Orbit Post Sitemap
The rooftop at 2 a.m. is so cold! After waiting a whole day for CPI data, all the results pointed to negative news, but the market reversed and taught everyone a lesson!
Just two days ago, I wrote that $ETH is the weakest among the three major coins, but tonight it directly teaches me how to be human.
Core CPI rose 0.3% month-on-month to beat expectations, and the probability of a rate hike in September soared to 90%. Despite textbook negative news, $ETH surged 6.44% in 24 hours to 2,582, hitting a new high since the end of January!
BTC rose just over one point, while the ETH/BTC exchange rate jumped 3.9% in a single day.
The logic is simple: The market has never feared negative news but uncertainty. Walsh's knife hung for half a month, but once the boot hit, people dared to get back to work. The three major US stock indices all rose over 1% for the same reason.
ETH's biggest rally has its own reasons: in the 48 hours before September 7, $300 million was withdrawn from exchanges; spot ETFs saw net inflows over 1 billion in two weeks, and the previous week saw $24.3 million in capital outflows, turning the supply side into a full-blown increase.
The 2,530 resistance has been broken, and now 2,700 is the target, but between 2,723 and 2,822, tens of millions of ETH are trapped inside, and the rate hike on September 16 is still a real threat. If you want to chase the high, plan your fallback first! The stockholders are already numb...
#美国CPI环比加速, expectations for interest rate hikes are heating up LIQUIDITY IS NOT FOLLOWING PRICE
$ETH gained 3.34%, yet generated 640T USDT in trading value — nearly matching $BTC at 606T, while $SOL reached only 123T. This suggests the market isn’t short on capital; money is being used to rotate positions.
$BTC remains below MA20
$SOL recovered to $102
$ETH holds above $2,500.
Hidden signal: High volume without a strong breakout may mean the market is absorbing selling pressure, not chasing FOMO.
The question: Who is quietly accumulating here? 5u challenge 10000u
Day three
Funds continue to grow
Also reached 67u
After the good news of #美国CPI环比加速，加息预期升温
immediately went long near $ZEC 1097
Stop loss near 1083, successfully caught a wave
Longed $SNDK near 1691 and also made some profit
Overall, not bad 😌
$BTC was stopped out, missed the opportunity
No worries 🙂 opportunities come every day
Just afraid of no capital
Surviving is more important than anything‼️
#BTC现货ETF连续流出
#财报观察员：甲骨文AI云收入增121% This is not a rebound; this is like CPR for my short position account, right? Last night at dawn while watching the market, $ARB was hovering at a high level. I noticed the volume didn't keep up, and the resistance above was tight. Every time it tried to surge, it fell short of breath, so I signaled to open a short. Many were afraid it would surge again, but I just said don't catch a falling knife at the top; wait for confirmation of resistance before acting.
The market waits for the right moment, and profits come from holding.
Later, it was pressed down from 0.19556 all the way to 0.14145, with a +1383.97% profit on the short position giving a direct answer. That profit was satisfying; those on board must have woken up smiling. The earlier hesitation was real, but the outcome is truly sweet.
Profits are given by the market; holding on makes them yours.
I first closed 80%, pocketing the bulk, and moved the remaining 20% to break-even for protection; if it continues to drop, let the profits run, and if it rebounds, don't give the profits back. Don't be greedy for the last bit; take profits when you should.
Being out of position is not a sin; opening random positions is the mistake.
For friends who haven't entered yet, listen to me: now is not the time to rush in. Chasing shorts risks getting caught in a rebound squeeze. Wait for the next signal; there are still opportunities, so don't be anxious.
$ZEC $LAB $XRP in 24 hours +1.07% versus BTC +0.39% — difference +0.69 p.p.
With a position of 35% within the daily range, the question is simple: is this real relative strength or is the movement already fading? The same exchange, the same $BTC contract, but users in two locations see different liquidation prices.
This is not speculation; someone compared screenshots from both sides. Mechanistically, it makes sense: matching depth, liquidation engine, and market making quotes are separated by region, so prices naturally diverge.
The affected users are CN users using leverage; even if the direction is correct, they might still get liquidated locally. Spot users are not directly hit by this, but after liquidity is fragmented, slippage will ultimately be borne by everyone.
Verification is simple: monitor the mark price and funding rate at the same time in both regions. If they remain consistent over the long term, it means my judgment is wrong.
#BTC现货ETF连续流出
#伊朗允许BTC与USDT外贸结算 #加密财库分化：买币还是回购？ $BTC Long and Short Crowding List
Crowding is not about being bullish or bearish; the key is which side has higher costs and the price still can't move.
$SNDK current rate +0.0319%, settled +0.007% in the past 24 hours, at the 96th percentile of recent samples. Price and OI both increased over 15 minutes, indicating market heat is transmitting to position expansion. The high positive rate combined with rising positions is still driven by the long side; watch how much each additional position can bring in price gains.
$CNPY current rate -0.0219%, settled -0.443% in the past 24 hours, at the 33rd percentile of recent samples. The 15-minute decline is accompanied by risk exposure contraction; first observe the speed of position reduction, do not mistake it for new short positions. Position reduction has already occurred; next, see if the price can stabilize after position contraction.
$ZEC current rate -0.0078%, settled -0.035% in the past 24 hours, at the 3rd percentile of recent samples. The 15-minute decline and position reduction clearly indicate position exit and deleveraging. When holdings decrease, extreme rates may quickly revert; currently, it is more suitable to observe deleveraging rather than chase direction.I no longer directly infer price increases from "high TPS and many active addresses" when looking at SOL.
What truly determines valuation is Token Value Accrual.
The continuous growth of Solana's on-chain transactions, DEX volume, and app revenue only proves strong network demand, but the value generated by these economic activities is diverted by protocol revenue, validator earnings, MEV, and market maker profits, and does not all settle into $SOL.
