Everyone watches the pricealmost nobody watches the level 1.3% under ETH where the market makers switch from damping moves to amplifying them.
Case № 20260905 · 05.09.2026 08:20 UTC · BTC$79,669 / ETH$2,456 · analysis: Fable 5.1 by Anthropic
📅YESTERDAY
The code closed two scenarios from 03.09. "Burn-up through the short fuel" (weighed at 42%) — CONFIRMED: the high printed 81,761 against a 79,432 level.
"The corridor holds" (weighed at 35%) — NOT CONFIRMED: price stayed inside the band only 15% of the term.
Both verdicts describe one event: the overnight push broke the top of the corridor. The three scenarios from 04.09 stay open until tomorrow.
⚡THE GIST
• ETH stands 1.3% above the $2,424gamma flip — below that mark Deribit's market makers stop damping moves and start amplifying them.
• Insurance on exactly that stretch is the cheapest in half a year: ETH two-day vol sits in the 2nd percentile of its own hourly history since February.
• For the first time this week the Hyperliquid perp whales split by asset: in ETH they shifted $57M toward short in a single day — most of the entire weekly shift — while in BTC they're moving the other way.
• Spot BTC ETFs held the week's inflow at the previous week's level, while in ETH the flow thinned almost fourfold — $218M against $824M.
• Retail on Bybit perps got longer over the day: in ETH67% of accounts stand long against 64% yesterday — squarely against the whales.
• 📊 Our Polygon — Pendulum: standing position, −$181 on the day, +$8,641 in total since 11.05.
• 📊 Our Polygon — signals: best — Dust Strategy V2 +$90,719 at a 5% win rate · worst — Wheel Trail V8 −$7,088. Table at the bottom.
🐋WHALES: WHAT CHANGED — three arenas
Arena one — Hyperliquid futures. The top whales still stand mostly short: in BTC the short is 2.3 times the long, in ETH twice. But over the last day, for the first time this week, they parted ways by asset. In BTC they shifted $24M toward long — an even, methodical move with no spike. ETH is the reverse, and the pace is picking up: most of the week's shift toward short landed in the last single day. Overleveraged whales under watch: 56, the same as yesterday; 70% of them stand long against 74% yesterday. None is anywhere near a margin call. There's fuel under the market; nobody has lit it.
Arena two — Deribit options. Two separate whale positions here, and they point in different directions. The first — quiet accumulation in small lots, outside the blocks: in BTC, over the week, more than 17 thousand calls gathered at three neighbouring strikes around the current price. In ETH the book stacks both sides at once — calls above the market and puts below, and the puts together are the bigger pile. The second — today's block trades. In BTC that's a complex multi-leg structure on the September term and an outright purchase of calls above the market. In ETH — a roll of a call position to a later term with a higher strike, and a double purchase of September calls. Strikes, sizes and dates are in the paid part.
Arena three — money at the exchange border. Over seven days $114M crossed into Hyperliquid and $93M crossed out: about $22M more came in than left. But cash and coins move differently. The "bridge" channel — USDC cash — produced a net inflow. The "Unit" channel — the coins themselves — gave BTC back instead: withdrawals ran nearly three times deposits. More than half of all the outflow settled on exchange deposit addresses. Border snapshot as of 04.09.
This is positioning, not a direction forecast.
🧭CROSS-CHECK — five camps of witnesses
What the funds are doing. Spot ETFs are the only camp that measures purchases of the coin itself, not a leveraged position; the flow figures come from SoSoValue. The figures arrive with a lag: funds trade on business days only, and the latest day on record is Friday 04.09. Over the week 31.08–04.09 BTC funds took in $987M against $924M the week before: the flow isn't just holding, it's thickening slightly. ETH is the opposite — $218M against $824M, thinned almost fourfold; SOL's weekly inflow has all but stalled. The BTC buying came above all from BlackRock's IBIT and Fidelity's FBTC. This contradicts the whales in BTC: institutional buying of the coin runs exactly where the whales hold short. In ETH this camp rather confirms the whales.
