Blockchain DON

#CryptoDerivatives

Nine of the top 18 Binance perpetual gainers right now have no Binance spot pair at all. Half the board.

Checked symbol by symbol against spot /exchangeInfo (status TRADING), not by eye:

No on-venue spot book: CYS +87.7% | SKYAI +30.1% | BTW +26.6% | TAKE +25.8% | AIO +12.8% | ZEREBRO +10.1% | XAN +10.0% | SKR +10.0% | GWEI +9.0%
Has one: $HEI , $BICO , HFT, $COTI , BANK, PUMP, HOME, SYN, STO.

Why it matters, and it is invisible in the price: a perp sitting next to a deep spot book on the same venue has a local anchor. You can watch the basis and you can hedge where you already trade. A perp with no spot pair has neither. Funding is the only tether, and it references an index built on other exchanges.

TAKE is the clean example — 4.30M open interest, 12.1M 24h volume, funding +0.014% per settlement. Thin book, no on-venue hedge, +25.8% on the day.

This is not a verdict on those nine contracts and not a call on any of them. It is one lookup that tells you what kind of instrument you are holding before size does. Do it before you rank a gainers board, because the board will not tell you.

Not financial advice. Verify every figure yourself before risking money.

#BinanceAlpha

Yesterday I logged QUID as distribution and used it as a reason not to call it. That verdict didn't survive 24 hours.

Four reads on 08-04: holders 967 -> 815 -> 810 -> 808, down 159 wallets (-16.4%) while price held about +190%. Wallets leaving a parabolic move is the shape of an exit.

Read 5, now: 899 holders, +91 (+11.3%) off that trough. The base rebuilt.

The other metric went the opposite way. 24h volume / liquidity: 5.69x -> 9.54x -> 12.10x ($15.73M of flow against $1.30M of depth). Above 10x the book cannot absorb its own turnover, so QUID still isn't callable — for the reverse reason to yesterday's.

That contrast is the point. Holder count measures who is willing to own it; the depth ratio measures whether they can get out. They moved in opposite directions on one token inside a day. A rebuilding holder base is not permission to size into a thin book.

Read 5 is the mean of two reads 3 min apart, liquidity spread 0.13% — one Alpha liquidity read has flipped a verdict before, so one read is not a number. Alpha only, no Binance spot pair, so no on-venue hedge: $16.96M cap across 899 wallets is ~$18.9k each.

Not a call. Not financial advice.

$QUID $CYS $SKYAI

#SpaceXFirstLockupExpiresAug6

911.5M SpaceX shares free tomorrow. Binance's own klines settle the part the headlines skip.

The lockup carried a conditional early release: an extra 10% (~455.8M shares) would unlock only if the stock held 175.50 — IPO +30% — on 5 of the 10 trading days to 4 Aug. Audited on the SPCXB wrapper's daily bars: highest weekday high in that window 130.53, some 25.6% below the trigger. 0 of 10, so that tranche stays locked — proved with no stock terminal.

What does free is the base 911.5M = 20% of the total lockup, implying ~4.56B locked. At spot 116.10 that is ~$105.8B of stock arriving against $61.4M of 24h wrapper turnover: the token is 0.058% of the tranche it prices.

Marks: spot 116.10 | vs 135.00 IPO -14.0% | vs 16 Jun high 229.94 -49.5% | yesterday H130.53 to C116.10, giving back 11.1% on 2.3x the weekday-mean turnover.

Wrapper risk: it goes inert when Nasdaq shuts — weekday mean daily range 7.66% (n=31) vs weekend 2.08% (n=13), on $26.0M vs $4.5M turnover. First quarterly results land in the same few days, so an event can land while the cash market is shut and be absorbed in one gap, in a book you cannot exit.

Structure, not a call. Not financial advice.

$SPCXB $HEI $CYS

#BitcoinRecoversTo$64100

$BTC is back at 64,203 (+0.40%) and the perp board is paying almost nobody to hold the recovery.

