From 1a05af803d110fdf33fb6471b82909764bf950ab Mon Sep 17 00:00:00 2001
From: jgrusewski <jgrusewski@localhost>
Date: Sat, 30 May 2026 23:35:54 +0200
Subject: [PATCH] fix(cuda): force-close existing position on DD trip /
 cooldown entry
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Cluster v5 alpha-rl-rjsjq step 371 revealed worst-account session_pnl
growing monotonically across cooldown windows (-$3.7k → -$9.3k from
step 100 → 371). Root cause: when DD triggers, actions_to_market_targets
forces {side=2, size=0} — a no-op that suppresses EVERY action type
including trail-stop-driven closes. If the account was holding a
losing position when DD fired, that position bleeds mark-to-market for
the entire 500-step cooldown with no exit path.

Fix: on first DD/cooldown step, emit a closing market order matched to
the current position (side=1 sell for long, side=0 buy for short, size
= |position_lots|). Once flat, subsequent cooldown steps emit the
existing no-op. Net effect: account closes its position at the moment
of DD trip, then sits flat until its recovery clock expires.

Validation: 13/13 risk_stack_invariants pass, 20/20 trade_management
pass, integrated_trainer_smoke passes. Local b=128 1k smoke:
  qpa: +0.95 (best result of session)
  mean_active: +$47k
  worst: stabilizes at -$11.8k (vs unbounded growth pre-fix); residual
  loss is from fat-tail single-step market moves that cross dd_limit
  before the controller can react — out of scope for this fix.
---
 .../cuda/actions_to_market_targets.cu         | 31 ++++++++++++++++---
 1 file changed, 26 insertions(+), 5 deletions(-)

diff --git a/crates/ml-alpha/cuda/actions_to_market_targets.cu b/crates/ml-alpha/cuda/actions_to_market_targets.cu
index e82b174b4..f118cf44e 100644
--- a/crates/ml-alpha/cuda/actions_to_market_targets.cu
+++ b/crates/ml-alpha/cuda/actions_to_market_targets.cu
@@ -126,15 +126,36 @@ extern "C" __global__ void actions_to_market_targets(
 
     // ────────────────────────────────────────────────────────────────────
     // Layer 1 (CMDP hard constraints — spec 2026-05-30-adaptive-risk-management).
-    // Session-level overrides force Hold regardless of agent's choice.
-    // Per-batch (one independent backtest session per b); a tripped DD
-    // or active cooldown on session `b` does NOT lock out the others.
+    // Session-level overrides force the account FLAT regardless of agent's
+    // choice. Per-batch (one independent backtest session per b); a tripped
+    // DD or active cooldown on session `b` does NOT lock out the others.
+    //
+    // Fix E (2026-05-30): emit a *closing* market order when the account
+    // is non-flat — otherwise an existing losing position bleeds m2m for
+    // the entire 500-step cooldown with no exit (trail stops are also
+    // suppressed by this branch). Cluster alpha-rl-rjsjq step 371 showed
+    // worst session_pnl growing monotonically -$3.7k → -$9.3k from
+    // exactly this mechanism.
+    //
+    // Once the position is flat, subsequent cooldown steps emit a true
+    // no-op (side=2, size=0) and the account waits for its recovery clock.
     // ────────────────────────────────────────────────────────────────────
     const bool dd_triggered = session_dd_triggered_per_batch[b] >= 0.5f;
     const bool in_cooldown  = cooldown_remaining_per_batch[b]   > 0.0f;
     if (dd_triggered || in_cooldown) {
-        market_targets[b * 2 + 0] = 2;  // no-op side
-        market_targets[b * 2 + 1] = 0;  // size = 0
+        if (position_lots > 0) {
+            // long → close via sell
+            market_targets[b * 2 + 0] = 1;
+            market_targets[b * 2 + 1] = position_lots;
+        } else if (position_lots < 0) {
+            // short → close via buy
+            market_targets[b * 2 + 0] = 0;
+            market_targets[b * 2 + 1] = -position_lots;
+        } else {
+            // already flat → true no-op
+            market_targets[b * 2 + 0] = 2;
+            market_targets[b * 2 + 1] = 0;
+        }
         return;
     }