The single most important fact of the fortnight: the −8% mandatory de-risk line was breached on three separate days — July 2 at −9.03%, July 8 at −8.94%, July 9 at −8.70% — and each time the book recovered, sitting at −7.38% today. None of those days ran a full protocol; they ran quick triage scans that flagged the drawdown as amber, not fired. The cause was not a shock but a grind: real yield ground up to 2.30%, the rate wall the whole book is defensive against, dragging the debasement sleeve (GLD −16%, GDX −22%) and pressuring the AI names. Credit stayed flat throughout (OAS 0.76) — this was rate mechanics, not contagion. Regime stays CHOP: VIX at cycle lows keeps trying to force Bull, real yield at the band top keeps blocking it. Same wall, two weeks on.
[v5-24] is a mandatory de-risk on breach. It breached, three times. The honest reading is that a de-risk was technically triggered and not executed. Two things temper that: the breaches were shallow (worst −9.03%, ~1 point over) and the book has already recovered to −7.38% — de-risking now means selling into a partial recovery on a rate-driven grind with credit flat, the exact whipsaw the confirmation discipline exists to prevent. But it exposes a real gap: the rule has no written definition of whether it triggers on an intraday breach or a confirmed close. The live board treated it as amber-watch — confirmation-style — but the protocol text is silent. This ambiguity is now a load-bearing v5.2 revision item. And SCANES1 should have escalated to a full run on July 2, when the board showed −7.2%, inside the proximity flag. It didn't. That is a process lesson logged.
Real yield 2.30% (FRED/TradingEconomics, verified — not the Junglerock dashboard's 1.89; ERROR-004). It has ground up over the fortnight from ~2.15% to the top of its 2.15–2.30 band, and it is the mechanism behind the whole drawdown: the debasement sleeve (GLD −16.4%, GDX −21.9%, CEG −16.3%) is bleeding directly into it. The [v5-1] GLD full-exit at 2.50% is now just 0.20 away — the closest it has been, and unlike the −8% drawdown ambiguity, that line is unambiguous. If real yield tags 2.50% on a CPI surprise, the GLD exit fires cleanly. This is the live structural risk, and it is the reason the Bull gate stays shut even as VIX prints cycle lows.
No trades. Marks from the July 10 07:03 UTC snapshot — $221,953, +10.98%, drawdown −7.38% vs the $239,644 HWM. Over the fortnight the book round-tripped: −4.21% (Jul 1) → −9.03% (Jul 2) → recovered → −8.94% (Jul 8) → −8.70% (Jul 9) → −7.38% today. Three sub-−8% closes, all recovered. AI primary 48.6% (still over the [v5-25] cap), BIL 26.4% (above the 15% floor). The drawdown is not the AI sleeve — those held green from cost — it is the debasement sleeve (GLD/GDX/CEG) bleeding into the risen real yield.
| Ticker | Sleeve | Live wt | P&L | Note |
|---|---|---|---|---|
| EWY | AI / Tech | 18.1% · $40.3K | +$8,962 (+28.6%) | Held through the fortnight. Tripped the −15%-from-high trim flag July 2 (session 1 of 3), recovered before confirming. No trim — the ×3 discipline held against a one-session wobble. |
| SMH | AI / Tech | 17.0% · $37.7K | +$10,679 (+39.5%) | The semi complex steadied after HBM demand was confirmed at the June-quarter earnings. Still deeply green from cost. Held. |
| TQQQ | AI / Tech | 11.9% · $26.5K | +$8,187 (+44.7%) | 3× leverage cuts both ways through the chop. No rebuild — Bull sequence + JPY Zone + real yield <1.80% all fail. |
| GLD | Debasement | 4.6% · $10.2K | −$2,003 (−16.4%) | Bleeding into the rate wall. Clock: real yield 2.30% → 2.50% exit, 0.20 away. Hold, watch the line — the closest live risk. |
| GDX | Debasement | 3.4% · $7.7K | −$2,153 (−21.9%) | The book's weakest position. High-beta miners taking the worst of the real-yield headwind. If [v5-1] fires, this exits down ~22% — the known cost of a mechanical exit. |
| FXI | BRICS / China | 2.5% · $5.5K | −$479 (−8.0%) | China steady. Held. Votes separately from Brazil (EWZ stays exited). |
| VGK | Deglobal. | 7.4% · $16.4K | +$381 (+2.4%) | The steadiest sleeve through the fortnight. Amplify at EUR 1.20, ~0.06 away. |
| ITA | Defense | 4.6% · $10.3K | +$629 (+6.5%) | Oil range-bound on Iran de-escalation puts the up-trigger out of play. Held. |
| IBIT | Alt Monetary | 2.4% · $5.3K | −$668 (−11.1%) | BTC held better than the equity sleeve this fortnight. Zone 3 JPY blocks adds. |
| BIL | Liquidity | 26.4% · $58.6K | +$14 (+0.0%) | 26.4% live vs 33% target vs 15% floor. The dry powder — and the answer to the shakeout fear. Redeploys on a confirmed Bull reclaim, not into chop. |
| CEG | AI Power | 1.6% · $3.5K | −$685 (−16.3%) | Rate-gated shut, grouped with the debasement wound. Add gate (real yield <1.80%) fails hard at 2.30%. |
| Total · equity only | $221,953 | +$21,953 · +10.98% since inception · AI primary 48.6% (over cap) · BIL 26.4% · drawdown −7.38% vs HWM · breached −8% 3× this fortnight, recovered | ||
The −8% line breached three times, and the book didn't de-risk. That could read as ignoring the rule — it isn't. The breaches were shallow (worst −9.03%), rate-driven not systemic (credit flat throughout), and each recovered within a session or two. De-risking into those partial recoveries would have sold the exact dips the confirmation discipline exists to hold through — the same logic that kept EWY through the KOSPI crash and was proven right. But "no panic trade" is not "nothing to fix." The fortnight exposed that [v5-24] has no written intraday-vs-confirmed definition, and that the book had been carrying naked downside for three weeks. The response is not to override the rules reactively — it is to close the hedge gap deliberately and to sharpen the rule at the next revision. That is the difference between discipline and complacency: the book holds, and the system improves.