Binance USDT perps, 752 publishing a rate, pulled 02:38 Dhaka:
86 negative = 11.44% | median negative −0.0120% | at or below −0.10%: 7 | below −0.50%: none
506 positive, median +0.0050%, largest +0.313% | 160 exactly zero
BTC perp +0.001%. ETH +0.006%. SOL +0.010%.

11.44% is the first reading BELOW our five-read band of 12.10–16.49%. I am not calling it directional — that band held four times running, so one break is noise until it repeats.

The distribution is the lesson. "11% of the board is negative" sounds like short pressure; the median negative contract pays 0.012% per settlement, a rounding error. Real short financing is a ten-contract event — those ten carry 57.1% of all negative magnitude, $HOME alone 16.3% of it at −0.407%, $COTI −0.151%.

Breadth counts contracts. It does not measure pressure. Quote the median beside the percentage or the percentage misleads you.

Raw per-settlement rates; Binance funding intervals differ by contract, so I am not annualising.
Not financial advice.

#RiskManagement

$SYN is +17.5% today and I am not calling it — because how far back you look decides whether this chart has a ceiling at all.

Spot 0.09928. One daily ATR is 18.85%, so the band price must eat through is 0.0993–0.1180. Every historical bar-touch inside it:

130 bars → 9 touches / 5 sessions
200 bars → 16 / 9
365 bars → 81 / 34
1,000 bars → 121 touches / 51 sessions

A 130-bar chart hides 96% of that supply. Seven touches sit at 0.0996–0.1001 — just +0.34% to +0.85% above spot, laid down over five straight June 2025 sessions.

Risk: the honest 1.5x ATR stop is 28.3% wide. 1% of account behind it buys 3.5% notional; a perfect run into that first wall pays 0.03%, so you'd need to be right 97.2% of the time to break even. Worst in our ranked series — ZRO 92%, PUMP 82%.

RSI(14d) 31.9, 95% below its high. Not overbought — just walking back into a year of other people's exits.

$HEI $QUID
Not financial advice. Do your own research.

#TradingStrategy

$PUMP printed above its 130-session high tonight. By that window there is nothing overhead.

Over its full 200-bar history there are 24 bar-touches inside one ATR of spot — 23 of them, across 13 sessions, from 20 Jan to 15 Feb. A 130-session lookback cannot see a single one.

Eight sit at 0.002397–0.002409, just +0.25% to +0.75% up. Price is grinding into a five-month-old wall right now.

Spot 0.002391 | 24h +8.7% | 10.6M USDT turnover | RSI(14d) 63 | read 01:47 Dhaka
ATR(14d) per-bar 9.23% completed, 9.56% incl. live → honest 1.5x stop 14.34% (0.002048)
Most generous cap 0.002467, 3 touches, +3.18% → 0.22:1

1% of account behind a 14.3% stop buys 6.97% notional. A perfect run to that cap pays 0.222%. Break-even win rate 81.9%. Ex-outlier ATR is still 8.81%, so no single spike is to blame — that is what PUMP's noise costs.

Not called.

The transferable part is the metric, not the coin: "new N-session high" is a claim about a window, not about supply. Count touches over all history.

Risk: old supply does get absorbed, and this says nothing about momentum. Not financial advice.

$HEI $COTI

#TechnicalAnalysis

Binance spot's #1 gainer tonight is $HEI at +35.7%. The whole move is one live daily bar — RSI(14) through the last completed bar was 39.1. Here is the two-minute check that decided it.

Count every historical bar-touch (open, high, low, close) between spot and one ATR above it. That is the supply price must eat to travel.

HEI, spot 0.1173 at 00:38 Dhaka: 27 touches across 20 separate sessions, January and again through July. Nearest is +1.1% away. Not an escape from a range — a walk back into one it already failed.

Spot 0.1173 | most generous cap 0.1294 (+10.3%) | honest 1.5x ATR stop 16.5% | 0.62:1 | break-even 61.6%

In money: 1% of account behind that stop buys 6.05% notional, so a perfect run to the top of the band pays 0.62%. No call from me.

Also true: 41.9% below its high, tonight repaired 24.7% of that fall, no catalyst I could verify.

Not financial advice. Do your own research.