What the options market says. This camp measures one thing: what insurance costs on the Deribit book. Our fear and greed index stands at 86 out of 100 — the same as yesterday; that's extreme greed. The greed here comes from quiet, not from a rally. ETH two-day vol sits in the 2nd percentile of its own hourly history since February, BTC in the 6th: cheaper than in almost every hour of the past half-year. This contradicts the whales: the whales hold short, and the options market isn't paying for fear of a fall at all.
What the futures say. This camp measures who pays whom to hold a position. BTC funding on Hyperliquid is +0.006% in eight-hour terms against +0.010% yesterday: longs still pay shorts, but less — the pressure from the leveraged crowd is easing (what funding is). The BTC futures basis, meanwhile, sank to −33 from −6 yesterday: the term price slipped noticeably deeper below spot than it was. This confirms the whales: both measures moved over the day in the direction the whales stand.
What the crowd is doing. This camp measures how retail accounts stand against the whales. On Bybit perps ETH retail is 67% long against 64% yesterday — and in BTC retail took the same step toward long over the day. The whales stand mostly short in both assets, so the distance between them and retail widened over the day in both. The gap itself is structural and almost always present — the event here is its motion, not its existence. This confirms the whales: retail moved the opposite way exactly as the whales were building their ETH short.
Any signs of stress. This camp measures whether anyone has already been forced out of the market. Among the top Hyperliquid whales, one forced closure over the day, $2.5M, and it was a long — yesterday there were none. On OKX swaps $550k was liquidated over the day, 97% of it longs; the day supplied only a tenth of the week's volume. This confirms the whales: it was the long side being forced out, not the short one, and a move down did it.
Again: this is positioning, not a direction forecast.
🌡ALTSEASON — the classic and our barometer
The classic altseason index stands at 31.7 out of 100 — bitcoin-season territory: over 90 days most alts have lagged BTC.
Our barometer watches money, not prices, and counts it from the Hyperliquid perp registry. Alts' share of open interest is 43.0% — $4.70B of the whole market's $10.93B.
Hot funding covers 61% of contracts, and retail in the top-10 coins stands 67.1% long: the hottest are DOGE and XRP, both above 76%.
Whales in alts are structurally short: only 35.4% long across $1.30B of positions. The biggest of them is HYPE at $479M.
That gap is permanent; the signal isn't in it but in its CHANGE — and over the week it narrowed, as the whales added long from 32.6%.
Full barometer: indiciadesk.com/en/altseason
⚖️VERDICT OF THE DAY
continuation ⟳ — the overnight push was handed back, and the structure returned to where it stood two days ago: whales short, retail long, insurance cheap.
One new thing over the day, and it matters: the whales stopped acting alike in the two assets. In BTC they're trimming the short methodically; in ETH they're building it faster than earlier in the week. And ETH is the one standing right above the level where the market makers change regime.
Forecasts closed by the code over the day: 2 — one confirmed, one not; three more scenarios in play until tomorrow.
What this does NOT mean. It isn't a claim that ETH goes down. The gamma flip describes how the dealers will behave IF price gets there, not the probability that it does. The Radar records where the risk is densest and whether the sources of measurement agree with each other — not where price goes next.
💎SCENARIOS
The corridor holds — 40% · Burn-up through the short fuel — 30% · Slide into the long fuel — 30%
This is how we weigh the odds — not a promise; trigger and cancellation levels are in the full version.
🔒In today's full RADAR:
— the portrait of the wallet carrying the market's biggest paper gain — an ETH long entered in spring at low leverage, with its size and its distance to a margin call;
— case №546: an ETH strangle in its sixty-second day in play, with legs, term and payoff profile.
ANALYST access → indiciadesk.com/en/agent
💎DEEP
🔒WHALES IN DETAIL 💠
Portraits from the Hyperliquid registry, snapshot 08:16 UTC. Nicknames aren't treated as identification.
