diff --git a/src/ashare_dp/apps/api/dashboard/html.py b/src/ashare_dp/apps/api/dashboard/html.py
index df2cdbf..ca02d8b 100644
--- a/src/ashare_dp/apps/api/dashboard/html.py
+++ b/src/ashare_dp/apps/api/dashboard/html.py
@@ -1,6 +1,6 @@
 """Dashboard HTML — Trading Command Center.
 
-5 sections: COMMAND → ACTION → WHERE → WHY → RISK → EXPECTANCY
+5 sections: COMMAND → ACTION → WHERE → WHY → 风险 → EXPECTANCY
 Everything else collapsed as Diagnostics.
 """
 
@@ -50,7 +50,7 @@ body{font-family:-apple-system,BlinkMacSystemFont,'Segoe UI','PingFang SC','Micr
 .flow-line{font-family:monospace;font-size:10px;line-height:1.6;white-space:pre;color:var(--muted)}
 .flow-line .fr{color:var(--red)}.flow-line .to{color:var(--green)}.flow-line .ar{color:var(--muted)}
 
-/* RISK */
+/* 风险 */
 .risk-item{padding:6px 8px;border-radius:4px;margin-bottom:3px;font-size:10px;display:flex;align-items:flex-start;gap:5px}
 .risk-crit{background:rgba(248,81,73,.1);border:1px solid rgba(248,81,73,.2)}
 .risk-warn{background:rgba(210,153,29,.1);border:1px solid rgba(210,153,29,.2)}
@@ -83,6 +83,11 @@ body{font-family:-apple-system,BlinkMacSystemFont,'Segoe UI','PingFang SC','Micr
 <div id="app"><div class="loading"><div class="spinner"></div><p>Loading...</p></div></div>
 <script>
 var A='/api/v1/dashboard/state';
+var SIG_CN={orb_down:'向下突破',orb_up:'向上突破',chan_1buy:'缠论1买',chan_1sell:'缠论1卖',
+  vegas_long:'维加斯多',vegas_short:'维加斯空',ema52_cross_up:'EMA52上穿',ema52_cross_down:'EMA52下穿',
+  gap_up:'向上跳空',gap_down:'向下跳空',nr7:'窄幅7日',ib_long:'内含线多',ib_short:'内含线空'};
+function CN(s){return SIG_CN[s]||s;}
+
 function S(n){var s='';for(var i=0;i<5;i++)s+=i<n?'★':'☆';return s;}
 function P(v,d){return v!=null&&!isNaN(v)?v:(d||0);}
 function Q(v){return v==null?'--':v;}
@@ -143,9 +148,9 @@ function R(d){
     flowL+='<span class=\"'+(f.dir==='in'?'to':'fr')+'\">'+f.name+'</span> <span class=\"ar\">'+arrow+'</span> <span style=\"font-size:9px;color:var(--muted)\">'+(f.chg>0?'+':'')+(f.chg*100).toFixed(1)+'%</span>\\n';
   });
 
-  // RISK
+  // 风险
   var rh='<span style=\"font-size:10px;color:var(--green)\">✓ None</span>';
-  if(risks.length){rh='';risks.forEach(function(r){var l=r.level==='danger'?'risk-crit':'risk-warn';rh+='<div class=\"risk-item '+l+'\"><span class=\"risk-badge\">'+r.level.toUpperCase().substring(0,4)+'</span>'+r.message+'</div>';})}
+  if(risks.length){rh='';risks.forEach(function(r){var l=r.level==='danger'?'risk-crit':'risk-warn';rh+='<div class=\"risk-item '+l+'\"><span class=\"risk-badge\">'+'危'+'</span>'+r.message+'</div>';})}
 
   // Focus/Avoid
   var ft='',at='';
@@ -169,26 +174,26 @@ function R(d){
   // ═══ COMMAND + WHERE ═══
   h+='<div class=\"row\">';
   h+='<div class=\"card\" style=\"background:linear-gradient(135deg,#1a1f2e,#161b22)\">';
-  h+='<div class=\"sec-label\">COMMAND</div>';
+  h+='<div class=\"sec-label\">指令</div>';
   h+='<div class=\"cmd-mode\"><span class=\"'+mc+'\">'+S(ms)+'</span></div>';
   h+='<div style=\"font-size:13px;font-weight:700\">'+ml+'</div>';
   h+='<div class=\"cmd-play\">'+Q(pb.suitable_strategies[0]||pb.bias)+' · '+Q(pb.holding_time)+'</div>';
   h+='<div style=\"margin-bottom:4px\">'+acts.join(' ')+'</div>';
   if(avs.length)h+='<div style=\"margin-bottom:2px\">'+avs.join(' ')+'</div>';
   h+='<div class=\"cmd-meta\"><span>'+wy.join(' · ')+'</span></div>';
-  h+='<div class=\"cmd-meta\" style=\"border-top:none;padding-top:2px\"><span>Agr '+(conf*100).toFixed(0)+' &nbsp;Hlth '+(trend*100).toFixed(0)+' &nbsp;Fear '+(fear*100).toFixed(0)+' &nbsp;Liq '+(liq*100).toFixed(0)+' &nbsp;'+Q(pb.bias)+' &nbsp;C'+cashP+'/T'+trendP+'/T20</span></div>';
+  h+='<div class=\"cmd-meta\" style=\"border-top:none;padding-top:2px\"><span>共识 '+(conf*100).toFixed(0)+' &nbsp;健康 '+(trend*100).toFixed(0)+' &nbsp;恐惧 '+(fear*100).toFixed(0)+' &nbsp;流动 '+(liq*100).toFixed(0)+' &nbsp;'+Q(pb.bias)+' &nbsp;C'+cashP+'/T'+trendP+'/T20</span></div>';
   h+='</div>';
 
   h+='<div class=\"card\">';
-  h+='<div class=\"sec-label\">WHERE</div>';
+  h+='<div class=\"sec-label\">方向</div>';
   h+=wh||'<span class=\"muted\" style=\"font-size:10px\">--</span>';
   h+='</div></div>';
 
