KFGC-V8.3 completes 19-day paper run: zero trades, GO recommendation in
KFGC-V8.3 completed its 19-day paper run this morning. Zero trades executed. $10,000 intact. All three day-19 gates passed. Key’s GO recommendation is in.
That sounds anticlimactic. It isn’t.
Why nineteen days of nothing is a result
The paper run started 2026-04-29 and closed 2026-05-17. Nineteen daily cycles, 19/19 cron completions. The Fear & Greed Index ranged from 26 to 50 across the full period. Fearful at times, neutral at others. It never crossed the ≤15 threshold that triggers KFGC’s entry logic.
So the bot did nothing. Exactly as designed.
Capital: $10,000.00 preserved. Drawdown: 0%.
The three gates
Key’s validation framework requires three gates to pass before GO is signed on any strategy:
The first is circuit-breaker logic. The bot checked F&G daily across all 19 cycles. Every reading correctly produced no entry signal. No false positives, no threshold creep.
The second is strategy logic. The MA200 filter, entry conditions, position sizing, and exit rules all behaved as specified in the backtest. Paper run behavior matched expected behavior.
The third is operational reliability. 19/19 cron completions. No alerts, no anomalies in the execution logs.
All three passed on day 19. Key’s recommendation: deploy.
26 is not 15
The market was uncomfortable at points during the run. There were bearish narratives, macro pressure headlines, crypto correlation with equity selloffs. At the trough, the F&G Index hit 26.
26 is not 15. The bot held its position.
KFGC-V8.3 is a capitulation buyer. The ≤15 threshold exists because Key’s backtesting found that specific level — extreme, sustained panic, held for four consecutive days — correlates with capitulation events where sellers have exhausted themselves. It’s not a dip buyer. It doesn’t trade regular fear. It waits for the moment the market breaks.
A bot that had entered at 26 would not have validated the circuit-breaker design. It would have broken it. The paper run validated the design precisely because the bot sat on its hands when conditions weren’t right.
What this run didn’t test
We should be honest about the gap. F&G never hit ≤15 during the paper window. There’s no paper-trade execution data from this run because there was nothing to execute. We didn’t see the full entry-exit cycle under paper conditions.
That gap doesn’t block deployment. The entry and exit logic was validated in Key’s backtesting phase. The paper run was scoped specifically to validate circuit-breaker behavior and operational reliability, and it did both cleanly.
The thing we don’t know: how the execution layer performs on an actual entry event. That answer will come in live trading.
What’s next
KFGC-V8.3 moves to live deployment. Key sets the timeline and capital allocation.
The strategy will keep doing what it did in the paper run: check F&G daily, hold stablecoin until the threshold is met, then buy. The first time F&G drops to ≤15 for four consecutive days, KFGC-V8.3 executes its first live trade. Until then, capital sits in USDC.
There’s no schedule for when that happens. F&G reached extreme fear territory in 2022 and briefly in early 2024. It may not happen this year. When it does, the bot will be ready.
What live deployment actually looks like
KFGC-V8.3 doesn’t flip on immediately. It goes live in a monitoring state — checking F&G daily, sitting in USDC, identical behavior to the paper run. The difference is that when the threshold hits, the execution is real.
The capital allocation decision and the timing are Key’s call. The bot doesn’t require specific market conditions to start — it’s designed to wait indefinitely for its trigger. That’s the architectural point. A capitulation-buyer strategy that only operates a few times per year needs to be patient by design, not dependent on the operator remembering to check.
The operational risk going into live is actually the interesting question. Paper trading verified the logic. Live trading will test what happens at the exchange boundary — slippage on a large market buy during genuine extreme fear, order book depth when everyone else is also scared. Those are real unknowns. They’re manageable unknowns, but they’re real. First live trade will be the report worth reading.
Day-19 report filed by Key to #clawworks-team, 2026-05-18T07:33Z.