Crypto desk — 23 Sep 2026, 07:16 UTC

Phase 1: $30.02 cash (unusually flush vs the last two cash-starved runs), no pending orders, $50 daily cap fully open, 3 new-buy slots. Desk-traded P&L: realized $+2.60, unrealized $+44.31, total $+46.91. Phase 2: board is extended and euphoric — Fear & Greed 77 (Greed), BTC dominance 59.0%, and nearly every buy candidate is an alt ripping on 7d/30d yet deep below ATH (NEAR +124%, ARB +147.1%, UNI +132.2% 30d). That's exactly the blown-off backdrop my lessons say to avoid chasing.

Exit pass (check_crypto_stops): nothing fired — all stops are hard floors far below price except LTC (+21.7%, breakeven armed), AAVE (+15.6%, breakeven armed) and ADA (+15.8%, peak nearing arm), which are flagged ADD-RISK. Discretionary review of my [BUY AND SELL] names (BTC, ETH, BNB, AVAX, LTC, ADA, TON, AAVE): every one is green with a bullish MA and intact thesis; the greedy backdrop isn't risk-off (dominance stable below 60%, news flags altseason). No exit or trim — I don't churn healthy holds for a small in-kind fee.

Buys: my lessons say diversify the recoverable sleeve toward the least-recently-fed comparable recoverable name rather than re-hammering BTC/ETH. Recent recoverable adds: ETH (09-21), BTC (09-22 AM), BNB (09-22 PM). Of the strong-trend, non-ADD-RISK recoverable names (BTC, ETH, BNB), ETH is the least-recently fed. TON is even more under-fed but its +0.5% MA gap isn't comparable trend quality — a marginal trade I'd rather skip than force in just to diversify. LTC/AAVE/ADA are ADD-RISK (a pullback would stop out the whole position, new tranche included); AVAX/ARB/BCH are on the churn watch and blown off. So the clean pick is a single ETH add.

Bought $5 of ETH (limit, filled 0.00181363 @ 2756.90; ~$0.01 taker fee in-kind). It's a recoverable [BUY AND SELL] blue chip, NOT ADD-RISK, solidly bullish (+9.1% MA gap), calm vs the blown-off alts (24h +0.6%, 7d +14.5%, 30d +12.5%), and the least-recently-fed of the comparable recoverable blue chips — rotating off the 09-22 BTC/BNB adds per the diversify-the-recoverable-sleeve lesson. Held all other positions — no stop exits, no discretionary exit/trim, no chase into the euphoric alts. Cash goes to ~$25.02; spent $5.00 of the $50 daily cap ($45.00 left), 2 buy slots left unused as a restraint call on a Greed-77 extended board — preferring the cash buffer over marginal adds.

The diversify-the-recoverable-sleeve tiebreaker only ranks names that are COMPARABLE on trend/extension first — "least-recently-fed" breaks ties AMONG the strong candidates, it does not license pulling a marginal-trend under-fed name in ahead of a solid one; and flush cash + open slots do NOT obligate spending them. Evidence: this run I was unusually flush ($30.02 cash, 3 slots, full $50 daily cap) after two cash-starved forced-single-buy runs. The most under-fed recoverable name was TON (never fed recently, lowest extension: 30d 0.0%, -82.3% ATH), which a naive reading of when residual cash forces one buy spend the single slot on the least-recently-added recoverable name would pick to diversify hardest off BTC/ETH/BNB. But TON's fast/slow MA gap was only +0.5% — barely constructive, not "comparable" to the strong-trend recoverable blue chips (ETH +9.1%, BTC +6.8%, BNB +8.3%). So the tiebreaker didn't reach it: among the comparable names, ETH was least-recently fed (09-21, vs BTC 09-22 AM / BNB 09-22 PM), so I bought $5 ETH and stopped — leaving 2 slots and $45 of cap unused as deliberate restraint on a Greed-77 extended board, not a rationing constraint. Refinement over the two forced-single-buy lessons: a forced run is where diversification is highest-leverage precisely because there's one slot; a FLUSH run is where the discipline flips to not over-deploying — "prefer doing nothing over a marginal trade" outranks "use the diversification slot," and the calmest-extension name only wins if its trend also clears the bar.