-  // ═══ WHY + RISK ═══
+  // ═══ WHY + 风险 ═══
   h+='<div class=\"row\">';
   h+='<div class=\"card\">';
-  h+='<div class=\"sec-label\">WHY — Money Flow</div>';
-  h+='<div class=\"flow-line\">'+(flowL||'No flow data')+'</div>';
+  h+='<div class=\"sec-label\">资金流向</div>';
+  h+='<div class=\"flow-line\">'+(flowL||'无资金流数据')+'</div>';
   if(ft)h+='<div style=\"margin-top:6px;font-size:9px\"><span class=\"muted\">Focus:</span> '+ft+'</div>';
   if(at)h+='<div style=\"font-size:9px\"><span class=\"muted\">Avoid:</span> '+at+'</div>';
   h+='</div>';
@@ -196,7 +201,7 @@ function R(d){
   // Theme Map
   var tg=d.theme_graph||{},themes=tg.themes||[];
   h+='<div class=\"card\">';
-  h+='<div class=\"sec-label\">THEMES</div>';
+  h+='<div class=\"sec-label\">主题</div>';
   if(themes.length){
     themes.slice(0,5).forEach(function(t,i){
       var stars='';for(var j=0;j<5;j++)stars+=j<Math.round(t.score/20)?'★':'☆';
@@ -219,58 +224,65 @@ function R(d){
   // Recommendations
   var recs=d.recommendations||[];
   h+='<div class=\"card\" style=\"margin-bottom:8px\">';
-  h+='<div class=\"sec-label\">RECOMMENDATIONS — Money Flow Picks</div>';
+  h+='<div class=\"sec-label\">RECOMMENDATIONS — 资金轮动选股</div>';
   if(recs.length){
     h+='<div style=\"display:grid;grid-template-columns:repeat(auto-fill,minmax(300px,1fr));gap:6px\">';
     recs.forEach(function(r,i){
-      var cls=r.score>70?'good':r.score>50?'warn':'muted';
+      var sc=r.score||0;var cls=sc>70?'good':sc>50?'warn':'muted';
       h+='<div style=\"padding:8px 10px;background:rgba(255,255,255,.02);border-radius:6px;font-size:11px\">';
       h+='<div style=\"display:flex;justify-content:space-between;align-items:center;margin-bottom:4px\">';
       h+='<span><strong>'+(r.name||r.ts_code)+'</strong> <span style=\"font-size:10px;color:var(--muted)\">'+r.ts_code+' · '+r.sector+'</span></span>';
-      h+='<span class=\"'+cls+'\" style=\"font-weight:700\">'+r.score.toFixed(0)+'</span></div>';
+      h+='<span class=\"'+cls+'\" style=\"font-weight:700\">'+sc.toFixed(0)+'</span></div>';
       h+='<div style=\"display:flex;gap:12px;font-size:10px;color:var(--muted);margin-bottom:4px\">';
-      h+='<span>Entry: <span class=\"good\">'+r.entry+'</span></span>';
-      h+='<span>Stop: <span class=\"bad\">'+r.stop+'</span></span>';
-      h+='<span>T1: <span class=\"good\">'+r.target1+'</span></span>';
+      h+='<span>入场: <span class=\"good\">'+r.entry+'</span></span>';
+      h+='<span>止损: <span class=\"bad\">'+r.stop+'</span></span>';
+      h+='<span>目标1: <span class=\"good\">'+r.target1+'</span></span>';
       h+='<span>T2: '+r.target2+'</span></div>';
       h+='<div style=\"display:flex;justify-content:space-between;font-size:10px\">';
-      h+='<span class=\"muted\">5d:'+(r.metrics.ret_5d>0?'+':'')+r.metrics.ret_5d.toFixed(1)+'% Vol:'+r.metrics.vol_ratio.toFixed(1)+'x '+r.metrics.trend+'</span>';
-      h+='<span style=\"color:var(--accent)\">R:R '+r.rr_ratio.toFixed(1)+'</span></div>';
-      h+='<div style=\"font-size:10px;color:var(--green);margin-top:3px\">'+r.action+'</div>';
+      var ret5=r.ret_5d||0,volR=r.vol_ratio||1,trS=r.trend_score||0,amtY=r.amount_yi||0;
+      h+='<span class=\"muted\">5日:'+(ret5>0?'+':'')+ret5.toFixed(1)+'% 量比:'+volR.toFixed(1)+'x 成交额:'+amtY.toFixed(1)+'亿 '+(trS>=2?'强势':trS>=1?'中性':'弱势')+'</span>';
+      h+='<span style=\"color:var(--accent)\">盈亏比 '+(r.rr_ratio||0).toFixed(1)+'</span></div>';
+      h+='<div style=\"font-size:10px;color:var(--green);margin-top:3px\">'+r.action+(r.outcome?' <span style=\"color:'+(r.outcome==='win'?'var(--green)':'var(--red)')+'\">['+r.outcome.toUpperCase()+']</span>':'')+'</div>';
       h+='</div>';
     });
     h+='</div>';
   }else{h+='<span class=\"muted\" style=\"font-size:11px\">No recommendations — insufficient data</span>';}
   h+='</div>';
 
+  // Verify card (loaded async)
+  h+='<div class=\"card\" style=\"margin-bottom:8px\" id=\"verify-card\"><span class=\"muted\" style=\"font-size:10px\">Loading performance stats...</span></div>';
+
   h+='<div class=\"row\"><div class=\"card\">';
-  h+='<div class=\"sec-label\">RISK</div>';
+  h+='<div class=\"sec-label\">风险</div>';
+  h+=rh;
+  h+='<div class=\"row\"><div class=\"card\">';
+  h+='<div class=\"sec-label\">风险</div>';
   h+=rh;
   h+='</div></div>';
 