Therefore, I focus more on three sets of data:
First, Real Economic Value (REV) and fee revenue;
Second, the net supply change between SOL's burn, staking demand, and new issuance;
Third, whether on-chain growth can form a sustained SOL demand sink.
If network revenue grows but token capture does not increase accordingly, even high on-chain activity may only mean "ecosystem prosperity with stagnant coin price."
So my current assessment of SOL's fundamentals is not about "how busy Solana is."
Rather:
How much of the network-generated cash flow ultimately flows back into SOL's valuation system.
This is the fundamental that truly matters for trading.Bitcoin's recent move has been thrilling! As soon as the CPI was released, it first dropped to $76,046, then surged wildly by $3,255, directly hitting $79,301, a 24-hour increase of 2.68%. The most outrageous part is that the core CPI was still on the hot side, yet it sharply reversed in a V-shape!
Why? Because what the market fears most isn't bad news, but "uncertainty." Before the release, the probability difference for a rate hike was as high as 15 percentage points, but once the data came out, the tension eased. Plus, with a long-short ratio of 1.114 and balanced positions, there was basically no forced selling pressure; as shorts covered, the price took off. The drop below $76,046 was clearly a fake fall, and the price recovered stronger and stronger.
But don't get too excited yet! Between $80,000 and $82,000, about 8% of Bitcoin supply is stacked, with the ETF cost line and the 50-week moving average at $81,081 all clustered there. Last week, it just touched $82,284 before pulling back. Historically, after a golden cross, the four previous times saw a rise followed by a correction. A 2.68% daily gain isn't enough to break through this wall!
What’s even more painful is that the spot demand indicator has dropped back to -145,000 BTC, Coinbase premium is -0.036, and ETF net outflows hit $308 million in a single day, the worst in two months. This rebound looks more like short covering plus uncertainty digestion, not a return of real demand. $ZEC $SOL $SNDK
#美国CPI环比加速，加息预期升温 #BTC现货ETF连续流出 #OKX预言家：来星球玩预测
With ETFs halted over the weekend, liquidity is thin, oil prices have surged to 109, and tensions in the Middle East continue. The "Clear Act" vote is on September 15, and the Federal Reserve decision on September 16. Brothers, is this a real breakout or a false peak? Let's discuss in the comments—do you dare to chase it? If ETH's zero-knowledge proofs rely solely on cloud providers, it can't be considered truly complete.
In the second quarter, the Ethereum Foundation funded several local multi-GPU proving system projects, with one goal being to enable L1 block proofs to run on proprietary hardware and establish open operational protocols.
Why should $ETH holders care about this? Because even if zero-knowledge proofs are technologically advanced, if their actual operation depends only on a few large cloud platforms, the network might just be replacing the original verification bottleneck with a new infrastructure centralization.
Local deployment allows testing stability across different hardware environments and helps more teams understand the real computational power, bandwidth, and maintenance capabilities required by the proving system.
This doesn't mean every household needs to line up GPUs. The key point is that proof generation cannot rely on a single supply chain, nor can the entire system lose capability if one cloud service goes down.
Decentralization doesn't require everyone to do exactly the same work, but that critical functions have multiple independently operable paths.
If $ETH increasingly depends on proof verification in the future, then whether the proving infrastructure has competition and redundancy will become a security issue. No matter how elegant the model is, it must be tested under real machine stress.Good morning everyone, last night's market was somewhat unexpected.
The rise was reasonable, but the rapid short squeeze followed by a quick rally really surprised me.
Some analysts jumped out saying it was due to accelerating CPI and rising expectations of interest rate hikes.
So I want to ask: based on the rapid rally after the short squeeze yesterday, is this good news or bad news?
Then some people said: that's the bad news fully priced in, the market has already digested it in advance.
So, what's the use of analyzing the news?
Before 8 PM, the market had actually already made things clear.
1. BTC/USDT Perpetual $BTC
BTC is oscillating within a range; we mostly use 1-hour and 15-minute charts to analyze the market.
On the 15-minute chart, it's clear that BTC was previously declining along a trendline, then broke above the trendline.
The key point is that the price pulled back to test the trendline, proving its support is valid.
At this point, we should mainly be bullish.
I was surprised that the price actually formed a wick here (I had no idea what the CPI was).
Now if the price breaks this support level again, what should we do?
Mainly be bullish; only consider shorting if it breaks below the bottom of this consolidation zone.
2. ETH/USDT Perpetual $ETH
ETH reacted most violently last night. I shorted below 2550, but it unexpectedly broke above 2600.
I then stopped out.
Next, we wait for more consolidation in price.
After the market frenzy, it will calm down; we patiently wait for the next opportunity.Don't just focus on Pons. StonkFun has surpassed Pons in daily revenue for two consecutive days (1,355,000 vs 1,210,000), following the #Solana path by spreading buyback pressure across the top 10 ecosystem tokens.
Pons spends 80% of its fees buying back its own token, which is more concentrated but also more vulnerable. Do you prefer the concentrated buyback of #Pons or the diversified ecosystem approach of StonkFun? The Top Dark Horse of BTCFi Crashes: 69 Million Ghost Tokens + Lack of Transparency = ?
⚠️ This article is based on publicly available on-chain information and does not constitute any investment advice.
This bull market has seen the emergence of many star projects in the BTCFi sector, and $CORE was once hailed by many as the top dark horse in the field. With Satoshi-Plus hybrid consensus, binding to Bitcoin’s hash power, and a hard cap of 2.1 billion tokens mirroring BTC, this highly attractive narrative drew in a large number of retail investors and capital bets. Everyone believed that with the backing of Bitcoin’s hash power and the total supply locked by the whitepaper, this project was sufficiently secure. Until the validator reward vulnerability exploded on 8.31, this dark horse revealed a huge hidden risk: 69 million ghost tokens looming overhead, combined with a lack of transparency during a major event, severely damaging the project’s narrative.
