The Shock Ran. The Door Held.

Friday’s jobs print was the unexpected event.

Twenty-nine thousand payrolls. The street wanted ninety.

Within a few hours Bitcoin had been to $87,249 and back to $83,850.

The question was never whether news can move a price. It was whether the move would find a book — and whether that book would still be there after the first rush.

As of 5 October 2026, 8:20 a.m. Central / 13:20 UTC, Coinbase BTC-USD was about $86,149. The last completed UTC daily candle is 4 October: open $84,742, high $86,793, low $84,700, close $86,507 on 2,623 BTC. The 5 October candle is incomplete and is not used as a completed-day fact. Coinbase ticker · Coinbase daily candles

1. The central answer

Observed: an identifiable macro shock traveled through Bitcoin in hours, not in the ten-minute hole Issue 28 reconstructed for 10 October 2025. Peak to trough on 2 October was about $3,399 / 3.9% in 375 minutes on the 5-minute Coinbase tape. The $82,000 door was not tagged. The afternoon low was $83,850. The daily close was $84,505. Four sessions later, 4 October produced the first completed close above $86k since 22 September — on Sunday volume of 2,623 BTC.

Calculated: from the 28 September low of $82,510 to the 2 October high of $87,249 is about +5.7% in four days. Using completed closes through 4 October, SMA20 / SMA50 / SMA100 / SMA200 were approximately $82,856 / $79,026 / $71,267 / $71,502. RSI14 (Wilder) was 68.5. MACD remained positive at about +$2,016, histogram −$63.

Inferred: speed and absorption are both in the tape, and they are not the same fact. The event was fast enough to matter. The book was deep enough, in this instance, to hold the named near-term door. That does not prove the book will hold a larger shock, a thinner session, or a second close above $86k. The desk decision remains WAIT.

2. What Paramaribo said before this week

Issue 28 froze on 30 September at about $83,034, with 29 September as the last completed day ($83,638). It named H-COMPRESSION as a speed hypothesis, not a crash forecast. It said $82–83k had been probed twice and had held on closes. It said two UTC closes above $86k with persistent ETF creations would upgrade the repair path. It said if October through mid-November passed without sequential stress, do not slide the window to December.

This letter does not get to rewrite that. It scores it.

Issue	Frozen price	What we said	Next gate	What happened afterward
27 · 26 Sep	~$83,976	$87.4k failed. $82–83k held. ETF slowing. WAIT.	$86k up / $82k down	28 Sep low $82,510, still closed above $82k
28 · 30 Sep	~$83,034	Wrong clock: setup months, event minutes. $82–83k held twice. Do not predict Oct 10.	Two closes >$86k vs close <$82k then $80k	2 Oct jobs-day wick $87,249→$83,850; 4 Oct close $86,507 on thin Sunday volume; $82k never lost

3. The inherited doors vs this tape

$82–83k: held again. The completed 30 September candle, which Issue 28 could not use, lowed $82,912 and closed $83,556. The 2 October flush lowed $83,850. No completed close below $82k. Observed.

$86k: one completed close above it, 4 October, $86,507. That is not two. Volume was 2,623 BTC against 8,762 BTC on the 2 October rejection day and 14,078 BTC on 21 September. A Sunday close is a close. It is a weak close. Issue 28’s upgrade rule has not printed.

$87.4k: not reclaimed. 2 October’s $87,249 high is a retest of the rejection zone, 148 dollars below 21 September’s $87,397. Distinguishing a sweep from acceptance still matters. This was a sweep.

$80k, $77k, $74–76k, $71–72k, ~$60–63k, ~$57.7k: not visited.

4. Can a shock travel fast enough?

Yes, in this sample. Not at October 2025 speed.

