/extract_training_data.py
@@ -424,12 +424,18 @@ class BatchDataLoader:
         for mid, tid, pid in self.cur.fetchall():
             starting_players[(mid, tid)].append(pid)
 
-        # 5) Build combined cache
+        # 5) Build match_id → mst_utc mapping for temporal filtering
+        match_mst = {}
+        for m in self.matches:
+            match_mst[m[0]] = m[7]  # m[0]=id, m[7]=mst_utc
+
+        # 6) Build combined cache — NO DATA LEAKAGE
+        # goals_form: avg goals from last 5 matches BEFORE this match (not this match!)
+        # squad_quality: only uses pre-match info (lineup, key players) — no current-match goals/assists
         all_keys = set(participation.keys()) | set(events.keys())
         for key in all_keys:
             mid, tid = key
             part = participation.get(key, {'starting_count': 0, 'total_squad': 0, 'fwd_count': 0})
-            evt = events.get(key, {'goals': 0, 'assists': 0, 'unique_scorers': 0})
 
             # Count key players in starting XI
             starters = starting_players.get(key, [])
@@ -437,22 +443,30 @@ class BatchDataLoader:
             kp_total = len(key_players_by_team.get(tid, set()))
             kp_missing = max(0, kp_total - kp_in_starting)
 
-            # Squad quality: composite score
+            # Squad quality: composite score — ONLY pre-match info (no current-match goals/assists!)
             squad_quality = (
                 part['starting_count'] * 0.3 +
-                evt['goals'] * 2.0 +
-                evt['assists'] * 1.0 +
                 kp_in_starting * 3.0 +
                 part['fwd_count'] * 1.5
             )
             # Missing impact: how many key players are missing
             missing_impact = min(kp_missing / max(kp_total, 1), 1.0)
 
+            # goals_form: avg goals from last 5 matches BEFORE this match
+            current_mst = match_mst.get(mid, 0)
+            team_history = self.team_matches.get(tid, [])
+            recent_goals = [
+                tm[2]  # team_score
+                for tm in team_history
+                if tm[0] < current_mst  # only matches BEFORE this one
+            ][-5:]  # last 5
+            goals_form = sum(recent_goals) / len(recent_goals) if recent_goals else 1.3
+
             self.squad_cache[key] = {
                 'squad_quality': squad_quality,
                 'key_players': kp_in_starting,
                 'missing_impact': missing_impact,
-                'goals_form': evt['goals'],
+                'goals_form': round(goals_form, 2),
             }
 
     def _load_cards_data(self):