The root cause of the vulnerability was a code defect in the protocol’s reward distribution module, which allowed a few malicious validator nodes to repeatedly claim block rewards. Within just three days, 255 million CORE tokens were mined prematurely in one go. These tokens were originally node rewards meant to be slowly released over decades on an 81-year cycle. The project team repeatedly emphasized that the incident did not breach the 2.1 billion total supply cap and no new tokens were minted out of thin air. However, the total supply cap is only the ultimate ceiling; the token release schedule was completely out of control, representing a typical overspending issuance, and the tokenomics written in the whitepaper were broken by a code bug.
After the crisis broke out, the project urgently launched a v1.0.26 hard fork, adopting a forward upgrade approach without rolling back historical transactions, so ordinary users’ holdings would not be wiped out. The hard fork destroyed and reclaimed 186 million abnormal tokens on-chain, restoring the total supply on the ledger to 2.1 billion. But the hard fork could not solve a fatal leftover problem: about 69 million abnormal tokens had already been transferred out of the reward pool into external wallets before the fork execution, making them unrecoverable through on-chain upgrades—these are the so-called ghost tokens in the market. This portion of tokens is unaffected by destruction and remains floating above the market, posing a constant risk of a dump.
What has sparked ongoing controversy in the market is the issue of information transparency after the incident. To this day, the community continues to ask how long the vulnerability had been lurking, the full list of involved validator nodes, the address distribution and complete transaction trails of the 69 million ghost tokens. The project team has only issued brief announcements and has yet to release a full technical post-mortem report. After a major security incident, avoiding core issues and obscuring information is a key reason institutional funds remain cautious and hesitate to enter on a large scale.
Many retail investors previously held a misconception: as long as there is Bitcoin hash power backing, the entire public chain is unbreakable. The truth is, Bitcoin hash power only secures the underlying hash layer against 51% attacks; token reward distribution and node verification rules belong to the upper-layer application code. No matter how strong the underlying hash power is, if there is a bug in the upper-layer code, the token release rules will fail. CORE’s crash this time brutally shattered the illusion of hash power omnipotence.
Looking back at CORE’s roadmap, it planned LST liquidity staking, SatPay payments, and asset management protocols, aiming to generate real business revenue from ecosystem fees and use profits to buy back tokens, building a positive value flywheel. But reality is harsh: the current ecosystem fee volume is very small, insufficient to offset the selling pressure from token releases. Price increases rely more on staking incentives than real business profits. After the vulnerability incident, multiple exchanges immediately suspended CORE deposits and withdrawals. Although trading later resumed, on-chain yield staking products were delisted and the project’s risk rating was raised—this is the market’s most direct risk warning.
Objectively, CORE’s code is open source and the on-chain ledger is verifiable, with no multi-level referral rebates like traditional Ponzi schemes, fundamentally different from such scams. But not being a Ponzi scheme does not mean there is no huge investment risk. Overspending issuance, leftover ghost tokens, insufficient disclosure of major event information, and upper-layer code vulnerabilities are all risks holders must face.
Other projects in the same sector like STX and MERL have not experienced major consensus-layer security incidents; their audits and governance disclosures are more transparent, and incremental bull market funds clearly tilt toward these projects. A hard fork can only fix numbers on the ledger; the trust investors lost is hard to rebuild quickly with a single technical upgrade. To regain market recognition, the project team must release a complete security post-mortem report and publicly disclose on-chain tracking data of the abnormal tokens.
CORE’s crash sounds a warning bell for all investors in the BTCFi sector. When choosing a public chain, don’t be fooled solely by hash power narratives or scarcity of total supply. Code security, token release schedule, and project information transparency are the three core evaluation criteria. No matter how good the sector is, once tokenomics get out of control and information becomes a black box, even the brightest dark horse may bury huge risks.Finally got tricked again
This ETH chart got me a bit hyped
Bought more after breaking 2600
Didn't expect the sudden drop like this
Retail investors are crying
Looking at the chart
ETH surged to 2667
I thought it was about to take off
Went long right away
But then
It dropped back to 2511
Down over 150 points
Account value shrank immediately
This move
Is a classic case of chasing highs and getting trapped
The 5-minute chart is so deceptive
It pulled from 2432 to 2667
Up over 200 points
Looks like a breakout
MA5, MA10, MA20 all trending up
Who would have thought
That was the main force baiting longs
Once you chase in
They slam the price down
Left hanging at the peak
ETH is now at 2511
MA5(2513), MA10(2514), MA20(2513)
All three moving averages pressing from above
Short-term trend has turned bearish
If BTC can't hold tonight
ETH might probe lower
2450 or even 2400
Don't rush to add positions
If it rebounds above 2550
Reduce or close positions first
Don't stubbornly hold on
Chasing highs and getting stuck
Is a common retail investor mistake
I made this mistake too
Remember next time
Chasing longs after breaking resistance
Is the main force's favorite trap to harvest $ETH $BTC $ZEC
#美国CPI环比加速，加息预期升温 #财报观察员：甲骨文AI云收入增121% #美国CPI环比加速，加息预期升温
US August CPI data shows inflation pressure rebounding:
CPI MoM: +0.4% (July +0.1%)
📈 CPI YoY: +3.4%
🔥 Core CPI MoM: +0.3%, higher than market expectation of +0.2%
Core logic:
1️⃣ Energy prices are the main driver
Oil prices have climbed back near $100, gasoline price increases are pushing overall inflation up.
2️⃣ Fed rate cut expectations cool down, rate hike probability rises
With CPI combined with previously strong PPI and employment data, the market is repricing, and the probability of a Fed rate hike in September has clearly increased.
3️⃣ Market impact
📉 Short-term negative for risk assets
US growth stocks: financing costs rise, valuations under pressure
Crypto market: USD liquidity tightens, short-term volatility increases
Gold: theoretically under pressure, but safe-haven demand may offset some pressure
📈 Beneficiary directions
US Dollar Index ↑
US Treasury yields ↑
Energy sector ↑
My view:
The biggest change in this CPI is not "inflation out of control," but the market switching from a "rate cut trade" back to a "high interest rate maintenance or even rate hike trade."