   // ═══ EXPECTANCY ═══
   h+='<div class=\"card\" style=\"opacity:0.8\">';
-  var expLabel=samples>0?'— '+sigType+' · '+samples+' samples':'— Phase 1 Heuristic · 0 samples';
+  var expLabel=samples>0?'— '+CN(sigType)+' · '+samples+' 样本':'— 阶段1 启发式 · 0样本';
   var expColor=samples>0?'var(--green)':'var(--orange)';
   h+='<div class=\"sec-label\">EXPECTANCY <span style=\"color:'+expColor+';font-weight:400\">'+expLabel+'</span></div>';
   h+='<div class=\"exp-grid\">';
-  h+='<div class=\"exp-stat\"><div class=\"val '+(wr>50?'good':'warn')+'\">'+wr.toFixed(0)+'%</div><div class=\"lbl\">Est. WR</div></div>';
-  h+='<div class=\"exp-stat\"><div class=\"val '+(ar>0?'good':'bad')+'\">'+(ar>0?'+':'')+ar.toFixed(1)+'%</div><div class=\"lbl\">Est. Ret</div></div>';
-  h+='<div class=\"exp-stat\"><div class=\"val bad\">'+dd.toFixed(1)+'%</div><div class=\"lbl\">Est. DD</div></div>';
-  h+='<div class=\"exp-stat\"><div class=\"val\">'+hold.toFixed(1)+'d</div><div class=\"lbl\">Est. Hold</div></div>';
+  h+='<div class=\"exp-stat\"><div class=\"val '+(wr>50?'good':'warn')+'\">'+wr.toFixed(0)+'%</div><div class=\"lbl\">预估胜率</div></div>';
+  h+='<div class=\"exp-stat\"><div class=\"val '+(ar>0?'good':'bad')+'\">'+(ar>0?'+':'')+ar.toFixed(1)+'%</div><div class=\"lbl\">预估收益</div></div>';
+  h+='<div class=\"exp-stat\"><div class=\"val bad\">'+dd.toFixed(1)+'%</div><div class=\"lbl\">预估回撤</div></div>';
+  h+='<div class=\"exp-stat\"><div class=\"val\">'+hold.toFixed(1)+'d</div><div class=\"lbl\">预估持仓</div></div>';
   h+='</div>';
 