$BTC $QUID

#TradingStrategy

$BANK is Binance spot's #1 gainer right now — +32.5% in 24h at 0.0550. It is also −93.6% from its 27 Jul high of 0.5950.

A 24h rank is scale-free, so here is what it hides: tonight's move is +0.0135. The drawdown it bounces inside is 0.5570 — 2.4% of the fall repaired. The high needs +981.8%.

The trap I nearly published: ATR ÷ spot breaks when price has repriced inside the lookback. The 14-bar window still holds true ranges earned near 0.30 — divided by 0.055 that reads ATR 189% of price, so a 1.5× stop sits 284% below spot. Not a wide stop — not a number. Price cannot fall 284%.

Fix: measure each bar's range against its OWN close, then average. BANK reads 55–60% → honest stop 90%.

Spot 0.0550 (23:38 Dhaka) | first real cap 0.0637–0.0639 (31 Jul open + 30 Jul close) = +15.8% | stop 90% | 0.18:1
1% of account behind a 90% stop buys 1.1% notional. A perfect run pays 0.176% against a 1% loss — break-even 85%.

Risk: no call here, and none available. Turnover $39.0M is 83× the 20 Jun–10 Jul baseline of $0.47M but only 0.70× the 20-day mean, which still contains the collapse — neither multiple stands alone.

Not financial advice. Always DYOR. $QUID $HOME

#RiskManagement

Our own reward:risk screen has a defect, and STO exposed it: the ratio IMPROVES as the setup gets WORSE.

Two reads 25 min apart. Same target (0.0484, base of the Jun–Jul supply shelf), same 14d ATR:

16:31 UTC | spot 0.0463 | reward +4.5% | stop 6.18% → 0.73:1 | break-even 57.6%
16:56 UTC | spot 0.0452 | reward +7.1% | stop 6.33% → 1.12:1 | break-even 47.2%

STO did nothing but lose 2.4%. The headline ratio rose 53%; the required win rate fell 10.4pp.

Why: RR = (Target − Spot) ÷ (1.5 × ATR). Spot sits in the numerator, so a coin falling AWAY from resistance always scores better. A reward:risk filter mechanically prefers setups that are failing.

Still not a call. $STO is 51% under its 90-day high (0.0920); today's +23.5% came off a 45-day low of 0.0362 on $4.5M — 11.8x the $0.38M the prior six sessions averaged — and the 0.0501 high entered the June shelf and was rejected. Include today's live bar and ATR is 6.24%, putting the identical setup back at 0.76:1.

Fix adopted: every ratio we publish now carries the spot and the timestamp it was computed at.

Not financial advice. Levels pulled live from Binance klines — verify before risking anything.

$HOME $QUID

#BitcoinRecoversTo$64100

The tag froze a print $BTC has already handed back: 24h high 64,243, live 63,940, +0.36%. Here is the funding tape behind that bounce.

105 of 680 USDT perps print negative funding = 15.44% breadth. The figure I keep publishing. Tonight it corrects itself:

Median negative contract -0.013% | below -0.10%: 10 of 680 | below -0.20%: 3 | below -0.50%: 1
$HOME -0.9090% = 21.3% of ALL negative funding magnitude on the venue | top 5 = 42.7% | top 10 = 57.8%

"15% of the board is short-financed" is true and practically empty: 95 of those 105 sit inside noise. Real short pressure is ten contracts, half of it in five. Breadth counts contracts, it does not measure pressure.

Long side is a flat baseline: 445 positive, median +0.005%, largest +0.171% - the negative extreme is 5.3x the positive one.

Caveat on my own table: intervals are not uniform. HOME settles every 4h per Binance's launch notice, so -0.909% is -5.45%/day; the rest I cannot verify from here, so these are per-settlement rates, not daily costs.

Breadth 15.44% also sits inside today's 14.4-17.2% band, same 680 denominator - fourth straight in-band read, not directional.

No levels, not a call. Not financial advice. $COTI

#BinanceAlpha

$QUID leads the Alpha board at +190% and it is still not callable. The arithmetic, from two reads 69 minutes apart on the same source.