For the crypto market:
⚠️ Short-term focus on USD and US Treasury yield trends.
If the 10-year Treasury yield continues to rise, BTC and ETH may face pressure.
But if:
✅ CPI negative impact is priced in early by the market
✅ US Treasury yields peak and fall back
✅ ETF funds flow back in $IOST Last night I was still calculating if this month's instant noodle money was enough, and this morning I was already thinking about whether to add sausage.
Before going to bed last night, I glanced at IOST, and after the pullback, it surprisingly held steady, with obvious signs of bottom-fishing below. I only left one sentence at the time: if the pullback doesn't break, you can try going long. After going long, I went from 0.0007429 to 0.0009031, +216.17%. After coming out, I felt comfortable, brothers.
Don't lose patience in the oscillation and then try to regain dignity in a one-sided move. Risk control done upfront is called rational; cutting losses later is called a brave decision.
This round, take profit at 70%, protect the remaining 30% at cost price. Take profits when you should, don't be greedy for the last bit. Hold if the trend is intact, run if it breaks, don't fall in love with the candlestick.
For friends who haven't gotten on board yet, listen to me: chasing highs easily leaves you stuck at the peak, there will be more opportunities later. Wait for the next shot, and see when the new structure emerges.
$ETH $ADA On-chain anomalies are analyzed by address behavior, not news. In the past four hours, passive buy orders have continuously appeared in the 0.0780 to 0.0800 range; individual orders are small but frequent, indicating that a giant whale is suppressing costs to accumulate. The 0.0835 to 0.0850 range above is a dense area of previous trapped positions, and there is currently no sign of a volume breakout.
Just parked the car under the shade and took a bite of bread, and at the 0.0790 level, support orders appeared again. The naked K-line price tested 0.0775 twice but was quickly pulled back, raising the low point, which shows funds are defending this position firmly. Funding rates are slightly negative, shorts are crowded, and a short squeeze could happen at any time.
Operationally, the current price is 0.0801800, entry range is 0.0785 to 0.0805, with a stop loss at 0.0768—if it breaks below, admit the mistake. Take profit is first targeted at 0.0845, and if it holds above, push to 0.0870. Do not chase highs; re-enter on pullbacks.
$LAB
#伊朗允许BTC与USDT外贸结算
@OKX星球 $BCH / $BTC / $ETH
I’m watching these three for different reasons.
$BCH → momentum
Bitcoin Cash is catching attention with a strong move today.
But I don't automatically chase a big green candle.
A strong move tells me to investigate.
It doesn't tell me to enter.
$BTC → direction
Bitcoin remains my first filter.
If BTC is weak, I become more careful with everything else.
If BTC starts building strength again, I’m more comfortable looking for opportunities.
$ETH → rotation
Ethereum is the one I watch for signs that capital is willing to move beyond Bitcoin.
That's what makes this combination interesting.
BCH is showing momentum.
BTC tells me about the broader market.
ETH tells me whether the market is willing to take more risk.
Three different signals.
One market.
I don't need to catch every move.
I just want to recognize the strongest setups before I decide whether they deserve my capital.
#USCPIReignitesHikeOdds #OutcomesOnOrbit The row for Friday isn't complete yet, but some have already taken "continuous outflow" as a conclusion.
What I trust is whether the form is fully filled out. The total for the first three days of Farside is about −449.5 (−46.6 on the 8th, −120.2 on the 9th, −282.7 on the 10th) which is already nailed down; on the 11th Sep, the IBIT on the row is still a dash, and the page total has just appeared as 6.0 (MSBT +3.8, HODL +2.2).
So this layer is not about reporting who is bearing the bulk—on the fourth trading day, don't treat an incomplete row as final. OKX BTC spot is about 77282. #BTC现货ETF连续流出 $BTC $ZEC
ZEC surged from 480 to 1299 in one month, and now the funding rate has turned negative.
Let's lay out the numbers first, no hype:
In the past 30 days, ZEC rose steadily from 480.41 to 1299.00, with a maximum increase of 170%. Currently at 1159.41, it stands at the 82.9% percentile of the 30-day range.
24h change +6.75%, high 1218, low 1053.77 — daily volatility 15.6%. Trading volume 1.73 billion USDT, open interest 125 million USDT.
But the real interesting part is not the rise, but the changing structure:
① Funding rate turned negative, currently -0.00631% (annualized about -6.9%), perpetual premium is also negative, spot-futures basis -0.093%. To translate: perpetual contracts are starting to trade at a discount, indicating long leverage is retreating and shorts are willing to pay to enter.
② Long-short account ratio is 0.47, with significantly more short accounts than long accounts. This is a signal worth watching in a coin that doubled in a month.
③ Two-way liquidations. This morning from 07:18 to 07:19 shorts were liquidated at 1165–1168, and at 06:52 and 07:42 longs were liquidated at 1153–1155. Both sides are being harvested, indicating huge divergence at this level. Brothers, putting together the recent CPI, rate hike expectations, US Treasury yields, oil prices, US stocks, and the actual movements of $BTC and $ETH, my personal conclusion is:
We still can't say the bear market is completely over, nor can we directly declare that a full bull market has arrived. More accurately, the market is in a "post-bear market recovery and new trend confirmation phase."
There are indeed many bearish factors: CPI shows inflationary pressure still exists, and the market's expectation for a Fed rate hike in September has clearly intensified, which is always a pressure on risk assets.
But on the other hand, the most notable thing this time is: despite such strong bearish expectations, the market did not collapse continuously and directly; both BTC and ETH have shown clear support. This indicates that funds have not completely withdrawn.
So I think it looks more like:
❌ Not a confirmed major bull market yet
❌ Nor a traditional one-sided bear market
✅ More like a large-scale bottoming with a tug-of-war between bulls and bears and a trend selection.
