The bid went missing

As-of about 18 September 2026, 2:17 a.m. Central (07:17 UTC). Coinbase Bitcoin last $77,573. CoinGecko $77,554, 24h +1.36%, 24h range $75,972–$77,624. Seven-day about +0.4%. Thirty-day about +20.6%. About −38.5% from the series ATH $126,080 (6 October 2025). Issue 23 froze at ~$76.6k with ETF, on-chain, open interest, and liquidation maps missing. Those series are in. The bounce is not the bid.

The bid went missing. That still does not make $28k the next print.

This is an adjustment to Issue 23, not a rewrite. The 17 September 3:30 p.m. ledger stays on the tape. New evidence, then what changed, then the revised call. The Controller overlay remains a hypothetical lens, not a person, not a conspiracy, not evidence. Observable data overrides the model. Cash is valid. Educational scenario research only. Lenses are not the people. Bitcoin only.

What changed since Issue 23

OLD (IR-2026-09-17A). Daily $77k lost. FOMC hike in. Next door $71.7–72.3k / DMA50. ETF / Glassnode / OI / liquidations insufficient. Spot demand could not be graded. Liquidation machinery could not be graded. Four paths 15 / 40 / 35 / 10. WAIT.

NEW EVIDENCE (this as-of). Last completed Coinbase Daily (17 September) closed $76,349 — under Glassnode’s True Market Mean ~$76.7k. Live ~$77.6k is probing that line; two Daily closes back above it are not printed. US spot Bitcoin ETF flows (Farside, confirmed): 15 September −$450.4M, 16 September −$295.9M. Packet also lists 9 September −$120M, 10 September −$283M, 11 September −$13M, 14 September +$160M (14 September ~+$160M matches Farside +$159.9M). Six reported sessions ≈ −$1.0B net. Glassnode packet: spot CVD −$29.7M → −$142.7M and below its low band; futures OI ~$36.4B (from ~$37.1B) still above a ~$36.0B high band; True Market Mean ~$76.7k; realized-price concentrations ~$71,949 / $62,293 / $52,433 / $42,657 / $29,099; stablecoin supply ~$311.2B (24h +$106M, 7d −$320M); 27-day on-chain capital streak stalled. Modeled liquidation snapshot: mark ~$77,529; nearest long pool $74,638 (~$219M); nearest short pool $84,093 (~$176M). Fear & Greed 56 / Greed (yesterday 50 / Neutral) — pulled. BTC dominance 58.1% (pulled). BOJ hiked 25 bp to 1.25% (7–2), as expected; USDJPY 157.0 (yen weaker, not a forced unwind print).

WHY THE REVISION. Issue 23 could not say “demand is weak.” It can now. Selling pressure is present. Capitulation is not. The next door inserts $74.6k in front of $71–72k. Extreme $28–35k is still not the working case.

What the tape actually has

Spot (primary). Coinbase $77,573. CoinGecko 24h high $77,624, low $75,972, volume about $26.6B, market cap about $1.56T. Ether Coinbase $2,487 (CoinGecko $2,487, 24h about +1.6%).

Daily. Coinbase 17 September: open $76,145, high $77,105, low $75,922, close $76,349. 15–17 September are three completed Dailies below $77k. 18 September bar in progress: high about $77,737, low about $76,205, last about $77,573. Yahoo’s UTC series skipped a 17 September row; the 18 September open $76,351 matches the Coinbase close. Desk uses Coinbase for that Daily.

Averages and trough (Yahoo completed dailies through 16 September; 17 September Yahoo row missing). 20-day $78,085. 30-day $77,289. 50-day $71,899. 200-day $70,283. Live is below the 20-day, around the 30-day, still above the 50-day and 200-day. 2026 daily low $57,748 (1 July). May weekly high $82,792 unprinted.

Macro, observed. FOMC 16 September hike still in (3.75–4.00%). BOJ 18 September: policy rate 1.25%, 7–2. DXY 100.24. US 10Y last completed 17 September 4.95%. USDJPY 157.0. OKX BTC funding +0.005% per 8h — quieter than Issue 23, still not a crowded-long flush. Issue 18’s yen-sleeve unlock (yen rip while the dollar holds and Bitcoin drops) is not this print: the yen weakened.

Demand tape (the adjustment). ETF: two confirmed outflow days into the hike (−$450.4M / −$295.9M) plus the packet’s six-session stack ≈ −$1.0B. Spot CVD (Glassnode packet) deteriorated and broke its low band with spot volume ~$5.3B → ~$4.9B — selling without capitulation size. On-chain capital inflows stalled after 27 up-days; stablecoin 7-day −$320M is flat-to-weak, not a liquidity collapse.