The market's biggest gain.ETH LONG $49.0M, leverage only 4×, entry $1,936, uPnL +$10.3M, liquidation at $1,019 — 58% below spot. Account $11.7M. Held for at least a week, and unchanged in size across it. The low leverage is why the position survived the August slide: its cushion is more than half the price.
The market's biggest loss.ETH SHORT $198.0M, leverage 5×, entry $2,221, uPnL −$18.6M, liquidation at $3,650 (+49% from spot). Account $105.3M. Over the week the same address added 27,166 ETH — the short isn't being lifted, it's being built. The same player in BTC. SHORT $171.4M, leverage 5×, entry $72,152, uPnL −$16.0M, liquidation at $124,931 (+57%), 460BTC added over the week. Together the two positions carry $34.6M of unrealised loss — and neither is anywhere near a margin call.
The risky side.BTC LONG $47.9M at 40× leverage, entry $79,368, liquidation at $72,484 — only 9% below spot, on a $6.8M account. The same wallet holds an ETH LONG of $100.3M at 25× with liquidation at $2,345, 4% from price. The fresh player.BTC LONG $39.8M, leverage 20×, entry $79,730, liquidation at $74,534 (−6%). A week ago this position didn't exist: it was opened mid-week and is already slightly underwater.
Block structures in the snapshot. In BTC, on the 11.09 term, a complex multi-leg position of 5,000 contracts: sold 2,000 calls at $84,000, bought 1,000 calls each at $82,000 and $88,000. Another, 1,500 contracts across the 11.09 and 30.10 terms, built on puts in the $62,000–$76,000 band. Separately, on 25.09, an outright purchase of 1,000 calls at $82,000. In ETH — a roll of a call position from the 11.09 term to 25.09 with the strike lifted from $3,000 to $3,500, 6,000 contracts in total; and two identical purchases of 2,500 calls at $2,800 for 18.09. Today's snapshot gives no net premium for these structures — we publish only what's in the data.
Case №948 · ETH condor at $1,700/$1,900/$2,800/$3,000, size 11,500, entered at $2,462, expiry 25.09.2026 — eleventh day in play.
Профіль виплат на експірацію 2026-09-25. Кит отримав $91k премії. Беззбитковість: $1 867 і $2 830.
Case №546 · ETH strangle at $1,000/$4,000, size 7,500, entered at $1,806, expiry 25.06.2027 — sixty-second day in play.
Профіль виплат на експірацію 2027-06-25. Кит заплатив $530k премії. Беззбитковість: $859 і $4 141.
Closed over the day: №1042, BTC bear call spread, thesis "down", price +0.0% — the whale about −$2k; №1033, BTC bull call spread, thesis "up", price +3.9% — the whale broke even. One of the two closed at a loss. ΔOI over the week, both ways:BTC+6,346 calls at $80,000, +6,310 at $82,000 and +4,348 at $79,000. ETH: +23,729 calls at $2,800 against +20,151 puts at $2,400 and +15,541 at $2,200. The BTC book stacks only the upside; ETH stacks both sides at once.
🔬Who exactly
Both of the biggest shorts are Abraxas Capital Mgmt (Heka Funds), two addresses of one house: ETH$198.0M and BTC$171.4M. The third-largest ETH short is Fasanara Capital, $70.8M at 15× leverage, uPnL −$9.1M on a $43.3M account.
Both of the riskiest longs — BTC at 40× and ETH at 25× — are held by Machi Big Brother: $148M combined on a $6.8M account.
A $21.6MBTC short at 40× sits behind an address self-signed "DoshiAtoll"; that's a nickname, not an identification. Its liquidation is $82,340, only 3% above spot: the position closest to triggering in the whole top.
Money by address. The USDC outflow went in tranches to Coinbase deposit addresses — $9.99M, $9.90M and $7.50M; the coin channel sent $6.19M and $5.53M in BTC to Wintermute. Coming in, by contrast: two tranches of $15.0M USDC each from Arbitrum to addresses with no public label.