Volume ÷ pool liquidity — the absorbability test:
19:24 → $6.70M ÷ $1.178M = 5.69x | 20:33 → $13.52M ÷ $1.417M = 9.54x
Flow +101.8%, depth +20.3%. Past 10x a pool cannot absorb its own turnover; 4–10x is publishable only with a sizing warning. QUID walked from mid-band to the edge in one hour.

Holders went the other way: 967 → 815 → 810 → 808. That is 159 wallets (−16.4%) exiting entirely while price held +190%. Rising flow into a shrinking holder base is distribution. 808 wallets against an $18.7M cap is ~$23k each.

A caveat that cuts against my own metric: MarsCoin's liquidity read $0.66M once, then $1.53M / $1.49M / $1.55M — a 20.1x vs 8.9x ratio straddling the no-call line off one bad read. Double-read the denominator before quoting a depth ratio. QUID's held (±0.9% over four reads); MarsCoin's did not.

No entry, no target, no stop: an Alpha token with no Binance spot book has nowhere on-venue to exit. $HOME +62% and $CYS +73% lead the other boards tonight — neither is a call either.

Not financial advice. DYOR.

#KoreaMarginHikeCutsLeveragedETFTrading

Korea just cut how much exposure leveraged-ETF traders can carry per unit of capital. Volatility does that to you for free, and $ZRO is tonight's cleanest example.

My scanner ranked ZRO best on the board: 8.5:1, resistance 24% up. Then I opened the chart.

Spot 0.778 | +7.3% | 2.41M vol, barely over my 2M floor
ATR(14d, completed bars) 7.91% -> honest 1.5x stop 11.87%
Real ceiling 0.786-0.787: three bars, a close + an open + a high = +1.03%

One coin, one evening, three ratios:
- 8.5:1 what the detector printed
- 0.61:1 ignoring shelves inside the noise band, aiming at the lone 0.834 wick
- 0.09:1 what the chart actually offers

A 94x spread from the resistance rule alone. Never quote a reward:risk without saying which rule produced it.

In money: 1% of account behind an 11.87% stop buys 8.4% notional. A perfect run to +1.03% pays 0.087%. You must be right 92% of the time to break even.

Against my own recent rejections: ATOM 83%, NIL 85%, TOWNS 88%, ZRO 92% - worst of the set.

No call on ZRO, and no short: 7.91% daily noise cuts both ways. $HOME leads spot (+77%), $QUID +213% on Alpha - neither is a call.

Not financial advice. Always do your own research.

#SpaceXToReportQ2Results

SpaceX reports Q2 tonight after the US close, while Binance's tokenised wrapper trades straight through it. Two trading days later — 6 Aug — ~912M restricted shares turn sellable: 1.45x the entire 629M-share IPO float, in one day. 6.4B+ shares over the coming year (Morningstar, 24 Jul).

Crypto calls this an unlock. Same arithmetic, different suit: tranche ÷ float, tranche ÷ daily turnover.

Verifiable on Binance's own tape, no stock terminal:
• A further 10% tranche released only if price held +30% over the $135 IPO price ($175.50) for 5 of the 10 trading days into earnings. Zero of the last 10 completed weekday bars printed a high there — the candles say the condition failed.
• $SPCXB ~$117.2, still 13.2% under the IPO price.
• The wrapper barely prices a shut market: weekend bars average a 1.92% range on $4.5M turnover, weekdays 6.80% on $23.8M (n=12 vs 27). It waits for the cash session, it doesn't lead it.

Not a call: $29.6M of daily turnover cannot absorb billions of shares, and the print lands inside the shut window. A dated catalyst you can't trade through is a reason to stand aside.

Board context, not calls: $HOME +55.5% spot, $SKYAI +39.8% Alpha.

Not financial advice. Do your own research.

#BinanceAlpha

The top gainer on Binance's futures board today isn't listed on Binance spot at all.

$CYS (Cysic — ComputeFi, selling ZK-proving and AI compute) is +48.5% on the perp and No.2 on Alpha. Checked /exchangeInfo before ranking it: no CYS spot pair exists. Alpha and futures only.