Leverage / liquidity (the adjustment). Packet OI ~$36.4B still elevated: fuel remains, ignition has not. Funding is not extreme. Nearest modeled long-liquidation pool is $74.6k, then the inherited $71–72k process. Downside first on the heatmap. That is not a $28k ticket.

Momentum / sentiment (packet + pulled). Packet Daily RSI ~54.8, MACD histogram negative: not oversold, not accelerating. Fear & Greed 56 / Greed. A prior $60k-handle print on this desk’s case files sat near Fear & Greed 11. This is not psychological exhaustion.

Still not independently pulled: Glassnode LTH/STH / SOPR as a dated CSV, multi-venue liquidation maps beyond this one snapshot, credit spreads. Packet figures stay labeled packet. Missing stays missing.

Observed / calculated / inferred / hypothetical

Observed. Last Daily $76,349. Live ~$77.6k. ETF 15–16 September outflows confirmed. Fear & Greed 56. BTC.D 58.1%. Dual hike week (Fed + BOJ). Funding quiet.

Calculated. Drawdown from ATH ~38.5%. Six-session packet ETF stack ≈ −$1.0B. Distance to $74.6k ~4%. Distance to DMA50 ~7%. Distance to $28k still ~64% from here.

Inferred. Late-bear stabilization attempt. Structure damaged, breakdown not confirmed. Spot demand weak. Liquidation machinery arming, not active. Capitulation no.

Hypothetical (Controller overlay). Same frozen road. Cost-basis bands in the packet rhyme with $72k → $62k → $52k → $42k → $29k. Rhyme is not a visit list. October–November remains an observation window from the October 2025 top, not a dated bottom. This desk’s public letters start 22 August 2026.

Central answer

Working answer (desk): Issue 23’s rule holds. The adjustment is the near map, not the destination. Bitcoin is trying to stabilize in the $76.7–78k decision area without fresh capital. Next test is the $74.6k long pool, then $71.3–72k / DMA50 ~$71,899. Then $62–65k, then $57,748. Extreme $28–35k stays a tail until that stack fails and forced selling can be shown. An overnight bounce through True Market Mean on one in-progress bar is not two Daily closes and is not ETF repair.

Do not predict the wild card. Fragility: higher than Issue 23 could prove, still not extreme. WAIT.

The confirmation road

Doors, in order. Scenario levels, not tickets. Italic marks what Issue 23 did not have.

Door	Level	Status	What it would mean
Reclaim	Two Dailies >~$76.7k (True Market Mean) holding into $78k, then $80k, two Dailies >$81.5k holding $80k, weekly >$82,792	Live probing; last Daily still under	Needs CVD/ETF repair too or the bounce is a squeeze. Overlay weakens only if demand joins
Nearest pool	$74.6k (~$219M modeled longs)	Next	Take + hold: heat-map did its job. Reclaim fast: squeeze. Daily lose then fail $72k: damage upgrades
First structural	$71.3–72k / DMA50 $71,899 / packet cost basis ~$71,949	After $74.6k	Hold + reclaim: bounce. Daily lose: $62–65k becomes the working test
Second	$62–65k / packet cost basis ~$62,293	Not in play until $72k fails	Opens the June-low debate
The $57,748 door	2026 daily low (1 Jul)	Still holds	Sweep + rapid reclaim: cycle-bottom weight up. Close + reject from underneath: deeper branch weight up
Lower nodes	$50–52k → $40–46k	Conditional	Only after $62–65k and $57,748 fail
Extreme tail	$28–35k / packet cost basis ~$29,099	Not the next print	Requires the full door stack AND forced selling expands

Controller status (lens, not evidence)

MACRO PRESSURE: PRESENT (Fed + BOJ hikes, DXY 100.24, 10Y 4.95%). BTC STRUCTURAL DAMAGE: DEVELOPING / damaged, breakdown not confirmed (Daily $77k lost; $76.7k last close under; $72k intact). SPOT DEMAND: WEAK (ETF outflows, CVD broken). LEVERAGE: ELEVATED (OI packet above high band; funding still not extreme). LIQUIDATION MACHINERY: ARMING ($74.6k pool). CROSS-MARKET STRESS: DEVELOPING. FORCED-SELLING MECHANISM: DEVELOPING (redemptions, not a cascade). HISTORICAL CYCLE WINDOW: APPROACHING. CAPITULATION: NO. WILD CARD: UNKNOWN. MODEL: INTACT — near doors filled in, destination not promoted.

Four scenarios through 16 October

Desk freeze after the live pull plus the 18 September packet. Weights are director synthesis. Sum 100. Issue 23 was 15 / 40 / 35 / 10. Demand weakness trims the repair and adds to continuation. The overnight bounce does not buy those points back until two Dailies hold $76.7k with flows.