🔒THE WEEK'S DYNAMICS 🔬
Flows — our registry of Hyperliquid whales; gamma levels and VRP — the Deribit book.
Net flow with acceleration.BTC: $148M toward long over seven days, and the pace inside the window is even — the turn toward long is methodical, with no spike. ETH is the mirror: $82M toward short over the week, but spread very unevenly — the weekly shift came mostly in the last day. So in BTC it's the week's trend; in ETH it's yesterday's decision.
Gamma flip shift.BTC$68,380, up 2,431 over the week; the call wall came down 2,000 over the same period and the put wall rose 2,500 — the walls converged, and the corridor narrowed from both sides (what the flip is). ETH: the flip rose 176 over the week and now sits right under spot.
VRP trend. The premium is −2.0 (IV37.8 against realised HV39.8), and over the week it thinned by 2.2. A week ago the premium was still positive — the regime flipped inside the window: options became cheaper than the market actually moves.
Changing of the guard. Over seven days 180 confirmed structures were collected: 57 bullish against 52 bearish, 20 on volatility and 51 on range. The lean has all but vanished — a week ago the bullish edge was visible. The biggest structure of the window is a risk reversal on 5,000 ETH.
🔒CROSS-CHECK IN DETAIL 💠
The nearest BTC expiry, 06.09. Max pain $80,000 across both books combined (Deribit $80,000 · Bybit $80,000), put wall $79,000 on 207 contracts, call wall $80,000 on 184. Spot stands between them. The term is thin: a day to expiry, and the combined interest on these walls is under four hundred contracts — at that volume walls don't hold price.
The September term, 25.09, carries 43% of all BTCOI: max pain$72,000 (Deribit $72,000 · Bybit $75,000), put wall $70,000 on 8,722 contracts, call wall at the same strike on 10,965. Our study across 362 expiries says an exact match of max pain with spot happens in roughly a third of cases — a weak reference point, not a magnet (what max pain is).
ETH. The 06.09 term — max pain$2,450, put wall $2,200 (2,322 contracts), call wall $2,440 (1,314). The 25.09 term holds 42% of OI with max pain at $2,150 and the densest call wall at $3,000 on 41,294 contracts.
The market-maker corridor is BTC$78,500–$81,000 and ETH$2,400–$2,480 — spot sits inside both, close to the middle.
Skew by tenor. 25-delta skewBTC−1.7, ETH−0.2: in BTC calls are pricier than puts, in ETH the two sides have levelled. BTC put/call — month 0.52, week 0.53, now 0.55: downside protection is being built, yet the money still goes to the upside.
Market-maker stress test ±2%. At −2%BTC runs into the $79,000 put wall and slows; at +2% it passes the $80,000 call wall, where resistance thins. ETH is different: at −2% price breaks through the gamma flip and the regime turns negative — the dealers switch from damping to amplifying. BTC gamma vacuum: $86,000–$88,000 upward and $73,000–$74,000 downward; in ETH the upside is dense, the downside empty at $2,300–$2,350.
💎SCENARIOS IN DETAIL 💠
⏱ 2 days, to 07.09.2026
• The corridor holds — 40%.BTC stays inside the $78,500–$81,000 band, i.e. within the market-maker break-even corridor. Cancellation level: a touch of $82,000. We cut the odds from yesterday's 45% on our own tally — see the review below.
• Burn-up through the short fuel — 30%.BTC touches $81,300 — the entry to the zone holding $29.1M of short fuel. Cancellation level: a touch of $78,000.
• Slide into the long fuel — 30%.BTC touches $78,000 — the upper edge of the zone with $18.2M of long fuel. Cancellation level: a touch of $81,300.
This is how we weigh the odds — not a promise; the code will compute the verdict from spot history.
🔭WHAT TO WATCH
💠 The ETH gamma flip at $2,424 — spot is 1.3% above it, and that's where the dealers' regime changes first; the reference holds until the nearest expiry, 06.09.