The Bureau of Labor Statistics’ September employment report landed on 2 October: nonfarm payrolls +29,000 versus a Reuters consensus near 90,000; unemployment 4.2%; August revised lower. That is the identifiable unexpected event. Secondary wires said October Fed-hike odds collapsed from a roughly 60% area toward the mid-teens. This freeze did not scrape CME FedWatch live. Treat the odds as secondary and perishable. Reuters instant view via Y94

On Coinbase, the 5-minute high of the day was $87,249.05 at 12:30 UTC. The 5-minute low after that high was $83,850.02 at 18:45 UTC. Elapsed time: 375 minutes. The 12:00 hour opened $86,412 and printed the high. The 18:00 hour printed the low and closed $84,238. The daily candle closed $84,505 — another close that hides the range if you only keep the last print. Coinbase 5-minute tape

Issue 28’s reference event, 10 October 2025, took about thirteen minutes for the last $3,000 of a $107,000 wick, inside an hour that opened and closed near $114,000. Friday was slower, shallower, and two-sided. It still crossed several round numbers before a daily candle settled. H-COMPRESSION is about whether several gates can fail in one continuous event. On 2 October they did not. The shock ran. The first named door held.

5. Was there enough liquidity to absorb it?

In this instance, at the $82–83k door: yes. The flush stopped $1,850 above $82,000. Spot volume on the rejection day was real (8,762 BTC). Weekend follow-through was not a cascade. By 4 October the market was printing above $86k again.

That is not the same as proving depth at $80k, $77k, or $60k. Those books were not tested. Order-book depth, CME basis, and a primary liquidation heatmap remain MISSING. Venue bid/ask at this freeze is a one-cent Coinbase spread; that is a snapshot, not a depth series.

What can be said without the missing plumbing: the near-term bid that Issue 26 named in advance has now absorbed three probes (24 Sep $82,709; 28 Sep $82,510; 2 Oct $83,850). Absorption at one gate is evidence about that gate. It is not a general proof that “there is enough liquidity.”

6. The ETF bid blinked, then came back

Farside, retrieved 5 October: 30 September −$148.7M ended the nine-session inflow streak Issue 28 was watching. FBTC accounted for −$125.6M of that day. 1 October flipped to +$102.7M, with IBIT +$195.6M against FBTC −$60.7M. 2 October’s completed row is +$189.9M (IBIT +$158.2M, FBTC +$29.3M). Some secondary trackers had a provisional +$31.7M for 2 October with IBIT blank; the completed Farside print supersedes that. 3–4 October were a weekend. 5 October’s Farside row is dashes totaling 0.0 — incomplete, not a zero. Farside Bitcoin ETF flow table

Who supplied the ETF bid on the green days, and who supplied Bitcoin into 30 September’s red day, remains MISSING. Cash versus in-kind, AP inventory, and hedge offsets are still not in the packet. The vehicle-level fact is simpler: the bid went red for one session and returned for two. Price on 2 October still failed at $87k while creations were green. ETF absorption divergence has not been retired. It has been interrupted.

7. Leverage is still not the October 2025 machine

OKX BTC-USDT perpetual open interest was $2.36B / 28,478 BTC at the Issue 28 freeze (29 September in that packet). This freeze: 4 October $2.50B / 29,020 BTC. That is a rebuild of about 2% in coin terms from 29 September, and it remains 6.3% below the 20 September peak of 30,981 BTC. Average realized funding stayed small: 0.0040% per eight hours on 2 October, 0.0030% on 4 October, current implied about 0.0069% per eight hours. OKX OI

Deribit BTC options remain call-heavy: call OI about 225,700, put OI about 126,379, put/call 0.56. DVOL was not retrieved (API 400). Dealer gamma was not derived. Fear & Greed is 70 / Greed. CoinGecko dominance 59.23% (Issue 28: 58.28%). DefiLlama pegged-USD about $313.3B versus $313.1B. None of that is a crowded-long liquidation setup of the 9 October 2025 kind.

The current vulnerability, if it exists, is still more macro than crypto-native leverage. Friday was a test of that sentence. The test did not break $82k.