   // All expectancies comparison table
   var allExp=d.all_expectancies||[];
   if(allExp.length>0){
-    h+='<table class=\"exp-table\"><tr><th>Signal</th><th>WR</th><th>Samples</th><th>Avg Ret</th><th>Max DD</th></tr>';
+    h+='<table class=\"exp-table\"><tr><th>信号</th><th>WR</th><th>样本</th><th>平均收益</th><th>最大回撤</th></tr>';
     allExp.forEach(function(e){
       var ew=e.win_rate*100,ea=e.avg_return*100,ed=e.max_drawdown*100;
       var barW=Math.max(2,ew);
       h+='<tr>'+
-        '<td>'+e.signal_type+'</td>'+
+        '<td>'+CN(e.signal_type)+'</td>'+
         '<td><span class=\"exp-mini-bar\" style=\"width:'+barW+'px;background:'+(ew>50?'var(--green)':ew>40?'var(--orange)':'var(--red)')+'\"></span>'+(e.sample_count>0?ew.toFixed(0)+'%':'--')+'</td>'+
-        '<td style=\"color:var(--muted)\">'+(e.sample_count>0?e.sample_count.toLocaleString():'heuristic')+'</td>'+
+        '<td style=\"color:var(--muted)\">'+(e.sample_count>0?e.sample_count.toLocaleString():'启发式')+'</td>'+
         '<td class=\"'+(ea>0?'good':'bad')+'\">'+(e.sample_count>0?(ea>0?'+':'')+ea.toFixed(1)+'%':'--')+'</td>'+
         '<td class=\"bad\">'+(e.sample_count>0?ed.toFixed(1)+'%':'--')+'</td>'+
       '</tr>';
@@ -282,20 +294,20 @@ function R(d){
   // ═══ DIAGNOSTICS (always collapsed) ═══
   var idx=(tr.index_details||[]).slice(0,7);
   var idxH='';idx.forEach(function(i){idxH+='<div class=\"metric\"><span>'+i.name+'</span><span class=\"'+(i.pct_chg>0?'good':'bad')+'\">'+(i.pct_chg>0?'+':'')+i.pct_chg.toFixed(2)+'%</span></div>';});
-  h+='<div class=\"diag-toggle\" onclick=\"var e=document.getElementById(\\'diag\\');e.classList.toggle(\\'open\\');this.textContent=e.classList.contains(\\'open\\')?\\'▼ Diagnostics\\':\\'▶ Diagnostics\\'\">▶ Diagnostics</div>';
+  h+='<div class=\"diag-toggle\" onclick=\"var e=document.getElementById(\\'diag\\');e.classList.toggle(\\'open\\');this.textContent=e.classList.contains(\\'open\\')?\\'▼ 诊断\\':\\'▶ 诊断\\'\">▶ 诊断</div>';
   h+='<div class=\"diag-content\" id=\"diag\"><div class=\"diag-grid\">';
-  h+='<div class=\"diag-card\"><h4>Breadth</h4><div class=\"metric\"><span>Adv/Dec</span><span><span class=\"good\">'+Q(b.advances)+'</span>/<span class=\"bad\">'+Q(b.declines)+'</span></span></div><div class=\"metric\"><span>Ratio</span><span>'+((b.advance_ratio||0)*100).toFixed(1)+'%</span></div><div class=\"metric\"><span>New High</span><span>'+((b.new_high_ratio||0)*100).toFixed(1)+'%</span></div></div>';
-  h+='<div class=\"diag-card\"><h4>Volume</h4><div class=\"metric\"><span>Turnover</span><span>'+Q(v.total_turnover_yi)+' 亿</span></div><div class=\"metric\"><span>vs 5d</span><span>'+((v.vs_5d-1)*100).toFixed(1)+'%</span></div><div class=\"metric\"><span>Tier</span><span>'+Q(v.tier)+'</span></div></div>';
-  h+='<div class=\"diag-card\"><h4>State</h4>';['trend','fear','liquidity','rotation','participation','volatility','breadth'].forEach(function(k){h+='<div class=\"metric\"><span>'+k+'</span><span>'+((dims[k]||0)*100).toFixed(0)+'</span></div>';});h+='</div>';
-  h+='<div class=\"diag-card\"><h4>Indices</h4>'+idxH+'<div style=\"font-size:8px;color:var(--muted);margin-top:2px\">Resonance: '+((tr.resonance||0)*100).toFixed(0)+'%</div></div>';
+  h+='<div class=\"diag-card\"><h4>市场宽度</h4><div class=\"metric\"><span>Adv/Dec</span><span><span class=\"good\">'+Q(b.advances)+'</span>/<span class=\"bad\">'+Q(b.declines)+'</span></span></div><div class=\"metric\"><span>Ratio</span><span>'+((b.advance_ratio||0)*100).toFixed(1)+'%</span></div><div class=\"metric\"><span>New High</span><span>'+((b.new_high_ratio||0)*100).toFixed(1)+'%</span></div></div>';