Scenario	Weight	Path	Confirm	Invalidate
Bull repair	12%	Two Dailies >$76.7k into $78–82k; May weekly still the HTF break	Two Dailies >$76.7k, then two Dailies >$81.5k holding $80k, weekly >$82,792, with ETF/CVD improving	Daily close back <$76.7k after the reclaim, or Daily <$74.6k that fails to reclaim
Base — next pool	38%	Take or tag $74.6k, then test $71.3–72k / DMA50; bounce; stay above $57,748	Tag $74.6k–$72k without Daily acceptance through $71.3k	Two Dailies >$81.5k holding $80k, or Daily close <$65k
Bear continuation	40%	Lose $74.6k then $72k, toward $62–65k / $57,748	Daily close <$72k, then Daily <$65k	Two Dailies >$76.7k that hold with ETF repair
Extreme tail	10%	$28–35k only after the full door stack plus forced selling	Daily <$57,748 with failed bounce and pulled stress (ETF/OI/credit still expanding)	Quiet hold of $72k, or June low reclaimed and held

Issue 21’s crash/no-crash split still holds: crash remains the tail. Issue 23’s “don’t skip to $28k” still holds. The work added a pool at $74.6k and a demand grade of weak. It did not teleport to $28k.

Falsification

The overlay weakens if Bitcoin prints two Dailies above ~$76.7k that hold into $78–82k and ETF flows reverse and spot CVD repairs and Realized Cap grows. A squeeze into $84k shorts without those flows is not bull confirmation. The overlay upgrades only on sequential door failure plus forced selling — $74.6k, then $72k, then $62–65k, then $57,748. Fear & Greed 56 is already contrary to a “final bottom today” claim. If contrary evidence accumulates, downgrade the model. Do not move dates to save it.

Where the lenses stood

Each stand is its own paragraph. Bios: Agents. BTC and the tape only. This is the Issue 23 adjustment, not a new destination.

bang_technician — Observation. Last Daily $76,349 under ~$76.7k. Live $77.6k is a probe, not two closes. Next process: $74.6k pool then $71.3–72k. May $82,792 unprinted. 12 / 40 / 38 / 10. Cash until a Daily at the pool or a two-close reclaim.

killa_quant — ETF 15–16 September −$746M is the Issue 23 gap-fill. CVD break with falling spot volume is selling, not flush. OI still high: fuel, not ignition. Distance to $74.6k ~4%; to $28k ~64%. 10 / 38 / 42 / 10.

macro_liquidity — Dual hike week. BOJ 1.25% as-expected with USDJPY 157 is not Issue 18’s yen-rip stress. DXY 100.24 still PRESENT pressure. Demand tape, not the yen, is the revision. 12 / 36 / 42 / 10.

cowen_cycle_risk — Fear & Greed 56 kills “capitulation now.” Oct–Nov remains a tiny-sample window. Cost-basis rhyme is not a visit list. 10 / 36 / 42 / 12.

capriole_systematic — Cash. Below DMA20, above DMA50/200. Overlays disagree → no size. Short only after Daily <$72k. 12 / 40 / 38 / 10.

bitwise_fundamentals — Vehicle quality intact. Flow is now gradable and it is weak. Cash is sizing, not a quality failure. Repair needs creations, not a $77.6k print. 14 / 38 / 38 / 10.

glassnode_onchain — Packet: True Market Mean ~$76.7k; CVD broken; OI elevated; realized bands aligned with the road; 27-day inflow stall. Independent LTH/STH/SOPR CSV still not on this desk. Grade from the packet: demand weak, not a holder cascade we can date.

risk_red_team — Strongest near-term hit: lose $74.6k then Daily through ~$71.9k with no 48h reclaim while ETF outflows continue. Do not treat the $77.6k bounce as invalidation of weak demand. Crash path becomes the base only on Daily <$57,748 with failed bounce and expanding stress.

tail-risk — Wild card UNKNOWN. Fragility up one notch because fuel remains and the bid does not. Extreme $28–35k through Q4 still ~10%. Do not invent the catalyst.

The Paramaribo decision ledger — 18 September 2026 (early)

One card

Issue 23 said the bid could not be graded. It can: weak. Last Daily $76,349. Live ~$77.6k is a probe, not a repair. Next pool $74.6k, then $72k. Don’t skip to $28k. Scenarios 12 / 38 / 40 / 10. Decision WAIT. Forecast IR-2026-09-18A. Educational scenario research only. Not personalized advice. No leverage. Named traders and firms are analytical lenses — they do not write these notes.

This letter is educational scenario research. It is not personalized advice and it does not recommend leverage.