8. The bond market did not get out of the way

Do not say the jobs print “fixed” yields. Yahoo snapshots at this freeze, last regular US session still in the print: 10-year 5.307% (Issue 28: 5.255%), 30-year 5.662% (Issue 28: 5.594%), 5-year 5.073%, 13-week 3.99%, DXY 102.27 (Issue 28: 101.30). Secondary session copy had the 10-year dipping toward 5.16–5.21% on the payrolls miss, then reversing toward 5.28%. This freeze’s TNX is higher than Issue 28, not lower. US 10-year · US 30-year

Equities, last regular close: S&P 500 7,723 (Issue 28: 7,671), Nasdaq 27,191 (Issue 28: 26,798), VIX 16.12, NVDA 233.95. Credit ETFs still soft: HYG 76.91, LQD 101.83, TLT 77.48. Gold 4,179, WTI 89.41, USDJPY 158.3. Cross-asset stress remains uneven: long bonds and the dollar are the hostile tape; equities and Bitcoin bounced. Synchronization is not a crash cluster. High yields are still a hurdle rate. They did not, on Friday, mechanically liquidate the $82k door.

9. H-UP is in play. Acceptance is not.

Issue 28 preserved the possibility that Bitcoin goes up first: confidence, a short squeeze, a sweep of $87k, an appearance of health. The path from $82,510 to $87,249 is that clause on the tape. It may already have fulfilled another upside attempt. Or $90k / $92–96k may still exist as untested liquidity. They were not visited. Two UTC closes above $86k with weekday volume and persistent creations would be the first evidence of acceptance rather than another sweep. We have one Sunday close. We do not have that evidence.

H-NON-ANNIVERSARY still holds. Today is 364 days after the 6 October 2025 Coinbase high. 10 October 2026 is five calendar days away. This letter does not predict it. A quiet 10 October would not kill H-COMPRESSION. A quiet October through mid-November would. Do not move the window if the second thing happens.

10. What we are watching

Node	Status at this freeze	What confirms it	What it is not
~$87.4k	Retested 2 Oct ($87,249); not accepted	Daily closes through it with volume and flow	A new all-time-high door
~$86k	One completed close (4 Oct, Sunday, 2,623 BTC)	A second UTC close, preferably a weekday	Issue 28’s upgrade, yet
~$82–83k	Held on three probes including 2 Oct $83,850	Continued daily-close holds while OI stays orderly	Proof of depth below it
~$80k / $77k	Not visited	—	Scheduled stops
~$60–63k / ~$57.7k	Not visited; 2026 low still $57,718	—	This week’s working test
~$38k	Severe-tail reference only	Sequential failures plus forced selling	A prediction

11. Why this reading could be wrong

The Sunday $86,507 close can be overfit into “reclaim.” Volume argues otherwise. A second close above $86k today or tomorrow would make this issue look late, not clever. ETF creations of +$190M are not the +$999M impulse of 21 September; treating them as equivalent would repeat Issue 27’s slope mistake in reverse. Yields are higher than at the last freeze — the jobs print did not clear the hurdle rate. OKX OI has started to rebuild; “leverage is clean” is less true than it was on 29 September, even if it is still not October 2025. Dominance at 59.2% can be Bitcoin bid or altcoin stress; this packet cannot separate them. Cycle day 364 is how anniversary thinking sneaks back in. Refuse it.

The other error is the bear one: deciding that because Friday was only 3.9% and six hours, H-COMPRESSION is dead. One absorbed shock does not prove the next book is thick. October 2025 was a different machine with more fuel. This week showed the speed half of the question can print without the cascade half.

12. What would kill the leading hypotheses

Kill the immediate repair path: a failure to put a second close above $86k, then a loss of $82k without a 48-hour reclaim. Kill H-COMPRESSION as a 2026 event: October through mid-November passes without sequential stress. If that happens, grade it. Do not slide to December.