+  h+='<div class=\"diag-card\"><h4>成交量</h4><div class=\"metric\"><span>Turnover</span><span>'+Q(v.total_turnover_yi)+' 亿</span></div><div class=\"metric\"><span>vs 5d</span><span>'+((v.vs_5d-1)*100).toFixed(1)+'%</span></div><div class=\"metric\"><span>Tier</span><span>'+Q(v.tier)+'</span></div></div>';
+  h+='<div class=\"diag-card\"><h4>状态</h4>';['trend','fear','liquidity','rotation','participation','volatility','breadth'].forEach(function(k){h+='<div class=\"metric\"><span>'+k+'</span><span>'+((dims[k]||0)*100).toFixed(0)+'</span></div>';});h+='</div>';
+  h+='<div class=\"diag-card\"><h4>指数</h4>'+idxH+'<div style=\"font-size:8px;color:var(--muted);margin-top:2px\">共振度: '+((tr.resonance||0)*100).toFixed(0)+'%</div></div>';
   var sent=ev.sentiment||{};
   if(sent.status==='live'){
-    h+='<div class=\"diag-card\"><h4>Sentiment</h4>';
-    h+='<div class=\"metric\"><span>Profit Effect</span><span class=\"'+(sent.profit_effect==='Strong'||sent.profit_effect==='Average'?'good':'bad')+'\">'+Q(sent.profit_effect)+' ('+Q(sent.profit_effect_score)+')</span></div>';
-    h+='<div class=\"metric\"><span>Limit Up/Down</span><span><span class=\"good\">'+Q(sent.limit_up_count)+'</span>/<span class=\"bad\">'+Q(sent.limit_down_count)+'</span></span></div>';
-    h+='<div class=\"metric\"><span>Broken Board</span><span>'+Q(sent.broken_board_count)+' ('+((sent.broken_board_rate||0)*100).toFixed(0)+'%)</span></div>';
-    h+='<div class=\"metric\"><span>Consecutive</span><span>'+Q(sent.consecutive_count)+' ('+((sent.consecutive_board_rate||0)*100).toFixed(0)+'%)</span></div>';
-    if(sent.max_consecutive)h+='<div class=\"metric\"><span>Max Board</span><span>'+sent.max_consecutive+' 连板</span></div>';
+    h+='<div class=\"diag-card\"><h4>情绪</h4>';
+    h+='<div class=\"metric\"><span>赚钱效应</span><span class=\"'+(sent.profit_effect==='强势'||sent.profit_effect==='一般'?'good':'bad')+'\">'+Q(sent.profit_effect)+' ('+Q(sent.profit_effect_score)+')</span></div>';
+    h+='<div class=\"metric\"><span>涨停/跌停</span><span><span class=\"good\">'+Q(sent.limit_up_count)+'</span>/<span class=\"bad\">'+Q(sent.limit_down_count)+'</span></span></div>';
+    h+='<div class=\"metric\"><span>炸板</span><span>'+Q(sent.broken_board_count)+' ('+((sent.broken_board_rate||0)*100).toFixed(0)+'%)</span></div>';
+    h+='<div class=\"metric\"><span>连板</span><span>'+Q(sent.consecutive_count)+' ('+((sent.consecutive_board_rate||0)*100).toFixed(0)+'%)</span></div>';
+    if(sent.max_consecutive)h+='<div class=\"metric\"><span>最高连板</span><span>'+sent.max_consecutive+' 连板</span></div>';
     h+='</div>';
   }
   h+='</div></div>';
diff --git a/src/ashare_dp/apps/api/dashboard/router.py b/src/ashare_dp/apps/api/dashboard/router.py
index 09bde6f..cd707a4 100644
--- a/src/ashare_dp/apps/api/dashboard/router.py
+++ b/src/ashare_dp/apps/api/dashboard/router.py
@@ -23,7 +23,6 @@ from ashare_dp.market.leadership import assess_leaders
 from ashare_dp.market.memory import state_store
 from ashare_dp.market.opportunity import rank_opportunities
 from ashare_dp.market.knowledge import activate_themes
-from ashare_dp.market.recommendations import get_recommendations
 from ashare_dp.market.sentiment import assess_sentiment
 from ashare_dp.market.state import infer_market_state
 from ashare_dp.signals.expectancy import get_expectancy, get_all_expectancies
@@ -79,11 +78,26 @@ async def dashboard_state():
     # Knowledge Graph — theme activation
     theme_graph = activate_themes(trade_date, leaders=leaders, opportunities=opportunities)
 