What will truly decide the bull or bear market going forward is still Fed policy. If the market can withstand after the rate hike is implemented, and even see a "bearish realization rally," that would be very positive for the medium-term trend; but if rate hike expectations continue to rise and US Treasury yields keep climbing, the market still faces the risk of another pullback.
#美国CPI环比加速，加息预期升温 #BTC现货ETF大额流入后转负 Something unusual is definitely going on, brothers!
I knew there had to be something fishy with ETH pumping so high. This short trade was perfect, and there's more dumping to come!
---
💰 Last night’s short was so satisfying
Asset: ETHUSDT
Direction: Short
Entry price: 2,603.46
Exit price: 2,550.1
Leverage: 20x
Return: +39.01%
---
📉 Market signals
ETH surged to 2,667 then quickly dropped, a classic spike-and-dump. Now around 2,513, with MA5 (2,514) and MA10 (2,513) flattening, and MA20 (2,523) pressing down overhead. After the spike and drop, short-term likely to continue retracing to 2,450-2,480 for support. With rising rate hike expectations and accelerating month-over-month CPI, macro pressure remains unresolved. ETH’s independent rally has been overextended.
---
📌 What’s next?
· Short again if it rebounds to 2,550-2,580 and stalls
· Stop loss at 2,620
· Target first 2,450, if broken then 2,400
When something’s off, there’s always a catch. The higher it pumps, the harder it will dump. More dumping to come!
$ETH $BTC
#美国CPI环比加速，加息预期升温
#财报观察员：甲骨文AI云收入增121% Long and Short Crowding List
Look at the rate to see who is paying, and look at the price and position to see if the money has brought direction.
$CNPY Current rate -0.0285%, settled -0.443% in the past 24 hours, at the 33rd percentile of recent samples. Price rises while expanding positions, the market has new positions supporting it, but open interest alone cannot determine the long or short side. It is not a crowded position yet; direction judgment depends on price and position response.
$SNDK Current rate +0.0241%, settled +0.007% in the past 24 hours, at the 93rd percentile of recent samples. Price is rising, and positions are also increasing, short-term funds are expanding risk exposure. High positive rate combined with rising positions, bulls still control the rhythm; when price response weakens, reassessment is needed.
$ZEC Current rate -0.0063%, settled -0.035% in the past 24 hours, at the 4th percentile of recent samples. Price rise is not accompanied by position reduction, new positions have participated, but continuation depends on subsequent price response. The short side is still paying fees, price and open interest are rising simultaneously; what can be confirmed now is that shorts are under pressure, the short squeeze is not yet complete.The bulls took a hit but got back up themselves: $ETH key levels and scenario for today
Wow, an hour ago $ETH was hammered down 0.72% in 15 minutes, then quickly bounced back to 2515. I lean towards buying the dip, volume ratio 2.13 caught the dip.
In short—oil price dropped combined with rising US Treasury yields, profit-taking and leverage liquidation resonated to push it down. When it falls, it gets bought up.
Bullish logic: first, MA7 has been pressing MA30 for 22 days, daily ADX at 54 indicates a strong trend; second, RSI at 57.8 is strong but not overbought; third, external support with BTC at 77288 only moving +0.616%.
Resistance above: 2523 (this morning's high) → 2537 (SAR resistance)
Support below: 2511 (this morning's low) → 2411 (4h SAR support)
Watershed level: 2511. Holding this supports the dip-buying scenario; breaking below means retreat first, then consider buying again at 2433.
Conclusion: More likely to see consolidation digesting crowded longs—account ratio 2.39, fee 0.0001/8h, leverage is not cheap. Scenario key levels: break above 2537 with volume to target 2600; break below 2511 wait for 2433 to 2450 before acting.
Strategy—buy dips above 2511 and take profits at 2523; if it breaks down, don't wait for a rebound, exit immediately.
I’ve prepared the opening scenario in advance, follow it to stay on track.
$ETH $BTC$AERO This profit makes me feel both anxious and cautious, afraid that the market will react tomorrow and blacklist me.
Just after lunch while watching the market, AERO's rebound was weak, with heavy signs of a bull trap, volume didn't keep up, and the upside wasn't that easy. I knew the short logic hadn't changed, a reminder not to chase the longs, wait for it to fall back on its own, high-level pressure is no joke.
From 0.6409 down to 0.5708, +219.06%, nailed it, can treat myself to a good meal, this gain feels comfortable, the earlier grind was worth it, timing was spot on.
For stocks you're not confident in, a glance keeps you sober, buying a lot is foolish. Don't lose patience in the choppy market, then try to regain dignity in a one-sided move.
First close 80%, keep 20% at cost price for protection, if it continues to drop let the profits run. Take profits when you should, brother, watch your gains.
For friends who haven't gotten in yet, listen to me, chasing highs easily leaves you stuck at the peak. Wait for a more comfortable position in the next round, patiently await good news.
$ADA $SOL CPI again exceeded expectations, and the market immediately speculated on rate hikes, yet $BTC and $ETH strengthened, which seems contradictory at first glance.
But the market does not trade based on headlines. Prices reflect advance pricing, position distribution, and expectation gaps—the truly important factor is "how much unknown remains."
CPI rose 0.4% month-over-month and 3.4% year-over-year, which is not insignificant. However, detailed items show that energy contributed the most, with gasoline prices up over 25% year-over-year; core inflation is still slowly declining. Therefore, this data did not bring new policy signals. The Fed not acting in September was already fully priced in, and CPI alone is unlikely to change that.
There are substantial changes in the funding side.
BTC ETFs have seen net inflows for three consecutive weeks, totaling $3.8 billion, the strongest period this year; a single day inflow of $730 million is the third highest this year. Behind this is allocation capital, not just slogans.
For ETH, Robinhood’s L2 has been live for two months, with daily revenue surpassing $1 million and trading volume reaching $1 billion, showing signs of ecological activity.
