Analysis without illusionsthe whales' walls, and the forecasts of ours that didn't hold.
Case № 20260830 · 30.08.2026 08:18 UTC · BTC$78,146 / ETH$2,458 · analysis: Fable 5 by Anthropic
📅YESTERDAY
The code judged the pair from 27.08 against spot history — and both scenarios closed in the green. "The corridor holds under the ceiling" (weighed at 45%): CONFIRMED — in range 92% of the time, close 77,671. "Burn-up through the short fuel" (32%): CONFIRMED — high 80,496 against the 80,400 level. A rare day for this journal; five more scenarios are still in play — full review at the bottom.
⚡THE GIST
• Leverage came back to the market faster than price: the Hyperliquid registry counts 60 overleveraged whales against 49 yesterday — more dry fuel on a day when price barely moved.
• The fuel grew on both sides, but unevenly: $92.7M of short fuel above the market against $19.6M of long fuel beneath it — the asymmetry compressed from twentyfold yesterday to fivefold (Hyperliquid liquidation map).
• The top-whale BTC book reversed inside its own window: $32M went long over the last day against a short week; ETH keeps adding short — the two assets point opposite ways.
• The cascade stopped and price returned inside the MM break-even corridor: daily liquidations on OKX swaps shrank to 1% of the weekly volume against 29% a day earlier — and it was shorts being closed now, not longs.
• 📊 Our Polygon — Pendulum: standing position, +$288 on the day, +$6,567 in total since 11.05.
• 📊 Our Polygon — signals: best — Dust Strategy V2 +$93k at a 5% win rate over 37 trades · worst — Wheel Trail V8 −$7.1k. Table at the bottom.
🐋WHALES: WHAT CHANGED — three arenas
Futures (Hyperliquid). The weekly flow of the top BTC book is still short — $92M over the week — but over the last day the flow reversed into long, and the book's tilt compressed from $395M to $362M short. A first, still faint sign that the push into short is losing pace.
ETH is the mirror: the weekly flow is short and the last day added more — acceleration, with the book's tilt growing to $205M short. Yesterday the sides stood the other way round — in one day the assets swapped roles.
Fragility: a quarter more overleveraged wallets, their combined position size up to $864.2M against $745.1M yesterday, 62% of them longs — the tilt inside the group has evened out a little.
Options (Deribit). Two positions. Quiet accumulation: over the week, small off-block lots in BTC are being stacked both in calls above the market and in puts beneath it — a two-sided hidden build; in ETH the edge is with the puts. Recognized structures: the off-block tape over two days shows eight in BTC and two in ETH — among the fresh ones sold calls dominate, a position for quiet and for rolling terms to later dates.
Money. We measure the Hyperliquid border over the seven days to 30.08 through two channels: the "bridge" is USDC cash, "Unit" is coins. In total $153.5M came in against $396.6M out — the outflow prevailed, though weaker than a day earlier. The point isn't the net but the split: cash is leaving, with nearly all of the withdrawal landing on exchanges, while coins are coming in — through Unit more than three times as much BTC came in as went out.
This is positioning, not a direction forecast: who stands where, not where price will go.
🧭CROSS-CHECK — four camps of witnesses
What the options market says. This camp measures how much people pay for protection against a move. Our fear and greed index stands at 86 out of 100 — extreme greed for the second day running, well above a week ago. Protection meanwhile is cheaper than the market's actual movement: BTCDVOL (Deribit) 36.9 against realized volatility of 47.9. This contradicts the whales: the top book is short, yet the options desk isn't paying for fear.
What the futures say. Funding shows who pays whom for the tilt in perpetual futures: when longs pay shorts, the crowd is pulling up. The BTC 8-hour funding from the day's snapshot is +0.0007% against +0.0002% yesterday — longs are paying more. This contradicts the whales: the perp shore leans up while the top book stands short.
What the crowd is doing. Retail perp accounts on Bybit are the market's opposite shore. In BTC54% of them are long — the same as yesterday; in ETH65% against 64%. The gap with the short whales is structural; the event would be a move in it — and over the day there is none. This camp is silent: the line-up hasn't changed, no new testimony.
Any signs of stress. We measure stress by forced closures. Among the top Hyperliquid whales the day brought zero, and OKX swaps liquidated just $49k — crumbs against a day earlier, and every position closed was a short. This contradicts the whales: if anything broke over the day, it was shorts — the pressure that crushed longs yesterday is gone.
Three camps of four stand against the top book — and again, this is positioning, not a direction forecast.