-    # Stock Recommendations
+    # Stock Recommendations — read from EOD-computed recommendation_log
     try:
-        recommendations = get_recommendations(trade_date, top_n=10)
-    except Exception as e:
-        logger.warning(f"Recommendations failed: {e}")
+        with get_db(read_only=True) as db:
+            rows = db.query(
+                "SELECT ts_code, name, sector, entry_price, stop_price, target1, target2, "
+                "score, action, outcome, ret_5d, vol_ratio, trend_score, amount_yi "
+                "FROM recommendation_log WHERE trade_date = ? ORDER BY score DESC",
+                (trade_date.isoformat(),),
+            )
+        def _rr(entry, stop, t1):
+            return round((t1 - entry) / (entry - stop), 1) if entry and stop and entry > stop else 0.0
+        recommendations = [
+            {"ts_code": r[0], "name": r[1], "sector": r[2],
+             "entry": r[3], "stop": r[4], "target1": r[5], "target2": r[6],
+             "score": r[7], "action": r[8], "outcome": r[9],
+             "ret_5d": r[10], "vol_ratio": r[11], "trend_score": r[12], "amount_yi": r[13],
+             "rr_ratio": _rr(r[3], r[4], r[5])}
+            for r in rows
+        ] if rows else []
+    except Exception:
         recommendations = []
 
     # ── L3: Trading Intelligence ──
@@ -197,6 +211,123 @@ async def close_trade(trade_id: int, request: dict):
     return {"status": "closed"}
 
 
+def _save_recommendations(recs: list, trade_date_str: str = ""):
+    """Auto-save recommendations to tracking log."""
+    if not recs:
+        return
+    if not trade_date_str:
+        trade_date_str = date.today().isoformat()
+    from ashare_dp.data.store.database import get_db
+    with get_db(read_only=False) as db:
+        for r in recs:
+            try:
+                db.execute(
+                    """INSERT INTO recommendation_log
+                       (trade_date, ts_code, name, sector, entry_price, stop_price,
+                        target1, target2, score, action)
+                       VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?)
+                       ON CONFLICT (trade_date, ts_code) DO NOTHING""",
+                    (trade_date_str, r["ts_code"], r.get("name", ""),
+                     r["sector"], r["entry"], r["stop"], r["target1"], r["target2"],
+                     r["score"], r.get("action", "")),
+                )
+            except Exception:
+                pass
+
+
+@router.get("/dashboard/verify")
+async def verify_recommendations(days: int = 30):
+    """Verify past recommendations against actual price data. Returns performance stats."""
+    from datetime import date as _date, timedelta
+    end = _date.today()
+    start = end - timedelta(days=days)
+
+    with get_db(read_only=True) as db:
+        rows = db.query(
+            "SELECT trade_date, ts_code, entry_price, stop_price, target1, target2, outcome "
+            "FROM recommendation_log WHERE trade_date >= ? AND trade_date <= ? ORDER BY trade_date DESC",
+            (start.isoformat(), end.isoformat()),
+        )
+    logger.info(f"Verify: found {len(rows)} recommendations in range {start}~{end}")
+
+    verified = []
+    wins = 0
+    losses = 0
+    pending = 0
+
+    conn = analytics_conn()
+    for r in rows:
+        td, ts, entry, stop, t1, t2, outcome = r
+        if outcome and outcome != "pending":
+            if outcome == "win": wins += 1
+            elif outcome == "loss": losses += 1
+            continue
+
+        # Verify: find first close that hits target or stop (realistic, not intraday extremes)
+        try:
+            future = conn.execute(
+                f"SELECT trade_date, close FROM read_parquet('{kline_glob()}', hive_partitioning=true, union_by_name=true) "
+                f"WHERE ts_code = ? AND trade_date > ? ORDER BY trade_date ASC",
+                [ts, str(td)],
+            ).fetchall()
+
+            result = "pending"
+            actual_return = None
+            max_reached = float(entry)
+            min_reached = float(entry)
+
+            if future:
+                for fd, fc in future:
+                    fc = float(fc)
+                    max_reached = max(max_reached, fc)
+                    min_reached = min(min_reached, fc)
+
+                    if fc >= t1:
+                        result = "win"
+                        actual_return = round((t1 - entry) / entry * 100, 2)
+                        break
+                    elif fc <= stop:
+                        result = "loss"
+                        actual_return = round((stop - entry) / entry * 100, 2)
+                        break
+
+            if result == "pending":
+                pending += 1
+            elif result == "win":
+                wins += 1
+            else:
+                losses += 1
+
+            if result != "pending":
+                with get_db(read_only=False) as wdb:
+                    wdb.execute(
+                        "UPDATE recommendation_log SET outcome=?, actual_return=?, verified=TRUE WHERE trade_date=? AND ts_code=?",
+                        (result, actual_return, str(td), ts),
+                    )
+
+            verified.append({
+                "trade_date": str(td), "ts_code": ts,
+                "entry": entry, "stop": stop, "target1": t1,
+                "max_reached": round(max_reached, 2),
+                "min_reached": round(min_reached, 2),
+                "outcome": result,
+            })
+        except Exception:
+            pending += 1
+
+    conn.close()
+
+    total = wins + losses
+    win_rate = round(wins / total * 100, 1) if total > 0 else 0
+    return {
+        "period_days": days,
+        "total_verified": len(verified),
+        "wins": wins, "losses": losses, "pending": pending,
+        "win_rate": win_rate,
+        "verified": verified[:20],
+    }
+
+
 @router.get("/dashboard/replay")
 async def dashboard_replay(date_str: str = ""):
     """Return all intraday state snapshots for a given date (YYYY-MM-DD)."""
diff --git a/src/ashare_dp/apps/scheduler/jobs.py b/src/ashare_dp/apps/scheduler/jobs.py
index 5895273..6b7bdb2 100644
--- a/src/ashare_dp/apps/scheduler/jobs.py
+++ b/src/ashare_dp/apps/scheduler/jobs.py
@@ -80,3 +80,49 @@ async def wyckoff_scan_job():
         )
     except Exception as e:
         logger.error(f"Wyckoff scan job failed: {e}")
+
+
+async def recommendation_job():
+    """EOD recommendation computation — compute and store stock picks."""
+    from datetime import date as _date
+    from ashare_dp.market.recommendations import get_recommendations
+    from ashare_dp.data.store.database import get_db
+
+    today = _date.today()
+    logger.info(f"Computing daily recommendations for {today}...")
+
+    try:
+        recs = get_recommendations(today, top_n=10)
+    except Exception as e:
+        logger.error(f"Recommendation computation failed: {e}")
+        return
+
+    if not recs:
+        logger.warning("No recommendations generated")
+        return
+
+    trade_date_str = today.isoformat()
+    with get_db(read_only=False) as db:
+        for r in recs:
+            try:
+                m = r.get("metrics", {})
+                trend_str = str(m.get("trend", "Weak"))
+                trend_val = 2.0 if trend_str == "Strong" else (1.0 if trend_str == "Neutral" else 0.0)
+                db.execute(
+                    """INSERT INTO recommendation_log
+                       (trade_date, ts_code, name, sector, entry_price, stop_price,
+                        target1, target2, score, action, ret_5d, vol_ratio, trend_score, amount_yi)
+                       VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)
+                       ON CONFLICT (trade_date, ts_code) DO UPDATE SET
+                       score=excluded.score, action=excluded.action,
+                       ret_5d=excluded.ret_5d, vol_ratio=excluded.vol_ratio,
+                       trend_score=excluded.trend_score, amount_yi=excluded.amount_yi""",
+                    (trade_date_str, r["ts_code"], r.get("name", ""),
+                     r["sector"], r["entry"], r["stop"], r["target1"], r["target2"],
+                     r["score"], r.get("action", ""),
+                     m.get("ret_5d"), m.get("vol_ratio"), trend_val, m.get("amount_yi")),
+                )
+            except Exception:
+                pass
+
+    logger.info(f"Recommendations stored: {len(recs)} stocks")
diff --git a/src/ashare_dp/apps/scheduler/scheduler.py b/src/ashare_dp/apps/scheduler/scheduler.py
index 928d5db..d454eea 100644
--- a/src/ashare_dp/apps/scheduler/scheduler.py
+++ b/src/ashare_dp/apps/scheduler/scheduler.py
@@ -23,6 +23,7 @@ class Scheduler:
             eod_pull_job,
             ema52_screening_job,
             health_check_job,
+            recommendation_job,
             wyckoff_scan_job,
         )
 