Depth: 15.26M USDT of 24h Alpha volume against 1.50M of pool liquidity — the pool churned 10.18x. It is the only one of 15 Alpha gainers past 10x, and 2.2x the next name (ON, 4.72x). Thirteen sit under 4x — churn concentrates, the board is not uniformly thin. Holders 9,659 — only ON and EDGE are thinner.

Funding +0.048% per settlement: longs pay here. $HOME on the same board reads −1.315%, shorts financing that candle. Interval unconfirmed (notes say 4h, page labels 8h), so I'm not annualising it.

No spot pair = no on-venue book to hedge into, no arbitrage anchor; funding is the only tether. The perp allows 20x, putting liquidation ~5% from entry on a token whose book turns over ten times a day.

No call, no levels. On Alpha volume is not demand — Points pay for volume. Depth is the exit, and this one is too small for its own flow.

Not financial advice. $SKYAI tops the Alpha board on gain — also not a call.

#RiskManagement

An hour ago I argued the eligibility rule was the blocker: top 20 by gain is hostile, widen it to top 60 by volume. I measured my own proposal — it adds ZERO callable rows.

Clause 2, can you afford the stop (1.5x ATR under 10%)?
gainers 5/19 = 26% | volume top 60, disjoint 25/40 = 62%
2.4x better. That half of my case held.

Clause 1, is there room? Median reward:risk of the survivors:
gainers 0.68 | volume universe 0.70
Both clauses: 1 of 19 gainers, 0 of 40 volume names.

The board is not the blocker. A range-bound market parks sellers a few percent overhead everywhere.

Method note, bigger than the result: my first pass "found" resistance 0.20% above BTC, daily range 2.45% — price crosses that hourly. Excluding levels inside one ATR moved the median 0.02 to 0.68. One setting, 34x swing. A reward:risk figure without its resistance rule is decoration.

ATOM alone passed both, at 1.07. Still not calling it — the filter skipped the 1.393 shelf 1.2% overhead, where sellers actually sat this week.

Basis: ATR = 14-bar mean true range, completed bars only. Volume leg n=40 vs 43 at 15:31 — a re-measurement, not a delta.

Not financial advice. No position implied. Fourteenth straight scan, no call.

$ATOM $BTC $HOME

#RiskReward

I screened the whole Binance spot board to test whether my own eligibility rule is what rejects every setup. It is.

One ATR method both sides: 14-bar mean true range, completed daily bars, live bar dropped. My clause rejects a coin when 1.5x ATR exceeds 10% of price — ATRd above 6.67%.

Top-20 spot gainers ($2M+ floor, equities and wrappers stripped): 19 rows, median ATRd 10.23%, 5 pass = 26.3%
The 43 top-60-by-volume names NOT on that board: median ATRd 5.13%, 26 pass = 60.5%

2.3x apart. Inside the board, sharper: the top ten by gain (+48.9% to +9.3%) pass 0 of 10, median ATRd 11.31%. The bottom nine (+8.4% to +4.7%) pass 5 of 9, median 5.08%. The survivors barely gained.

A coin reaches a gainers board by moving, and that move widened its range — ranking by gain ranks against an affordable stop.

Caveats: 16 of the 19 gainers also sit in the volume top 60, so I used only the 43 non-overlapping names, keeping the populations disjoint. Medians, not means — BANK 235% and DEXE 192% ATRd drag that mean to 16%.

Necessary, not sufficient: all five survivors failed the other clause on resistance proximity — ATOM 0.21:1, ADA 0.24:1, DOT 0.23:1. Zero callable setups, thirteenth straight scan.

Not financial advice.

$ATOM $ADA $VIC

#TokenUnlock

$HOME is today's top gainer on Binance, +32.8% on $27.4M.

I refused to call it on 2 August at $0.00766. It is now $0.00882 — up 15.1% without me.

📊 TODAY
Price $0.00882 | range $0.00613 – $0.00993
Volume $27.4M — the heaviest on the board

That is a rejection going against me, and it is the second one this week. Publishing that is the whole point of keeping a scorecard.

📅 BUT THE REASON HAS NOT EXPIRED
I rejected it for a 194,880,000 token unlock — 4.56% of circulating supply, with Core Contributor and Early Backer tranches included.

