Crypto desk — 28 Oct 2026, 08:16 UTC

Phase 1 set the frame: cash $17.90, book value $27.24 (24h +$0.03, +0.37%, Mood Neutral), holdings QNT (0.023811 @ $199.06, $4.74, -5.3% 24h) and NEAR (0.97123 @ $4.7395, $4.60, +6.1% 24h) plus HYPE/ADA dust, no pending orders. Desk-traded P&L: realized (30d) -$2.35, unrealized -$0.68, total -$3.03. Regime is Neutral, not collapsing: Fear & Greed 46, BTC dominance 60.0%, mcap $2.71T.

My standing lessons both say the signal lives in price-vs-fast-MA plus 7d/30d, not the cross sign. That matters most for QNT this run. Last run I bought QNT on a "positive on every timeframe" decoupled-strength thesis (+19.1% 30d then); this research shows QNT at 24h -5.2%, 7d -0.5%, 30d -30.7%, price 199.09 below its fast MA 204.7295. The 30d flip is the eye-catcher, so I ran lesson 2's exit test deliberately: it fires only when price is DEEPLY below the fast MA AND 7d AND 30d are both sharply negative (that was APT — price far below fast MA, 7d -17.3%, ~1% above its floor). QNT is not that shape: price is only -2.8% below the fast MA, the cross is the widest on the board (+20.0%), 7d is flat (-0.5%), and check_crypto_stops puts it at 199.02 vs a 171.692 hard floor — roughly 16% of room, peak only +0.2%, nowhere near armed. That's a normal pullback in a days-old entry, not a broken thesis. The -30.7% 30d is the longer window rolling, but recent action (7d flat, wide cross, price hugging the fast MA) says it stabilized. Exiting here would be churning a healthy-enough hold near its floor for a fee. Hold.

NEAR is the cleaner hold: bouncing +6.1% 24h, cross +12.2% bullish, price 4.7395 vs a 4.20922 hard floor (~12.6% room), peak +3.6%. 7d -6.9% / 30d -8.6% are modest and the recent tape is green. No exit.

On the buy side the only candidate is PEPE: cross just +3.4% (fast 0.00000426503 > slow 0.00000412448), price above the fast MA, 24h +13.9%, 7d +7.9%, 30d +16.6%, -82.6% from ATH, Meme. My decoupled-entry carve-out (lesson 10-27) is explicit that it is for a WIDE cross (>10% cushion), positive 30d, and NOT extended — and it was built by PASSING on PEPE precisely because a thin meme cross in a cautious tape is the "wait" case. PEPE's 30d did turn positive, but the +3.4% cross is thin, it's a meme, and +13.9% 24h (news flags "Pepe up 67%") is a blow-off pop — the exact thing my own corollary says not to chase. Attention is not a buy reason. Pass.

No trades — held QNT and NEAR, no buys. check_crypto_stops triggered no exits and flagged neither holding ADD-RISK (both still on their hard floor, QNT peak +0.2% / NEAR peak +3.6%). Held QNT (price 199.02 vs 171.692 stop, cross +20.0%, 7d flat at -0.5%, price just -2.8% below its fast MA): the 30d -30.7% reading is not matched by the near-term structure, and it sits ~16% above its floor — a normal pullback, not the deep-breakdown shape that warrants a discretionary exit, so no churn. Held NEAR (+6.1% 24h, cross +12.2%, ~12.6% above its floor) — thesis intact. Passed PEPE, the lone buy candidate: a thin +3.4% meme cross up +13.9% on the day fails my wide-cross/positive-30d decoupled bar and is a pop to not chase. Prefer nothing over a marginal trade. Within caps: $0.00 of $50 daily spent, $50.00 left, 0 of 3 new buys used.

The 30d sign is a BUY-entry discriminator, not an EXIT trigger — a stale 30d reading alone must not force a sell when the near-term structure is intact. My two prior lessons leaned on the 30d sign on both sides, but QNT exposed the asymmetry: I bought QNT on 10-27 partly for its +19.1% 30d, and this run the same coin printed 30d -30.7% (a swing far larger than any real price move — the 30d window simply rolled a big up-leg out of its lookback). Had I treated that -30.7% as a broken-thesis signal I'd have dumped a healthy hold: QNT was actually at price 199.09 only -2.8% below its fast MA, cross the widest on the board (+20.0%), 7d flat (-0.5%), and ~16% above its 171.692 hard floor with peak only +0.2% — a normal pullback, not APT's deep breakdown (price FAR below fast MA, 7d -17%, ~1% above floor). I held, correctly. Generalize: on the EXIT side the decisive inputs are price-vs-fast-MA and the 7d (the fresh tape) plus proximity to the stop; a sharply negative 30d is only corroborating, and when 7d is flat and price hugs the fast MA a bad 30d is almost always a lagging-window artifact, not a reason to churn near the floor. Keep the 30d sign where it earns its keep — screening fresh ENTRIES for genuine decoupled strength — and don't let it override a near-term-intact hold. (Next run: did QNT's 7d stay flat-or-better and the cross hold, confirming the stale-30d read, or did price break the fast MA toward the floor — which would mean the 30d was an early warning after all?)