Bitcoin Weeklyビットコイン週報
Value crossed the 20 line —
16.6 → 23.5.
One call at the top. The evidence under it. The judgement at the end.
Deploy free capital — value is at 23.5 and nothing is flashing danger. This is the band where deployment has been rewarded.
Scheduled contributions continue either way — that never changes. This is about a lump you already have.
Where value has been — 12 months価値の軌跡
Each day ranked against its own history; the current value percentile is 23.5. It peaked at 70.3 on 2025-10-08 and bottomed at 14.6 on 2026-07-01. Weekly points.
Anything unusual?異常
- long-term holders are realising losses on average (SOPR 0.845)
Do anything?行動
- value crossed the 20 line (16.6 → 23.5)
- engine state changed: BEAR_DECLINE → RECOVERY
An item appears here only when a level crosses a boundary, the engine changes state, the sell guard lights, or an input goes stale. Steady readings produce no items on purpose — a report that always finds something will eventually invent something.
The Panel計器盤
Three indicators, three different jobs.
MVRV
the picture, and the buy side
1.294
26.0 percentile ▲ 6.0
Cost basis of the average coin. Below 1.0 the average holder is underwater. This is the buy-side number; do not average it with anything.
Puell multiple
the danger side
0.838
32.1 percentile ▲ 6.0
Miner revenue against its own trend. Best danger marker in the set — 96% of purchases in its top fifth were underwater two years later.
price ÷ 200-week avg
the confluence check
1.064
12.7 percentile ▲ 4.9
Price against its 200-week average. Third leg of the guard; only meaningful when it agrees with the other two.
Sell guard — CLEAR
Nothing in its top fifth. The danger confluence is not lit. Lights only when all three panel members sit in their top fifth at once. In the 2018–2024 sample that combination left 100% of purchases underwater two years later. It is a reason to stop adding and check sizing — never a sell trigger, and it rests on roughly three independent episodes.
If you were deploying capital today資本
The distribution you are stepping into
days since 2018 with MVRV in the 16–36 percentile band · n=548
Historical distribution conditional on today's reading. Not a forecast. It says what this zone has looked like, not what happens next.
The rule that survived every test. Keep buying on schedule regardless of this number. Nine timing strategies were tested against this data and five failed, including every version of "wait for a cheap week."
Use the panel to size a lump you already have — faster below 40, slower above 60 — and never to decide whether a scheduled contribution goes in.
Who is under stress売り手
What the week rules out.
| cohort | level | percentile |
|---|---|---|
| long-term holders | 0.845 | 21.2 |
| short-term holders | 1.0 | 39.0 |
| supply in profit | 0.536 | 22.5 |
SOPR below 1.0 means that cohort is realising losses on average. This is the one thing price cannot tell you: who is doing the selling. Use it to kill stories — if someone claims long-term holders are capitulating, this either shows it or it does not.
Week, month, quarter, year四つの地平
| span | BTC | value then | Δ value | Δ MVRV | Δ Puell |
|---|---|---|---|---|---|
| weeksince 2026-08-13 | +9.4% | 16.6 | +6.9 | +6.0 | +6.0 |
| monthsince 2026-07-21 | +4.3% | 20.1 | +3.4 | +3.3 | +10.2 |
| quartersince 2026-05-21 | −10.5% | 31.9 | −8.4 | −8.2 | +9.0 |
| YTDsince 2026-01-01 | −21.8% | 40.9 | −17.4 | −13.8 | −5.4 |
One week is noise. The same direction across four horizons is the difference between drifting and moving. Δ columns are percentile points.
Where we are in the cycle周期
This drawdown so far
Day 318 since the all-time high.
Completed cycles, for scale
peak to trough · the analogue set, all four of it
| cycle | duration | depth |
|---|---|---|
| 2011-06 → 2011-11 | 162d | −93% |
| 2013-12 → 2015-01 | 406d | −85% |
| 2017-12 → 2018-12 | 364d | −84% |
| 2021-11 → 2022-11 | 378d | −77% |
The Analyst Note所見
Value crossed the 20 line: 16.6 → 23.5
Bitcoin is at $69,400, up 9.4% on the week. Value rose with it, moving from 16.6 to 23.5 and out of deep value into cheap. The regime engine also changed from BEAR_DECLINE to RECOVERY on a fired rule. This is a real change in the evidence, but not a claim that risk has disappeared.
The distinction matters. The BUY call is not stronger because bitcoin became cheaper this week; it did not. The call survives the rebound because value remains cheap at 23.5 and the sell guard remains CLEAR at 0 / 3. Price improved, the state improved, and danger did not light. That is enough to deploy free capital without pretending the turn is confirmed.
