Crypto desk — 27 Sep 2026, 07:24 UTC

Phase 1 lessons frame today exactly: distinguish a slow dominance drift from a spike, and read risk-off by composition — a genuine dip trigger is a dominance SPIKE + Greed COLLAPSE that pulls the blue chips down, not any single soft indicator. Availability of cash/cap is not edge.

The board is the same regime those lessons describe. Fear & Greed 72 (Greed), BTC dominance 58.5% (flat vs the prior run's 58.5%, not spiking), mcap $2.90T, Mood "Neutral", book 24h $+2.96 (+0.49%). The blue chips are flat, not dipped: BTC +0.6% 24h, ETH +0.6%, BNB +0.3% — no low-extension pullback to anchor a fresh entry into. Meanwhile the leaders are more parabolic than ever: QNT +77.2% 24h / +169.7% 7d / +177.4% 30d (price 179.3 vs fast MA 77.30 — ~132% above it, yet still flagged Candidate = buy), NEAR +10.8% / +55.6% / +187.4%, DASH +12.6% / +23.6% / +83.7%, TON +10.0% 24h. Those are the momentum/blow-off traps my lessons warn against, and the screen's "buy" flag does nothing to discount their extension — that judgment is on me.

Exit pass: check_crypto_stops fired nothing — ETC 9.516 vs stop 8.4419, AVAX 11.14 vs 8.77113, BTC 84482.4 vs 68948.4, AAVE 155.557 vs 119.437, ARB 0.2265 vs 0.202872. All sit on the hard floor (none armed to breakeven/trail), so none is ADD-RISK.

No trades — held all 19 positions. Discretionary exit pass on my [BUY AND SELL] names (BTC, AVAX, ARB, AAVE, ETC): all MA-bullish, no broken thesis, no blow-off, and the backdrop is risk-on not risk-off — so no trim. No buy either: cash $241.87, full $50 daily cap and 3 slots open, no pending orders — but on a Greed-72 board with dominance flat (not spiking) and the blue chips flat (not dipped), there is no low-extension entry to anchor into, and the hot names (QNT +77% 24h chief among them) are blow-off risk. Availability isn't edge; the discipline is to keep waiting for a genuine risk-off dip rather than manufacture a marginal chase. Desk P&L at review: realized $-2.27, unrealized $+13.42, total $+11.15.

The Candidate = buy flag is a trend screen, NOT an extension filter: it fires the instant price clears the fast MA and says nothing about how far above it price sits — so a parabolic blow-off reads "buy" exactly like a fresh base breakout. The extension judgment is entirely mine; never treat "buy" on the screen as permission when 24h/7d/30d shows a vertical run. Evidence: QNT printed +77.2% 24h / +169.7% 7d / +177.4% 30d with price 179.3 sitting ~132% above its fast MA (77.30) — the single most stretched name on the board — yet the research still tagged it Candidate = buy right alongside flat blue chips (BTC +0.6% 24h above fast MA, ETH +0.6%). Same regime as the prior holds (Greed 72, dominance flat 58.5%, book +0.49% 24h, no stop fired, none ADD-RISK), no trade placed. Refinement over distinguish a slow dominance bleed from a dominance spike and read the risk-off signal by its composition: those govern WHEN to deploy; this pins down that the buy-candidate flag itself carries no extension information, so a "buy" tag on a coin up 77% in a day is a momentum trap wearing a green label — cross-check every candidate against its multi-timeframe run before trusting the flag.