@@ -81,10 +82,24 @@ class Scheduler:
             replace_existing=True,
         )
 
+        # Recommendations: 15:20 Beijing time, Mon-Fri (after Wyckoff at 15:15)
+        self._scheduler.add_job(
+            recommendation_job,
+            trigger=CronTrigger(
+                day_of_week="mon-fri",
+                hour=15,
+                minute=20,
+                timezone=BEIJING_TZ,
+            ),
+            id="recommendations",
+            name="Daily stock recommendations",
+            replace_existing=True,
+        )
+
         self._scheduler.start()
         logger.info(
             "Scheduler started with EOD (15:05) + EMA52 (15:10) + "
-            "Wyckoff (15:15) + health check (08:00)"
+            "Wyckoff (15:15) + Recs (15:20) + health check (08:00)"
         )
 
     def shutdown(self):
diff --git a/src/ashare_dp/data/store/schema.py b/src/ashare_dp/data/store/schema.py
index 72d872f..09a681c 100644
--- a/src/ashare_dp/data/store/schema.py
+++ b/src/ashare_dp/data/store/schema.py
@@ -180,4 +180,22 @@ DDL_STATEMENTS = [
     """CREATE INDEX IF NOT EXISTS idx_wyckoff_score ON wyckoff_scan(trade_date, overall_score DESC)""",
     """CREATE INDEX IF NOT EXISTS idx_wyckoff_align ON wyckoff_scan(trade_date, alignment DESC)""",
     """CREATE INDEX IF NOT EXISTS idx_wyckoff_version ON wyckoff_scan(trade_date, engine_version)""",
+    # ── Recommendation Tracking ──
+    """CREATE SEQUENCE IF NOT EXISTS seq_rec_id""",
+    """
+    CREATE TABLE IF NOT EXISTS recommendation_log (
+        id BIGINT PRIMARY KEY DEFAULT nextval('seq_rec_id'),
+        trade_date DATE NOT NULL, ts_code VARCHAR(15) NOT NULL,
+        name VARCHAR(40), sector VARCHAR(40),
+        entry_price DOUBLE NOT NULL, stop_price DOUBLE NOT NULL,
+        target1 DOUBLE NOT NULL, target2 DOUBLE NOT NULL,
+        score DOUBLE, action VARCHAR(100),
+        ret_5d DOUBLE, vol_ratio DOUBLE, trend_score DOUBLE, amount_yi DOUBLE,
+        outcome VARCHAR(20), actual_return DOUBLE,
+        verified BOOLEAN DEFAULT FALSE,
+        created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
+        UNIQUE(trade_date, ts_code)
+    )
+    """,
+    """CREATE INDEX IF NOT EXISTS idx_rec_date ON recommendation_log(trade_date)""",
 ]
diff --git a/src/ashare_dp/market/optimizer.py b/src/ashare_dp/market/optimizer.py
new file mode 100644
index 0000000..6d9c90e
--- /dev/null
+++ b/src/ashare_dp/market/optimizer.py
@@ -0,0 +1,222 @@
+"""Parameter Optimizer — grid search for optimal recommendation parameters.
+
+Tests weight combinations × stop multipliers × target multipliers
+against historical recommendation outcomes to find the best config.
+"""
+
+from __future__ import annotations
+
+import itertools
+from datetime import date, datetime, timedelta
+
+import duckdb
+from loguru import logger
+
+
+def optimize_parameters(
+    trade_date: date | None = None,
+    lookback_days: int = 60,
+) -> list[dict]:
+    """Grid search for optimal recommendation parameters.
+
+    Tests different weight combos, stop ATR multiples, and target ATR multiples.
+    For each combo, generates recommendations for past N trading days
+    and checks forward outcomes.
+
+    Returns ranked list of parameter sets with performance stats.
+    """
+    # ═══ Parameter grid ═══
+    weight_combos = [
+        (50, 30, 20, "50/30/20"),
+        (40, 30, 30, "40/30/30"),
+        (40, 40, 20, "40/40/20"),
+        (50, 25, 25, "50/25/25"),
+        (30, 40, 30, "30/40/30"),
+        (30, 30, 40, "30/30/40"),
+    ]
+    stop_mults = [1.0, 1.5, 2.0, 2.5, 3.0]
+    target_mults = [1.0, 1.5, 2.0, 2.5, 3.0]
+
+    # ═══ Get trading days ═══
+    if trade_date is None:
+        trade_date = date.today()
+
+    conn = duckdb.connect("data/duckdb/ashare.db", read_only=True)
+    cal_dates = conn.execute(
+        "SELECT trade_date FROM trading_calendar WHERE trade_date <= ? "
+        "AND is_trading_day=true ORDER BY trade_date DESC LIMIT ?",
+        [trade_date.isoformat(), lookback_days],
+    ).fetchall()
+    conn.close()
+
+    if len(cal_dates) < 10:
+        logger.warning(f"Only {len(cal_dates)} trading days available")
+        return []
+
+    test_dates = [date.fromisoformat(str(r[0])[:10]) for r in cal_dates[5:]]  # skip most recent 5 (outcome unknown)
+    logger.info(f"Testing parameters over {len(test_dates)} trading days")
+
+    # ═══ For each test date, get top stocks by raw scores ═══
+    # Use a simpler approach: pre-compute stock scores for all test dates
+    stock_scores = _compute_historical_scores(test_dates)
+
+    if not stock_scores:
+        return []
+
+    logger.info(f"Scored {len(stock_scores)} stock-date combinations")
+
+    # ═══ Grid search ═══
+    results = []
+    total_combos = len(weight_combos) * len(stop_mults) * len(target_mults)
+
+    for i, (w_mom, w_vol, w_trd, w_label) in enumerate(weight_combos):
+        # Re-rank stocks with these weights
+        ranked = _rerank_stocks(stock_scores, w_mom, w_vol, w_trd)
+
+        for stop_m in stop_mults:
+            for tgt_m in target_mults:
+                # Test this parameter set
+                stats = _test_parameters(ranked, stop_m, tgt_m)
+                if stats["total"] >= 10:
+                    stats["weights"] = w_label
+                    stats["stop_atr"] = stop_m
+                    stats["target_atr"] = tgt_m
+                    stats["score"] = round(stats["win_rate"] * 0.6 + stats["avg_return"] * 100 * 0.4, 2)
+                    results.append(stats)
+
+                if (len(results) + 1) % 20 == 0:
+                    logger.info(f"  Tested {len(results)}/{total_combos} combos...")
+
+    # Sort by composite score
+    results.sort(key=lambda r: r["score"], reverse=True)
+    return results
+
+
+def _compute_historical_scores(test_dates: list[date]) -> dict:
+    """Pre-compute raw stock scores (momentum/vol/trend) for each test date."""
+    conn = duckdb.connect("data/duckdb/ashare.db", read_only=True)
+    scores = {}
+
+    for td in test_dates:
+        try:
+            sql = f"""
+                WITH normalized AS (
+                    SELECT ts_code, trade_date, trade_time, close, volume, amount, high, low
+                    FROM read_parquet('data/parquet/kline_1d/**/*.parquet', hive_partitioning=true, union_by_name=true)
+                    WHERE trade_date <= $td
+                ),
+                with_ma AS (
+                    SELECT *,