Strengthen H-COMPRESSION: a continuous flush through $82k, $80k, and $77k in one session, with ETF outflows and rising liquidation intensity. Strengthen Model A: two UTC closes above $86k, then acceptance through $87.4k, with creations that price actually respects. Wild card remains UNKNOWN. Do not nominate payrolls, after the fact, as “the” wild card. It was a scheduled release. It was still a shock in the print. Those can both be true.

The Controller overlay, if used at all, remains a hypothetical inside a long-range scenario. The 2030 reset is a hypothetical assumption — not observed fact.

13. The experiment

Four paths through 15 December, director synthesis, sum 100. Scenario weights, not tickets, not advice.

Scenario	Weight	Path	Confirm	Invalidate
Model A / bull repair	22%	Second close >$86k, then $87.4k as digestion rather than a cap	Two UTC closes >$86k with ETF persistence, weekday volume, orderly OI	Failure to hold $86k, then close <$82k without a 48-hour reclaim
Base / orderly re-range	48%	Chop $82–87k while yields and the second close argue	Repeated closes inside the box; mixed flows	Two strong closes >$86k, or accepted loss of $82k then $80k
Model B / structural failure	22%	Gates fail, possibly several in one event	Daily close <$82k without reclaim, then $80k/$77k — or a continuous flush	Reclaim $86k with flow and volume
Severe tail	8%	Only after $57.7k fails with forced selling; ~$38k is a reference	Daily <$57.7k with failed reclaim plus pulled ETF/OI/credit stress	Hold of $82–83k into a durable $86k/$87.4k reclaim

Issue 28 was 18 / 48 / 26 / 8. The $82–83k hold through a real shock, plus one close above $86k, nudges repair up and structural failure down. The thin Sunday volume and still-hostile 10-year keep the base as the largest bucket. H-COMPRESSION remains a speed clause inside Model B, not a fifth scenario.

14. One card

An unexpected event can travel through this market in hours. Friday proved that much. Ten minutes is a different machine, and it did not print.

There was enough liquidity, this time, to absorb the first named door. $82–83k has now held three times. $87k has now failed twice. One Sunday close above $86k is a fact. It is not acceptance.

The bid blinked on 30 September and returned on 1–2 October. Leverage is rebuilding off a low base, not crowding. Yields are higher than a week ago. The window is October through mid-November, not 10 October.

Wild card: UNKNOWN.

Decision: WAIT.

Sources

• Coinbase BTC-USD ticker, daily candles, hourly 2 Oct, and 5-minute 2 Oct — spot, completed UTC structure, and the jobs-day wick; freeze 5 October 2026, 13:20 UTC.
• CoinGecko Bitcoin and global — about $86,140, 7d +3.1%, 30d +8.1%, ATH drawdown about −31.7%, dominance 59.23%.
• Farside Investors — 30 Sep −$148.7M; 1 Oct +$102.7M; 2 Oct +$189.9M completed. 5 Oct incomplete.
• OKX open interest and funding history — 4 Oct $2.50B / 29,020 BTC; 20 Sep peak $2.66B / 30,981 BTC.
• Deribit options — put/call about 0.56. DVOL missing this freeze.
• Alternative.me Fear & Greed — 70 / Greed. DefiLlama stablecoins — about $313.3B.
• Yahoo Finance — 10-year, 30-year, DXY, S&P 500, Nasdaq, VIX. Last regular US equity session in the snapshot is 2 October.
• September payrolls +29,000 vs ~90,000 — secondary wrap of the BLS print and the first yield reaction. FedWatch odds not live-scraped.

Educational-not-financial-advice disclaimer. This letter is educational scenario research, not personalized financial advice. It does not recommend buying, selling, leverage, or any position size. Levels are observation points, not instructions. Cash is a valid position when confirmation is absent. Named traders and firms are analytical lenses only. Observable evidence overrides every model, including the Controller. The 2030 reset inside that model is a hypothetical assumption, not observed fact.