1. What the system observed. On 2026-09-03, INTC's funding rate on Coinbase International (INTX) cleared the entry threshold (+20.1% APR) with the required persistence, and basis was tight (+0.18%). A position opened: short the INTC perp, long INTC stock via Interactive Brokers.

2. Why the trade was entered. Every entry gate passed — nothing unusual. Entry prices: perp $89.48, IB $89.32.

3. What the model expected. A funding-rate arbitrage trade with price risk cancelled between the two legs, earning the funding differential until an exit condition fired.

4. What changed. On 2026-09-07 at 23:34 UTC, funding flipped hard negative — an emergency trigger (FR = −52%), which bypasses any grace period. The perp leg closed immediately at $97.76 to stop the funding bleed. IB's market was closed at that hour, so the hedge leg deferred to the next open, sitting naked (unhedged, long-only exposure to INTC) for about 14 hours.

5. Funding, price movements, and costs. Funding earned over the ~4.4-day hold: +$8.84 — modest, as expected from a ~20% APR position held a few days. The perp leg (short) lost −$925.35 as INTC's price rose over the full hold. The IB leg (long) gained +$1,303.18 over the same window. Fees: −$11.50.

6. Final result. The IB leg closed the next morning (2026-09-08, 13:34 UTC) at $100.96 — INTC kept rising through the naked window. That gap, measured from the $95.80 reference price at the moment the perp closed to the $100.96 final print, is booked as naked-leg P&L: +$538.62. Net P&L (strategy accounting): +$535.96. True P&L (all-in, both legs' full price move): +$375.17.

7. What was learned. This is the mirror image of the ARM outage and the SNDK case below — same mechanism (perp closes first, hedge leg sits naked until the next open), opposite direction. The naked window is pure coin-flip directional exposure; it isn't inherently a source of loss. What differs trade to trade is which way the coin lands, and the ARM incident specifically was about the window being 15 days long instead of 14 hours — duration, not direction, is what turned a bounded risk into an unbounded one. See What happens when the perp closes before the hedge for the general mechanism.