Derivatives have also completed a round of deleveraging. Open interest first declined then rose, currently exceeding $9.6 billion, above the 180-day average. Compared to leverage-driven rallies, this structure is more solid.
On the interest rate front, Citi has pushed back rate cut expectations to June 2026. The unemployment rate has not worsened significantly, and the labor market remains stable, so "no rate cuts" have long been accepted. The 10-year US Treasury yield broke above 4.80%, and the 2-year is at 4.42%. Such a combination would usually suppress crypto, but this time it caused almost no reaction.
Conclusion: The fundamental pricing that should be set is complete.$ARB ARB large holders on the long side are collectively trapped, while shorts have already locked in floating profits.
The price is below the main long holders' cost; when market sentiment warms up, a corrective rebound will follow, with a bullish outlook.
Contract risk is extremely high, set stop losses, and avoid further downside risk #美国CPI环比加速，加息预期升温 #OutcomesOnOrbit #BTCETFFlipsNeg Everyone is watching $BTC. I'm watching the US bond market. Because if the Treasury continues to send stress signals, the most important question might no longer be: “Will Bitcoin breakout?” But rather: “WHO WILL BUY THE MASSIVE US DEBT — AND WHAT INTEREST RATE WILL THEY DEMAND?” This could be one of the most important financial questions for crypto in the next phase. The US has a very simple but extremely difficult problem to solve: Large debt. ↓ Needs to borrow more. ↓ Most contracts will not be affected, but a small remaining portion may hold real money
After replaying historical mainnet transactions, the Ethereum Foundation found that Glamsterdam's new Gas rules will not affect the vast majority of contracts. Some transactions may fail due to insufficient original Gas limits, but raising the limit can restore them; a few contracts may face more serious issues.
"Most have no problems" is certainly good news, but it should not be a reason to ignore the remaining contracts. Code on Ethereum is interconnected; an old contract may be relied upon by multiple applications and may hold assets that cannot be easily migrated.
This is also why historical replay before upgrades is important. Testing is not to prove the solution is definitely correct, but to find edge cases that only appear under real transaction conditions as much as possible.
For $ETH holders, exposing problems during the development phase is not scary. Problems being made public, located, and fixed in advance actually shows that the security process is working.
The real danger is maintaining an optimistic narrative by interpreting "a small portion affected" as "no need to worry at all." In financial infrastructure, even a small proportion of vulnerabilities can correspond to large amounts of money.Why is $82,000 difficult to break through? Perhaps some clues can be seen from the chip structure.
First, the chips held by short-term holders (STH) are distributed directly between $59k-$81k (red in Figure 1). Reaching $82,000 means all STH are in profit.
Some short-term speculative funds will choose to take profits, which is the first layer of selling pressure.
Second, although the chips held by long-term holders (LTH) are distributed across the entire price axis, the most concentrated chip peak is exactly at $81k-$82k (blue in Figure 1).
This part of LTH may not truly be believers; some are just passively holding long-term chips after being trapped by buying.
When the price approaches breakeven, they choose to exit. This is the second layer of selling pressure.
What’s worse, this is also the concentration area of super whales.
Except for two locations near $40k, holders with more than 100k BTC are all between $78k-$82k.
Whoever has the strongest fist calls the shots.
Therefore, $82,000 indeed has resistance in the short term, which is beyond doubt. The market needs time to digest differences and supply.
But! We can also understand it the other way around:
When the market regroups and the next time it can break through this peak in one go, ahead will be a smooth plain...As the weekend approaches, $DOGE bulls are trying to push the winning streak to the third week, but the real variable lies in tonight. At 20:30 Beijing time, the US August CPI will be released, with the market expecting a year-on-year rate of about 3.4%. This is also the last inflation data before the Federal Reserve's policy meeting on September 16, making the timing very tight. The gains in the past two weeks were mostly driven by retail and sentiment funds. With liquidity thin over the weekend and institutions absent, this type of capital comes quickly and leaves quickly, indicating a warming speculative appetite but an unstable foundation; the continuous rise has consumed some buying power, so the third week requires new funds to take over. Last week's nonfarm payrolls increased by 162,000, showing strong employment and the emergence of rate hike expectations. If the CPI is higher than expected, the dollar and US Treasury yields will rise, suppressing risk assets, and $DOGE's weekend momentum will most likely be interrupted; if the data is softer, liquidity expectations will improve, giving sentiment funds a reason to re-enter. Therefore, the keyword for this week is not "continuation" but "confirmation." In the first few hours after the data release, trading volume and price direction will provide the answer. The direction can be predicted, but positions still need to leave room.The Treasury's long-term bond repurchase on Wednesday night went through: a cap of 6 billion USD, but only about 5.19 billion in face value was actually accepted, not fully subscribed. The result? On Friday, the 10-year US Treasury yield surged to 4.974%, and the 30-year yield stood above 5.37%, both hitting multi-year highs. The market complained that this was like using a small kettle to put out a big fire, and Bassett directly retorted: "The people in front of the Bloomberg terminal are unhappy with me, but there's nothing I can do."
Adding Friday's CPI: overall month-on-month 0.4%, year-on-year 3.4% met expectations, but the core month-on-month 0.3% exceeded expectations, gasoline rose 3.9% adding fuel to the fire, and the Middle East situation is pushing oil prices up. The market is already raising the probability of a Fed rate hike in September. The Treasury wants to suppress yields, but inflation and oil prices are pulling in the opposite direction, a tug-of-war between the two.
On the $BTC side, after the CPI release, it first dipped to 76,000 then pulled back to 79,800, but with low volume over the weekend, it retreated back to 77,300. The 5% volatility was entirely event-driven. $ETH was clearly stronger, rising 2.7% in 24h to rebound to 2,667, with funds moving to more elastic assets.
My view: yields are high and expectations for rate hikes are heating up, which is a headwind in the short term, but funding rates are only +0.002%, leverage is not crowded at all, and someone caught the 76,000 dip. With low volume and sideways trading over the weekend, avoid chasing highs or panicking sells. The real direction will be given by next week's FOMC and the CLARITY Act vote.Brothers, last night the CPI data came out, and the market went on a roller coaster ride.