🌡ALTSEASON — the classic and our barometer
• the classic: 37.6 against 33.2 yesterday — still bitcoin-season zone, though the broad market pressed in a little;
• our barometer: alts' share of perp open interest 42.6% against 43.9% yesterday — alts gave back share;
• hot funding in alts: 54.5% against 54.8% — the local-long overheat persists;
• retail in the top-10 coins stands 66.0% long while whales in alts (Hyperliquid) are only 32.7% long — the whales are structurally short; the signal is in the CHANGE of the gap, not the gap itself.
Continuation ⟳. Yesterday's break didn't grow into a move: price returned inside the MM break-even corridor $77,000–$80,000 and settled almost on max pain$78,500, the cascade died, and the code closed both open scenarios from 27.08 as confirmed. But the structure underneath is heavier than yesterday: a quarter more overleveraged whales, five times more fuel above the market than below it, and BTC and ETH point opposite ways again — both in fuel and in the direction of the whales' daily flow.
Forecasts over the week (scenario journal, opened 23–29.08): 16 closed — 6 confirmed · 1 partial · 5 not confirmed · 4 cancelled at the boundary; 5 more in play.
What this does NOT mean. Neither that the bounce continues, nor that the whales know the direction. More fuel above the market means a bigger RANGE of upward move for the same push, not a higher PROBABILITY. The cascade stopping means the day had nothing left to break — not that nothing is left to break. And the return of leverage is as consistent with hunting a bounce as with managing margin.
💎SCENARIOS
By the journal: "The corridor holds under max pain" — 45% · "Burn-up through the short fuel" — 33% · "Slide into the long fuel" — 22%.
This is how we weigh the odds — not a promise; trigger and cancellation levels are in the full version.
🔒In today's full RADAR:
• over the last day a BTC short at fortyfold leverage entered the registry with its forced-closure boundary just two percent above the market — the thinnest margin in the whole registry, sitting on an account thirty times smaller than the position itself;
• the structure of the day — a four-leg ETH structure expiring today: the breakdown of all four legs, size and term inside.
ANALYST access → indiciadesk.com/en/agent
💎DEEP
🔒WHALES IN DETAIL 💠
Portrait one — the biggest BTC short in the registry (Hyperliquid).$132.5M at 5× leverage, entry $70,491 — $13.1M underwater unrealized, but the forced-closure boundary is far: $125,883, 61% above the market, on an $83.4M account. It hasn't stood its first week — a strategic side, not a manoeuvre.
Portrait two — the thinnest margin of the registry (the wallet from the teaser). A BTC short of $35.9M at 40×, entry $77,559, unrealized −$306k; liquidation $79,594 — just 2% above the market on a $1.1M account, thirty-odd times smaller than the position. A week ago the position didn't exist — a fresh and very nervous short.
Portrait three — a long on the edge.$40.3M at the same 40×, entry $78,829, liquidation $70,839 — 9% below the market on a $4.6M account: the first candidate for fuel if the market slips into the lower band.
Portrait four — the biggest gain of the registry. An ETH long of $49.2M at a moderate 4×, entry $1,936, unrealized +$10.4M; liquidation $1,014 — 59% below the market. The position appeared only this week — and it's already deep in the green.
Fresh structures (Deribit, over the last day). The registry carries no fresh premiums today — we give terms, sizes and the leg breakdowns. In BTC to 25.09.2026: sold 125× put and 125× call on the single $78,000 strike — 252 contracts of sold quiet: it wins if price stays close to the strike into expiry. Next to it a roll from 25.09 to 30.10: sold a 100× $80,000 call, bought a 100× $79,000 call — the same view moved to a later term and a lower strike. And separately a 150× $79,000 call sold — a position against growth above that level. Structure of the day (ETH) — four legs on 30.08, expiring today. Sold a 500× $2,460 call and a 500× $2,440 put, bought a 500× $2,420 put and a 500× $2,480 call — 2,000 contracts assembled into a narrow financed one-day corridor. The second view is the opposite: on the same 30.08, bought a 700× $2,500 call and a 700× $2,380 put — 1,400 contracts, a position for a strong move either way. Plus a roll of the $3,600 call between 27.11 and 30.10: sold 1,000×, bought 650× — same strike, a different term and a smaller size.
Case №948 · condor ETH to 25.09.2026
Профіль виплат на експірацію 2026-09-25. Кит отримав $91k премії. Беззбитковість: $1 867 і $2 830.