A rebound inside a larger drawdown
Across the short horizons, the move is coherent: over the week price is up 9.4%, value 6.9 points, MVRV 6.0, and Puell 6.0. Over the month, price is up 4.3% and value 3.4 points. The broader record is still negative: price is down 10.5% over the quarter and 21.8% year to date, while value is down 8.4 and 17.4 points on those horizons. RECOVERY describes the present move; it does not erase the path into it.
The panel is supportive without being euphoric. MVRV is 1.294 at the 26.0th percentile, Puell is 0.838 at 32.1, and price is 1.064 times its 200-week average at the 12.7th percentile. None is close to the sell guard's danger zone. Their agreement says valuation still offers room; it does not say the first recovery rule must persist.
The unresolved signal is who is still selling
Long-term holders are realising losses on average: LTH-SOPR is 0.845. Short-term holders sit at 1.0, while supply in profit is 0.536. A recovery rule firing while the patient cohort still accepts losses is tension, not confirmation. I read that as evidence that forced or tired selling has not fully cleared, even as price has bounced.
The cycle record does not settle that tension. We are at day 318 since the high, down 44.4%; the low so far was 53.1% below the peak on day 267. Those facts locate the market. They do not tell us whether RECOVERY is durable.
Two branches, with invalidators named now
Branch A — RECOVERY holds. The engine remains in RECOVERY, valuation stays away from danger, and the guard stays clear. A return to BEAR_DECLINE or a lit sell guard invalidates this branch.
Branch B — this is a relief move inside the bear decline. Long-term-holder loss-taking at 0.845 and the negative quarter and year horizons are the evidence for it. LTH-SOPR recovering above 1.0 while RECOVERY holds would invalidate this branch.
One detail underneath the weekly number is worth more than the number. Price has crossed its 200-week average six times since 13 July, and every one of those crosses held the line by a hair before losing it again. This week's is the first that did not: the ratio sits at 1.064 against five prior crosses that never got clear of 1.03. A market that spent five weeks unable to choose a side of that line has, for now, chosen one. That is the real content of a +9.4% week, and it is why the state changed.
I favour Branch A narrowly, and less than the draft of this note first did. The reason I have removed is “the state change was fired rather than carried”. That is a fair argument for most states and a weak one for this state: RECOVERY is the least precise label this engine produces. It fires on materially more days than it should, and it is the state most likely to be sitting on a market that is really doing something else. A fired RECOVERY is close to the weakest confirmation the engine can hand you, so it is not the leg to stand on. What remains supporting Branch A is real but plainer: the 200-week break above, three panel readings still cheap or in deep value, and no danger input unknown.
Two things push the other way, and both belong on the record. The first is the shape of this drawdown against the only four analogues we have. Completed cycles ran 162 to 406 days from peak to trough and took 77% to 93% off the price. This one is at day 267 to its low so far, at 53.1%. On the analogue set it is both shallower and shorter than anything that has completed. That does not mean a lower low is owed — four observations cannot owe anything — but a reader entitled to the bull reading is equally entitled to notice this drawdown has not yet done what the other four did.
The second is a blind spot rather than a signal. Reports are circulating that futures open interest is at or near record levels while volatility stays compressed — the setup that resolves through liquidations rather than through valuation, and it resolves in whichever direction hurts most. We are not in a position to confirm or deny it, and that is the part worth publishing: our derivatives history is weeks long, not years, so the comparison that claim rests on is one this desk cannot make. Note also what the guard does and does not say. It reads CLEAR at 0 / 3, but its members are MVRV, Puell and the 200-week average — all valuation. None of them can see leverage. A clear guard means no valuation danger, not no danger. When the instrument cannot see a risk, the honest move is to name the risk, not to let a clear panel imply an all-clear.
“Narrowly” is doing work here: one weekly transition is not a trend, and it is a transition into the one state I would trust least. The next issue should test persistence, not celebrate a label.
Standing caveat. Four completed cycles is four observations. Historical distributions are context, not forecasts, and the on-chain signal has been measurably decaying cycle over cycle. Nothing here is a reason to change a scheduled contribution.
Read this before trusting any number above. Every input is checked for staleness before it is rendered, and anything older than 14 days prints as UNKNOWN rather than as a confident figure. All inputs fresh this week.
The engine's label is RECOVERY, sourced fired — a rule fired, changing the previous state. Its edge is at naming the present (+19.9pp over price structure on days its rules fire), not at predicting. On-chain signal has been decaying cycle over cycle; treat everything here as context for sizing and for killing stories, not as timing. Generated against the regime engine · value percentile is expanding-window, no look-ahead. The full method, every refuted strategy, and the evidence for this panel are in the engine atlas.