$BTC fluctuated sharply between 76,671 and 78,521, now hovering around 77,200, up just 0.1% in 24 hours, basically unchanged. But $ETH was different, surging over 8% at one point, breaking above $2,600, hitting a seven-month high, now at 2,525, up 2.67% in 24 hours.
In the past 24 hours, the entire network liquidated $732 million. The most outrageous was ETH—shorts were liquidated over $300 million, while longs only liquidated $75.55 million. Shorts were wiped out completely. On the BTC side, longs and shorts each liquidated about $100 million, ending in a draw.
Why is ETH so strong while BTC is weak?
CPI year-over-year at 3.4% met expectations, but core CPI month-over-month at 0.3% slightly exceeded expectations, with gasoline prices rising 3.9%. Once the data came out, the market turned more hawkish—the probability of a September FOMC rate hike soared to 79%. High US Treasury yields put pressure on BTC as "digital gold," but for ETH, which has staking yields and an ETF narrative, investors are more willing to pay a premium.
Next week's FOMC is the key variable. With a 79% chance of a rate hike, if it happens, BTC's 77,000 support will be tested. ETH has surged too fast in the short term, with profit-taking pressure near 2,600.
Did your ETH shorts get liquidated this round? Let's talk in the comments👇
#OKX星球话题来啦 #美国CPI环比加速，加息预期升温 $BEAT Last night I was still calculating if this month's instant noodle money was enough, and this morning I was already thinking about whether to add sausage.
During the intraday plunge, every time BEAT surged it was just short of breath, with obvious resistance above, low trading volume, and strong selling pressure. I repeatedly reminded during the session to hold the short positions firmly, don't be scared off by small rebounds, if the trend isn't broken don't make rash moves, wait for it to give the answer itself.
From 0.1223 to 0.0944, +228.94% gave the answer, big profit, this profit feels good, the earlier hesitation was real, but the outcome is really sweet, those on board should have woken up laughing.
Risk control done upfront is called rational; cutting losses after losing is called decisive. Hold if the trend isn't broken, run if it breaks, don't fall in love with stocks.
Take profit on 80% first, keep the remaining 20% at cost price for protection, don't give back profits even if it rebounds. There's still opportunity, don't rush to reverse positions.
For friends who haven't gotten on board yet, listen to me, now is not the time to rush. Wait for a more comfortable position in the next round, I will notify immediately.
$SOL $BNB $ETH What does a true “bad news priced in” mean?
For example, the economic forecast and the Federal Reserve interest rate decision on 9/17
Fed raises by 25bp, as expected
But the Chair said there is no preset plan for the next rate hike
The dot plot did not show further upward adjustments
The market lowered the probability of a rate hike in December
US Treasury yields actually fell back
Then the market is very likely to say:
"The worst tightening expectations did not materialize."
At this point, it is entirely possible that:
The news headline is "Fed rate hike"
ETH actually rises.
Just like yesterday when the core CPI was 0.3% hotter than expected, ETH’s first reaction was a sharp surge—the market trades on the difference in expectations, not the news headline itself.
So on 9/17, you shouldn’t just ask "Will they raise rates or not?"
What you really should look at is:
① Will they raise by 25bp?
② Will there be more hikes afterward?
③ Has the dot plot shifted upward?
④ Has the inflation forecast been revised upward?
⑤ How are the 2Y/10Y US Treasury yields moving?
⑥ Does ETH actually drop in response to these bad news?
The rate hike in September itself is no longer the biggest bad news; the real big bad news is the market realizing this is not the last rate hike, but the start of a new round of tightening. This is what you really need to guard against on 9/17."All $XRP holders, please set your alarms for these two dates next week."
The recent fluctuations of $XRP have all been driven by events.
In August, it surged from 1.00 to 1.698, a 72% increase, fueled by the continuous 8-week net inflow of $1.68 billion into the spot ETF approved in March.
On September 1, 1 billion tokens were unlocked from monthly custody — ten years ago, this would have been an unshakable dump day, but this time Ripple directly re-locked 700 million, releasing only 300 million net. There was no dump that day, and veteran XRP holders understand how significant this change is: a bearish factor that has lasted over a decade was dulled for the first time. However, on September 4, weekly ETF inflows shrank sharply from 110 million to 19 million, with a single-day outflow appearing. As funds slowed, the price fell from 1.70 to 1.40 and stabilized. Tonight, CPI met expectations, and the market rose +3%.
The real drama is all next week: September 15 is the Senate procedural vote on the CLARITY Act, and the FOMC meeting is on the 16th. Regulatory benefits combined with no rate hike will open the space from 1.50 to 1.80–2.10; conversely, if 1.35 breaks, the dense cost zone of 4.8 billion tokens won’t hold, and the next support to watch is 1.21.
$XRP is not just consolidating without a story now; the story is all pinned on next week.
#美国CPI环比加速，加息预期升温 Single Coin Contract Fluctuation
$CNPY leverage funds have started to move; the price-position relationship and fee rates will explain where the pressure is coming from.
15m price-position readings +1.69%/+3.66%, the open interest did not decrease as the price moved up, indicating new exposure participation is clear. Buyer market orders account for 47.6%; when positions continue to increase, a simultaneous price rise indicates effective position expansion.The most dangerous illusion on the chessboard is thinking you still have time. The Federal Reserve's September interest rate decision is not yet finalized, but a million chips are already on the table—this is not the endgame, this is the midgame's killing zone. The real money makers don't just take it step by step; they have already calculated the position twenty moves ahead before making a move.