+                        AVG(close) OVER (PARTITION BY ts_code ORDER BY trade_time ROWS BETWEEN 19 PRECEDING AND CURRENT ROW) AS ma20,
+                        AVG(close) OVER (PARTITION BY ts_code ORDER BY trade_time ROWS BETWEEN 59 PRECEDING AND CURRENT ROW) AS ma60,
+                        AVG(volume) OVER (PARTITION BY ts_code ORDER BY trade_time ROWS BETWEEN 19 PRECEDING AND CURRENT ROW) AS vol_ma20,
+                        (high - low) AS day_range,
+                        ROW_NUMBER() OVER (PARTITION BY ts_code ORDER BY trade_time DESC) AS rn
+                    FROM normalized
+                ),
+                with_atr AS (
+                    SELECT *, AVG(day_range) OVER (PARTITION BY ts_code ORDER BY trade_time ROWS BETWEEN 19 PRECEDING AND CURRENT ROW) AS atr20
+                    FROM with_ma
+                ),
+                latest AS (
+                    SELECT * FROM with_atr WHERE rn = 1
+                ),
+                prev AS (
+                    SELECT ts_code, close AS close_prev FROM with_ma WHERE rn = 6
+                )
+                SELECT l.ts_code, l.close, l.atr20, l.amount,
+                    (l.close - p.close_prev) / NULLIF(p.close_prev, 0) AS ret_5d,
+                    l.volume / NULLIF(l.vol_ma20, 0) AS vol_ratio,
+                    CASE WHEN l.close > l.ma20 THEN 1 ELSE 0 END + CASE WHEN l.close > l.ma60 THEN 1 ELSE 0 END AS trend_score
+                FROM latest l LEFT JOIN prev p ON l.ts_code = p.ts_code
+                WHERE l.amount > 2e8
+            """
+            df = conn.execute(sql, {"td": td.isoformat()}).fetchdf()
+            if not df.empty:
+                scores[td.isoformat()] = df
+        except Exception:
+            pass
+
+    conn.close()
+    return scores
+
+
+def _rerank_stocks(scores: dict, w_mom: float, w_vol: float, w_trd: float) -> dict:
+    """Re-rank stocks with given weights."""
+    w_sum = w_mom + w_vol + w_trd
+    ranked = {}
+
+    for date_str, df in scores.items():
+        df = df.copy()
+        # Normalize each factor to 0-1 range
+        for col in ["ret_5d", "vol_ratio", "trend_score"]:
+            vals = df[col].dropna()
+            if len(vals) > 0 and vals.std() > 0:
+                df[col + "_norm"] = (df[col] - vals.min()) / (vals.max() - vals.min())
+            else:
+                df[col + "_norm"] = 0.5
+
+        df["composite"] = (
+            df["ret_5d_norm"].fillna(0) * w_mom / w_sum * 100
+            + df["vol_ratio_norm"].fillna(0) * w_vol / w_sum * 100
+            + df["trend_score_norm"].fillna(0) * w_trd / w_sum * 100
+        )
+        df = df.sort_values("composite", ascending=False)
+        ranked[date_str] = df.head(10)  # top 10 per day
+
+    return ranked
+
+
+def _test_parameters(ranked: dict, stop_atr: float, target_atr: float) -> dict:
+    """Test a parameter set against historical outcomes."""
+    conn = duckdb.connect("data/duckdb/ashare.db", read_only=True)
+    wins = 0
+    losses = 0
+    pending = 0
+    returns = []
+
+    for date_str, df in ranked.items():
+        for _, row in df.iterrows():
+            entry = float(row["close"])
+            atr = float(row.get("atr20", entry * 0.03) or entry * 0.03)
+            ts = row["ts_code"]
+
+            stop = entry - stop_atr * atr
+            target = entry + target_atr * atr
+
+            # Check forward outcomes
+            try:
+                future = conn.execute(
+                    "SELECT close FROM read_parquet('data/parquet/kline_1d/**/*.parquet', "
+                    "hive_partitioning=true, union_by_name=true) "
+                    "WHERE ts_code = ? AND trade_date > ? ORDER BY trade_date ASC",
+                    [ts, date_str],
+                ).fetchall()
+
+                for (fc,) in future:
+                    fc = float(fc)
+                    if fc >= target:
+                        wins += 1
+                        returns.append((target - entry) / entry)
+                        break
+                    elif fc <= stop:
+                        losses += 1
+                        returns.append((stop - entry) / entry)
+                        break
+                else:
+                    pending += 1
+            except Exception:
+                pending += 1
+
+    conn.close()
+    total = wins + losses
+    return {
+        "total": total,
+        "wins": wins,
+        "losses": losses,
+        "pending": pending,
+        "win_rate": round(wins / total * 100, 1) if total > 0 else 0,
+        "avg_return": round(sum(returns) / len(returns) * 100, 2) if returns else 0,
+        "avg_win": round(sum(r for r in returns if r > 0) / max(1, sum(1 for r in returns if r > 0)) * 100, 2),
+        "avg_loss": round(sum(r for r in returns if r < 0) / max(1, sum(1 for r in returns if r < 0)) * 100, 2),
+    }
diff --git a/src/ashare_dp/market/recommendations.py b/src/ashare_dp/market/recommendations.py
index d2d7f0f..8faf06d 100644
--- a/src/ashare_dp/market/recommendations.py
+++ b/src/ashare_dp/market/recommendations.py
@@ -137,7 +137,7 @@ def get_recommendations(
                     WHERE l.amount > $min_amt
                 )
                 SELECT *,
-                    (COALESCE(ret_5d, 0) * 40 + LEAST(vol_ratio, 3.0) / 3.0 * 30 + trend_score * 15) AS composite
+                    (COALESCE(ret_5d, 0) * 50 + LEAST(vol_ratio, 3.0) / 3.0 * 25 + trend_score * 12.5) AS composite
                 FROM scored
                 ORDER BY composite DESC
                 LIMIT 4
@@ -165,13 +165,13 @@ def get_recommendations(
                 atr_raw = row.get("atr20")
                 atr = float(atr_raw) if atr_raw and not (isinstance(atr_raw, float) and math.isnan(atr_raw)) else entry * 0.03
 
-                # Stop: MA20 or 2 ATR below entry
+                # Stop: 2.5 ATR below entry (optimal from grid search)
                 ma20 = float(row["ma20"] or entry)
-                stop = round(min(ma20 * 0.97, entry - 2 * atr), 2)
+                stop = round(min(ma20 * 0.97, entry - 2.5 * atr), 2)
 
-                # Targets
-                target1 = round(entry + 1.5 * atr, 2)
-                target2 = round(entry + 3.0 * atr, 2)
+                # Targets: 3.0 ATR (optimal), secondary 4.0 ATR
+                target1 = round(entry + 3.0 * atr, 2)
+                target2 = round(entry + 4.0 * atr, 2)
 
                 # Risk/reward
                 risk = entry - stop