Crypto desk — 21 Sep 2026, 19:16 UTC

The board is euphoric — Fear & Greed 79 (Greed), BTC dominance 59.2%, total mcap $2.92T — and wall-to-wall extended: NEAR +111.1% 30d, ARB +124.9%, UNI +105.2%, STX +51.7%, ENA +39.7%. That is exactly the blow-off-top screen my recent lessons keep flagging, so most candidates are momentum traps, not entries.

The material change this run is cash: quote is back to $50.56 after four cash-starved runs (a deposit landed on top of last run's $0.56), with $45.00 of the $50 daily cap left and up to 3 new buys of $5 available. That reopens real optionality — but a greedy, heavily-extended tape is not a signal to lean in; it's a reason to take only the calmest, most-recoverable names and keep dry powder for a plausible pullback.

check_crypto_stops fired nothing — all holdings above hard floors. It flagged AVAX (trail armed at 10.0104, +44.3%) and LTC (breakeven armed, +17.2%) as ADD-RISK, so both are off the add list. On the discretionary-exit pass over my [BUY AND SELL] book (ETH, BTC, BNB, BCH, ADA, LTC, TON, AVAX, AAVE) nothing broke: all green and MA-bullish. TON's marginal bearish cross (-0.2%) on a flat $5.73 hold isn't worth a churn fee, and AVAX's parabola is covered by its armed trail — no discretionary exits.

For buys I screened the whole Coins table for lowest extension + constructive MA + recoverability. BTC is the standout: MA gap only +7.1%, 7d +9.0% / 30d +11.4%, and just -31.8% from ATH — the least-extended large cap on the board, [BUY AND SELL] (recoverable), not ADD-RISK. ETH is the next calmest: MA gap +9.4%, 7d +8.9% / 30d +13.6%, -44.2% from ATH, also [BUY AND SELL] and not ADD-RISK. BNB (+8.4% gap, -41.6% ATH) was comparable but I passed on a third buy — on an F&G-79 tape I'd rather keep the dry powder than deploy a marginal third tranche.

Bought $5 BTC (filled 0.00005789 @ 86364.74) and $5 ETH (filled 0.00180397 @ 2771.66); ~$0.01 taker fee each taken in-kind. These are the two lowest-extension, constructive, recoverable blue chips on a euphoric board — spreading $10 across both rather than concentrating. Held all 27 existing positions — no stop exits, no discretionary exit or trim. Spent $10.00, so $35.00 of the $50 daily cap remains; cash goes to ~$40.56, deliberately kept idle rather than leaning into greed. Desk-traded P&L at review: realized $+0.21, unrealized $+35.31, total $+35.52.

When cash flips from starved back to ample, the buy caps (≤3 new buys × $5) are a ceiling, NOT a target — on a euphoric, heavily-extended board, deploy only into the one or two calmest low-extension recoverable names and bank the rest as dry powder; filling all three slots just because you finally can is leaning into greed. My last four fundable runs were one-buy-floor decisions, so all recent lessons optimize scarcity (leave ≥$5 idle; break ties to [BUY AND SELL] for recoverability — when cash is scarce prefer adding to a buy-and-sell name so deployed cash stays recoverable, the leave-5-idle buffer stops binding at the one-buy floor). This run a deposit restored quote to $50.56 (from $0.56) with $45 of daily cap and 3 buys available — the first genuine abundance decision in a while. The board, though, was F&G 79 (Greed) and wall-to-wall extended (NEAR +111% / ARB +125% / UNI +105% 30d), i.e. the blow-off-top screen at full tilt. So abundance did NOT change the core call — I still took only the two calmest low-extension recoverable blue chips (BTC +7.1% MA gap / -31.8% ATH; ETH +9.4% / -44.2% ATH, both [BUY AND SELL], neither ADD-RISK), passed a comparable-but-marginal third (BNB), and kept ~$40 idle. The refinement over the scarcity lessons: recoverability and low-extension selection still govern in abundance; what changes is that the binding discipline becomes restraint against the caps rather than rationing the last $5. Regime (greed + extension), not available cash, sets how many slots to fill.