The book has been unhedged since June 17. This session closes that gap — by rule, not reflex. The structure: an XSP 695/628 put spread, September 18 expiry, ~0.8% of book. The strikes are not round numbers — 695 and 628 are the technical "Tail Bearish 1 and 2" levels, so the hedge kicks in exactly where the confluence model says the first real breakdown is and covers down to the deep tail. Modelled cost ~$1,800; max protection ~$38,000; the payoff is 18:1 because it only pays on a genuine tail event (SPX below ~6,917). Monte-Carlo'd, it takes the probability of a >−15% book drawdown from ~2% to zero over the next two months.
The entry rule is the insight of the session. The hedge does not go on today. It rests at a limit that only fills into complacency — when the market grinds up or vol compresses, i.e. when VIX pushes below 15. That inverts the retail reflex: you buy protection when it is cheap and unwanted, not after the shock when it is expensive. VIX was 15.85 at structuring — close, not through. The order waits. Insurance bought before the fire, priced at the moment nobody thinks they need it. Status: pending, resting. The book remains unhedged until it fills.
Verified FRED/TE. At the band top, 0.20 away — the closest hard line and unambiguous. If it tags 2.50% on a CPI print, the GLD exit fires cleanly. The live structural risk.
Breached 3× (Jul 2/8/9), recovered each time. 0.62 from here. The rule lacks an intraday-vs-confirmed definition — now a load-bearing v5.2 revision item. No de-risk into the recovery.
Still over. Adds vetoed — including the AI-infra and AI-bio watchlist. Not a trim signal; the cap blocks new risk, doesn't force selling.
Flat throughout the fortnight's breaches. Confirms it was rate mechanics, not contagion — the reason the answer stays hold.
XSP 695/628 put spread, ~0.8% book, at a complacency-trigger limit. Closes the three-week hedge gap. Fires into the melt-up, not the shock.
Fourth week sub-18. VRP +7.90 (very positive). Deep contango. The calm that makes protection cheap to buy — and precedes shocks more often than follows them.
The cash that limits drawdowns and stands ready to redeploy on a confirmed Bull reclaim (real yield <1.80 + VIX <18 + MOU). You are not getting shaken out — you are positioned to participate.
BRL ~5.12, above the 4.90 line; z-score near norm, not weak-and-turning. Absolute and relative agree: out.
One email when a new protocol session is published. Nothing else.
[v5-23] Session close confirmed: NO TRADES — all 11 positions held · book $221,953 (+10.98%) from July 10 07:03 UTC snapshot · EVENT: [v5-24] −8% mandatory de-risk line BREACHED 3× this fortnight — −9.03% (Jul 2), −8.94% (Jul 8), −8.70% (Jul 9) — all recovered; now −7.38% · NO PANIC DE-RISK: shallow, rate-driven, credit flat (OAS 0.76) — selling the recovery = the whipsaw the confirmation discipline avoids · GAPS EXPOSED: [v5-24] lacks intraday-vs-confirmed definition (→ v5.2 revision, load-bearing); SCANES1 should have escalated to full run Jul 2 at −7.2% · TAIL HEDGE STRUCTURED: XSP 695/628 put spread (= JR Tail Bearish 1/2), Sep 18, ~0.8% book, resting GTC limit 3.00 debit, fills on VIX<15 complacency ([v5-16] corrected trigger) — book UNHEDGED since Jun 17 until fill · [v5-1] real yield 2.30% (FRED/TE verified, NOT JR 1.89), 0.20 from 2.50% GLD exit — nearest hard line · [v5-25] AI primary 48.6% over 40% cap — adds vetoed incl. AI-infra + AI-bio watchlist · [v5-2] JPY 161.60 Zone 3 · [v5-8] BRL 5.12 stay out of EWZ · [v5-27] DEFENSIVE · credit via LQD/IEF (ERROR-010), oil WTI spot not USO (ERROR-009) · Bagheera ES96/TY49/CL16/GC6.5 (trimmed equity, added gold) · YTD [MISSING ×9]