🔬 The liquidation price of the riskiest ETH long — $2,345, 4% from price; at 25× leverage that level is reachable within one day of average range.
🔬 The liquidation price of the 40× BTC short — $82,340, only 3% above spot, and squarely inside the short-fuel zone.
🔬 The funds' next trading day will show whether the thinning ETH flow is a one-off or a weekly thing.
🆓TAIL
📊POLYGON
(our own signals at real prices, not a backtest)
Pendulum · Wheel — standing position: −$181 on the day, +$8,641 in total, counted since 11.05.2026.
signal trades win last total since
Dust Strategy V2 395%−$1,125+$90,719 21.03
Dust Strategy V1 329%−$1,125+$35,906 20.03
Volatility Convergence V3 2245%−$507+$14,608 26.05
Skew 2.0 V2 3837%+$315+$11,762 26.05
Fear Flash V2 2133%−$255+$3,440 16.05
Who lost and why. Deepest in the red are Wheel Trail V8 −$7,088, Wheel Trail V6 −$6,081 and Against the Current · top V1 −$5,598. Wheel Trail doesn't lose on a single trade: the whole family — 151 trades at a 34% win rate and −$19,968 combined, with −$12,259 of that collected in the last 30 days. The loss isn't just systematic — it's accelerating. The most profitable signal, Dust Strategy V2, wins only 5% of its 39 trades and closed its last one at a $1,125 loss. Win rate isn't money.
📅REVIEW OF PREVIOUS CONCLUSIONS
(verdicts are computed by code from spot history)
Yesterday the system saw a break: the overnight push removed both bearish hypotheses, and the whales didn't lift their shorts. Over the day BTC gave back −1.4%, price returned inside the corridor, and the stretch between whale positioning and price narrowed on its own. A week ago the conclusion was "no clear tilt", and BTC added 2.7% — neutrality didn't hold then.
Over this day the code closed two scenarios from 03.09. "Burn-up through the short fuel" (weighed at 42%) confirmed: a high of 81,761 against a 79,432 level. "The corridor holds" (weighed at 35%) not confirmed: inside the band 15% of the term, closing at 79,667. The three scenarios from 04.09 remain open.
The tally by name, whole history: "The corridor holds under max pain" — 3 of 5 (+1 partially) · "Burn-up through the short fuel" — 5 of 15, cancelled at the boundary 3 · "Squeeze up into the short fuel" — 4 of 10, cancelled at the boundary 4 · "Slide under the gamma flip" — 4 of 10, cancelled at the boundary 4 · "The corridor holds" — 3 of 11 (+2 partially), cancelled at the boundary 3 · "Slide into the long fuel" — 2 of 5, cancelled at the boundary 2.
Types that don't work. "The corridor holds" confirmed 3 times out of 11 — 27% at odds we set at 35–45%: the system systematically overrates this type, and today we cut it to 40%. "Burn-up through the short fuel" after today's verdict stands at 5 of 15, i.e. 33% — which is exactly why it's weighed at 30% today and not higher. If "The corridor holds" doesn't level out over the next month, the type leaves circulation.
📏Scale calibrator
• What protection costs. Our fear and greed index — 86 out of 100. We compute it from the price of volatility and the option skew, not from news, which is why it diverges from the well-known index.
• Cheap or expensive for this market itself.ETH two-day tenor — 2nd percentile of the hourly history since February, monthly — 15th; BTC two-day — 6th. Cheaper than in almost every hour of the past half-year.
• Which way protection tilts.BTC 25-delta skew−1.7: calls pricier than puts. We don't yet have a reliable historical cut for skew. Terms: indiciadesk.com/en/glossary/
Not investment advice and not a recommendation to trade. Derivatives trading carries a high risk of loss; INDICIA DESK is a research publisher, not an adviser licensed by MAS, the SEC or the FCA — decisions and their consequences are yours.