Wings at $1,700/$1,900 below and $2,800/$3,000 above, 11,500 contracts, entered at spot $2,462 — day five in play; a position for price reaching expiry inside the range. Case №546 · strangle ETH to 25.06.2027
Профіль виплат на експірацію 2027-06-25. Кит заплатив $530k премії. Беззбитковість: $859 і $4 141.
A $1,000 put and a $4,000 call, 7,500 contracts, entered at $1,806 — day 56 in play; a structure for a strong move in either direction. Case №7 · risk reversal BTC — the whale exited early
Unwound 28.08 before expiry: an OI trace of −41% with price +25.9% from entry. Case №27 · risk reversal BTC — the whale exited early
Профіль виплат на експірацію 2026-09-25. Кит отримав $3k премії. Беззбитковість: $54 753.
A second identical exit on the same day, with the same trace. Case №975 · bear call spread ETH — settled
Профіль виплат на експірацію 2026-08-28. Кит отримав $325k премії. Беззбитковості в показаному діапазоні немає.
A "down" thesis with price moving −0.2%; by the registry's accounting the whale's result ≈ +$182. Case №976 · bear call spread ETH — settled
Профіль виплат на експірацію 2026-08-28. Кит отримав $238k премії. Беззбитковості в показаному діапазоні немає.
Same type, same day, same thesis: ≈ +$371 on the whale's side.
Quiet accumulation (Deribit, ΔOI over 7 days off-block).BTC: +8,199 calls at $85,000 (80% on 25.09), +7,020 puts at $70,000 (60% on 25.09), +6,897 calls at $82,000 (81% on 04.09). ETH: +17,888 calls at $3,000, +17,438 puts at $2,150 (89% on 25.09), +15,981 puts at $2,300. Small lots outside the block tape — a trace visible only in ΔOI.
🔬For the Analyst — names from the registry. The biggest BTC short from the first portrait belongs to Abraxas Capital Mgmt (Heka Funds): two of their accounts hold BTC shorts of $132.5M and $96.2M and ETH shorts of $131.8M and $99.9M — together more than $61M underwater. Against them Machi Big Brother is long ETH$100.3M at 25× with liquidation at $2,311 — 6% below the market, and Fasanara Capital is short ETH$69.9M at 15× with unrealized −$9.7M. The nicknames VBVIT and MoonpathCliff are self-signed, not identification; the remaining portraits are addresses with no public label. Money: the week's biggest cash withdrawals went to Coinbase Deposit — top operations of $9.99M, $9.78M and $6.85M.
🔒THE WEEK'S DYNAMICS 🔬
Over the week (Hyperliquid × Deribit, our registry): the BTC gamma flip rose $2,536 — to $66,029; ETH's stands at $2,288. The levels crawl up after price — the MMs are ceding space from below, not defending it. The fear premium on the nearest tenor (VRP) went below zero: −0.6 with implied volatility at 36.8 against realized 37.5, and over seven days it compressed by 7.2 — options are getting cheaper relative to the actual movement, not dearer. Changing of the guard: the fresh fortyfold BTC short entered the registry this week, the most profitable ETH long is also this week's newcomer, while the big old shorts sit unmoved. Structures over seven days — 292: 81 bullish, 112 bearish, 29 on volatility, 70 on range — bearish lead the count, but there's no single shared position.
🔒CROSS-CHECK IN DETAIL 💠
The frame by expiry (Deribit and Bybit books, combined). After the weekly expiry the base switched to deeper terms and the aggregate walls moved and thickened: the BTC put wall is now $75,000 (9,097 contracts), the call wall $82,000 (15,142) — a consequence of the change in the calculation base, not whales buying more protection. BTC 31.08:max pain$78,250 · put $75,000 (1,426 contracts) · call $82,000 (176). BTC 25.09:max pain$70,000, with the put and call walls converging on the same $70,000 (7,227 and 11,019 contracts) — September's gravity node. ETH: aggregate walls — put $2,200 (35,992 contracts), call $2,500 (79,004); 31.08:max pain$2,480 · put $2,200 (4,410) · call $2,500 (2,260); 25.09:max pain$2,100 · put $2,100 (38,190) · call $3,000 (40,666). An exact match of price and max pain at expiry happens only about a third of the time — a weak magnet (our review across 350 expiries).
Put/Call drift in BTC: month 0.44 · week 0.59 · now 0.56 — downside protection has been added for a month straight, the daily step is small; ETH: month 0.51 · week 0.56 · now 0.55 — the same, slower. In ETH protection is also cheaper than the actual move: DVOL51.2 against realized 65.0.
