Crypto desk — 19 Oct 2026, 19:16 UTC

Phase 1 carried two live lessons: (1) the paired reopen signal — dominance ticking down WHILE mcap stops shrinking — is a positive buy trigger, which is what put ARB in the book at 07:16; and (2) the counter-condition from 10-18, that a dominance move on a still-bleeding mcap is a shrinking pie / risk-off step that offers NO entry. This run, the regime moved against the trigger, not with it: F&G eased 50 → 48, mcap slipped $2.76T → $2.73T, dominance flat at 60.2%. The pie is shrinking again — precisely the "no entry" condition — so the ARB-style reopen read from 07:16 does not re-arm here.

check_crypto_stops fired no stops and flagged nothing ADD-RISK. My three sellable [BUY AND SELL] holdings are all MA-bullish and sitting on their hard stops: BTC 81812.1 vs stop 68678.7 (now -2.3%, peak +4.3%), AVAX 9.532 vs stop 8.77071 (now -10.9%, peak +12.2%), ARB 0.1929 vs stop 0.161294 (now -1.9%, peak +0.3%). None is a blow-off, none has a broken thesis, so no discretionary exit — I won't churn a healthy hold.

On buys: ARB is again the lone Candidate = buy (price 0.193035 above fast MA 0.191705 > slow 0.181718, +5.5%), and I already hold it, so an add is mechanically allowed (not ADD-RISK). But I bought it yesterday at 0.197 and it is -1.9% with peak only +0.3% — the reclaim is barely holding, not extending — and the regime turned the wrong way (shrinking pie). Adding into that, after a +7.6% 24h day, is a marginal chase, not a confirmed follow-through. Per my own lesson and the Decision Framework (constructive trend yes, but downside = a normal pullback stops the whole lot after a weak reclaim in a risk-off step), the disciplined move is to let ARB prove itself rather than double down.

No trades — held all 24 positions. No stop fired and no discretionary exit warranted (BTC, AVAX, ARB all MA-bullish, on hard stops, no blow-off, no thesis break). No buy: ARB was the sole candidate and I already hold it, but the paired reopen signal reversed this run (mcap $2.76T → $2.73T, F&G 50 → 48, dominance flat 60.2%) — a shrinking-pie / risk-off step that my 10-18 lesson says offers no entry — and ARB's own reclaim is barely holding (-1.9% from my 0.197 entry, peak +0.3%). Adding there would be chasing a +7.6% 24h day into a weakening regime. Preserved the daily cap ($5.00 of $50 still spent from the 07:16 ARB buy, $45.00 left). Cash $13.77, book value $724.65. Desk-traded P&L at review: realized $-63.88, unrealized $-20.44, total $-84.32.

The paired reopen signal is fragile — it can reverse the very next run, so a buy made on it is not a mandate to ADD on the next pass; re-check the pie before compounding. One run after the 07:16 paired-signal ARB buy (mcap $2.70T→$2.76T, F&G 47→50), this run gave the improvement straight back: mcap $2.76T→$2.73T, F&G 50→48, dominance flat 60.2% — the shrinking-pie / no-entry condition from "watch-dominance-roll-over"-needs-a-second-condition-mcap-stabilizing. ARB itself only barely held its reclaim (bought 0.197, now 0.1929, -1.9%, peak +0.3%) — a reclaim, not yet follow-through. This answers the open question from the-paired-reopen-signal-is-a-leading-buy-trigger-act-on-the-turn: the signal predicted a single reclaim, NOT durable follow-through, so treat a paired-signal entry as a probe to let prove itself, and require the pie to still be growing before adding to it — don't chase the same name into a pie that has resumed shrinking. Held all 24, no stop fired, $5 of $50 spent (prior run's ARB only).