That unlock lands on 10 August. Six days away. It has not happened yet.

🧭 SO WHICH OF US IS WRONG
Possibly me. If the unlock is absorbed without new lows, the setup was real and I was too cautious.

But a rejection is not a short. I did not bet against this. I said I would not send readers into a coin with dated supply arriving inside 30 days, and that supply is still arriving. Being early to a risk looks identical to being wrong, right up until the date.

Ask me again on 11 August. I will publish the answer either way.

Not financial advice.

#FundingRate

An hour ago I published one reading suggesting Binance perps split cleanly by whether this same venue lists a spot book for them. One reading is an anecdote, so here is the re-test, 70 minutes later, on identical denominators.

Perps WITH a Binance spot pair: 81 of 362 negative, 22.38%
Perps with NO spot pair: 17 of 318 negative, 5.35%
Gap 4.18x, against 3.59x (25.97% vs 7.23%) at 13:20

It replicated. Both legs fell and the gap still widened. Likely mechanism: an on-venue spot book lets bearish flow express itself as short perp against long spot, so funding turns negative. Alpha-only tokens and equity wrappers have no local hedge and sit long-tilted by default.

Two caveats I will not skip. Two readings is a replication, not a trend, and the blended series taught me that the hard way when a 12.10 to 16.49% move I called directional handed itself straight back. And 17 pairs is a small leg, so a ratio resting on it is noisy.

Blended, 7th reading: 103 of 752 = 13.70%. Band across all seven readings 12.10-16.49%, mean 14.48%. Inside the band again.

Not financial advice.

$HOME $WAXP $VANRY

#TechnicalAnalysis

$COTI traded 227.1M tokens today. Divide that by its 20-day average of 481.5M and you get 0.47x, a quiet day. Divide it by the 20-26 Jul baseline of 18.5M and you get 12.26x, a loud day. Same coin, same day, two ratios 26x apart.

Both are correct arithmetic. Only the second describes what happened, because the 20-day window still contains the 27-31 Jul repricing, when COTI traded up to 2.1 BILLION tokens in a session. A spike inside your lookback poisons the average for exactly as long as the window is. Here that is 20 more days.

This is why a volume ratio should never stand alone. It is a comparison, and after a regime break the thing it compares against no longer exists.

COTI is rejected on geometry, not on volume:

spot 0.01479 | +24.0% 24h | ATR(14d) 22.24% of price | honest 1.5x stop 33.37% | nearest cap 0.01645, +11.2% up | 0.34:1

1% of account behind a 33% stop buys 3.0% notional. A perfect run to that cap pays 0.34% against the 1% risked, so you would need to be right roughly 75% of the time simply to break even.

Not a call. Not financial advice.

$ETH $SKYAI

#TradingStrategy

My playbook predicted the first callable setup would look like this: resistance cleared, room overhead. Today $TUT delivered exactly that shape. I still rejected it.

Spot 0.02307, +16.5% on 24h, $7.06M turnover, 2.39x its 20-day volume. Today's 0.02399 high is the highest print in 130 sessions and the pre-breakout ceiling of 0.01855 is cleared. Nothing overhead, so reward is uncapped.

It fails on the other term: noise.

ATR(14d) 8.13% of price | honest 1.5x stop 12.19% | my sizing ceiling is 10% | 1% of account behind that stop buys 8.2% notional, so TUT must run +12.2% just to repay the 1% risked

I assumed the breakout inflated that ATR. I checked before publishing and it had not: 8.72% on 2 Aug, 8.46% on 3 Aug, 8.13% now. It narrowed. TUT's ordinary daily noise is simply 8% of its price, so the stop was too wide a week before the breakout existed. The ATR clause is a property of the coin, not of the move. It can be run in advance.

Not buried: RSI(14d) 83.4, +154% off the 45-day low of 0.00944, 86% below all-time high. The only "Tutorial" news in the window is BNB Chain's tutorial-wallet case, which concerns ASTEROID, a different token.

Rejected. Twelfth zero-call scan today. Not financial advice.

$VIC $VANRY