The tilt: BTCskew+1.2 against +2.3 yesterday — the payment for downside protection settled by half; ETH−1.2 — calls are dearer there, that desk looks up. Vol percentiles (hourly history since February): the BTC 2-day at the 24th against the 11th yesterday, the weekly at the 19th, the monthly at the 19th; ETH — the 28th against the 5th yesterday, the 18th and the 21st. Protection has started getting dearer, but from a very low base.
The gamma relief: a vacuum below the market at $72,000–73,000 in BTC and $2,300–2,350 in ETH — a break into those bands would accelerate. The MM stress test: at −2% ($76,657) BTC is still inside the positive-gamma zone — a pullback gets damped; at +2% ($79,786) price is still under the call wall — the ceiling holds. ETH's asymmetry is different: at +2% ($2,507) price passes the call wall — resistance thins out, so the same push is damped less on the way up.
💎SCENARIOS IN DETAIL
"The corridor holds under max pain" — 45% · ⏱ two days, to 01.09 · condition: the $76,500–$80,000 range holds · cancellation level: a touch of $75,500.
"Burn-up through the short fuel" — 33% · ⏱ two days, to 01.09 · trigger: a touch of $79,800 — entry into the $79.8–82.1k short-fuel band · cancellation level: a touch of $76,500.
"Slide into the long fuel" — 22% · ⏱ two days, to 01.09 · trigger: a touch of $76,600 — the top of the $74.3–76.7k long-fuel band, with the $72–73k gamma vacuum beneath it · cancellation level: a touch of $80,000.
The code will compute the verdict from spot history; this is how we weigh the odds — not a promise.
🔭WHAT TO WATCH
• the 31.08 expiry — tomorrow: max painBTC$78,250 and ETH$2,480 on the combined Deribit and Bybit books get tested first;
• a touch of $79,800 — entry into the short-fuel band (scenario two);
• a touch of $76,600 — the top of the long-fuel band, with the $72–73k vacuum beneath it (scenario three);
• the BTC gamma flip at $66,029 — far away, but its break is what changes the market's regime first (Analyst);
• in ETH the fuel lies below the market: $18.3M of long fuel in the $2,335–2,408 band against just $318k of short fuel above — the mirror of BTC (Analyst).
🆓TAIL
📊 POLYGON(our own signals at real prices, not a backtest)
Pendulum · Wheel — standing position: +$288 on the day · +$6,567 in total · counted since 11.05.2026.
signal trades win rate last total since
Dust Strategy V2 375%−$1,125+$92,969 21.03
Dust Strategy V1 3110%−$1,125+$37,031 20.03
Volatility Convergence V3 2148%−$535+$15,116 26.05
Skew 2.0 V2 3537%−$259+$12,470 26.05
Fear Flash V2 2133%−$255+$3,440 16.05
Deepest in the red: Wheel Trail V8 −$7,088 · Against the Current · top V1 −$6,189 · Wheel Trail V6 −$6,081 — many trades without the single rare big win that rescues the table's leaders. The most profitable — Dust Strategy V2 — wins only 5% of its 37 trades. Win rate isn't money.
📅REVIEW OF PREVIOUS CONCLUSIONS(verdicts are computed by code from spot history)
The pair from 27.08 reached its resolution, and both scenarios closed in the green. "The corridor holds under the ceiling" (weighed at 45%): CONFIRMED — in range 92% of the time, close 77,671. "Burn-up through the short fuel" (32%): CONFIRMED — high 80,496 against the 80,400 level.
Five remain open: the pair from 28.08 and yesterday's trio.
The score by name across the journal's whole history: "The corridor holds" — 3 of 10 (+2 partial), 3 cancelled at the boundary · "Squeeze up into the short fuel" — 4 of 10, 4 cancelled at the boundary · "Slide under the gamma flip" — 4 of 10, 4 cancelled at the boundary · "Burn-up through the short fuel" — 4 of 9, 1 cancelled at the boundary. The weakest type is "The corridor holds": 3 confirmed of 10 — and that denominator is exactly why the same type gets 45% today rather than 70%.
📏Scale calibrator
• What protection costs. Our fear and greed index — 86 out of 100: extreme greed, unchanged on the day; a week ago it stood at 73. 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. The BTC 2-day vol climbed to the 24th percentile of our hourly history since February from the 11th yesterday, the ETH 2-day to the 28th from the 5th: protection is getting dearer, yet it's still cheaper than in three quarters of that history.
• Which way protection tilts.BTCskew+1.2 — puts are still dearer than calls, but the payment for downside protection is half of what it was a day ago. 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.