My allocation strategy starts with discussing the value of the pieces. Spot assets are the king, the foundation, the bottom line that cannot be exchanged; contracts and options are the rooks and knights, the firepower projection, the exchangeable pieces that can be sacrificed to change the tempo. The three fronts of crypto, US stocks, and commodities are simultaneously at war, equivalent to a three-front war of attrition; those with thin forces will inevitably be restrained. I will allocate 40% to the central line—the spot base positions of Bitcoin and Ethereum, serving as the backbone of the troop chain. Thirty percent goes to tokenized US stocks, with XMSFT as the key square: it has a clear central control relationship with Nasdaq futures. If tech stocks break out on the interest rate decision night, this linkage line becomes my open line, allowing the rook to penetrate directly. Ten percent is for grid trading, slowly exchanging pieces within the oscillation range to capture the opponent's passing pawns; another ten percent is reserved for options premiums, buying downside protection in case of a crash; the final ten percent is the never-moved reserve force.
Dollar-cost averaging is a clumsy move, but for many, it is the only correct move. It seems like a dull pawn push, but it inadvertently establishes spatial advantage. True masters know that capturing pieces does not equal advantage; mobility is the real advantage.
During this midgame before September 17, the biggest fear is not a wrong direction but being forced to sacrifice pieces due to an overweight position. Margin is the number of times you can afford to be put in check, not your hope to win. The term market linkage essentially means the pieces in different areas of the board respond to each other—a breach on one line will trigger the collapse of the entire flank.
What I value most is the synchronization between XMSFT and the tech index. When it shows independent strength, it means you have an advanced rook deep in the opponent's territory; when it sinks with the broader market, it means the fulcrum of this line has been replaced. At this point, sacrificing contracts to preserve spot assets is the correct piece exchange logic.
Champions never explain their layout; they only smile when the opponent regrets their move. #okx1millionstrategistUnder the classical dome of Capitol Hill, another load of 630 pages of concrete was poured. The clarity amendment on September 10 is equivalent to pulling back the originally cantilevered glass curtain wall—what structural engineers fear most is a cantilever that is too large without shear wall support. They re-reinforced the registration rules for decentralized finance, narrowing the scope to spot and cash settlement of digital goods, while preserving the protection layers for self-custody and developers. This is not just renovation; this is moving load-bearing walls.
Those who truly understand know that a regulatory white paper is never a finished delivery; it is merely a construction blueprint. Now, the Democrats have inserted 114 amendments into this blueprint, equivalent to the client suddenly demanding adjustments to all MEP shafts before the main structure is topped out. The contractor did not throw a fit but instead agreed to incorporate them all, which itself is a highly complex structural coordination. But where is the critical point? On September 15, the cloture motion requiring 60 votes to proceed.
Sixty votes. These are sixty pile foundations. The Republicans are still short of seven cross-support columns, and these seven piles are precisely embedded in the softest soil layer—the cryptocurrency conflict of interest clauses were barely touched. Any supertall building, if the core tube’s eccentricity deviates, will sway uncontrollably in the wind. The Treasury Secretary’s public support is like the supervisor urging the schedule, but the supervisor’s signature cannot replace geological surveys.
Now look at the linkage on the side of tokenized US stocks. Mapping blue-chip stocks like Apple into on-chain certificates essentially adds a steel-structured cantilevered corridor outside the existing reinforced concrete framework. Whether the corridor looks good depends on the anchoring nodes—once the boundary of spot and cash transactions is nailed down by this amendment, the span of the corridor is limited, and freedom becomes a scarce commodity. Leaving the developer protection layer intact is like keeping the fire escape unblocked; this is the only thing that lets people sleep peacefully.
The foundation has been poured, but the reinforcement ratio is still on the negotiation table. September 15 is not the completion date; it is the last inspection before pouring. Missing seven votes means missing seven sets of stirrup spacing. #claritybessentpushWith 100x leverage, 2 BTC long positions lose $894, 3.44 BTC short positions lose $2,090, 15 ETH long positions lose $685, 12.9 ETH short positions lose $275. The same account, hit on both long and short sides. Guess what's the most painful part? It's not losing money, but this person used to be a famous air force leader, always thinking about shorting. Now he's changed his username, changed direction, switched back and forth, and every trade is riding the market's reverse rhythm. Only one ZEC long position is still profiting—50x, floating profit of $280, like a small consolation prize from the market. I stared at these charts for a long time. What really concerned me wasn't how much he lost, but what this "double kill" pattern was talking about. BTC opening average price is 77,393, mark price 76,946, less than 0.6% difference. ETH long positions average 2,470, short positions average 2,447, both directions are almost close to each other. This isn't a matter of direction judgment; it's the position being repeatedly squeezed within an extremely narrow range. 100x leverage means a 0.5% price fluctuation approaches the liquidation line, and BTC and ETH have recently been in a typical squeeze structure with low volatility and high open interest. Under this structure, what is the market trading? It's not a trend, it's stop-loss. Market makers and liquidity providers clearly know where the liquidation prices for both long and short positions are piled up; as long as the price sweeps back and forth within the key range, stop-loss orders can be triggered simultaneously. The "losses for both bulls and bears" you see are actually the inevitable result of the market harvesting high-leverage positions. Bullish$LPT 🐋 LPT (Livepeer) On-Chain Whale Holdings & Cost Analysis (As of 2026-09-12)
Important premise: The true total cost of wallets cannot be directly obtained on-chain; only the average on-chain purchase price can be calculated. Staking rewards that mint additional LPT dilute the real holding cost, so the costs below are estimates, not exact values. Also, many top holdings are staking contracts, treasuries, or CEX custody addresses and should not be considered trading whales.
1. Concentration of Holdings
• Top 100 addresses hold 75.66% of circulating supply; top 10 holders account for 51.07%; only 37 whale addresses hold 68.32% of total supply, showing high concentration.
• 13 addresses individually hold ≥1% of circulation; many large addresses are Livepeer staking contracts/treasuries, not trading whales.
2. Historically Known Trading Whales (with clear cost records)
1. Address starting with 0xba8
◦ Holdings: 114,637 LPT
◦ Entry period: April–August 2023, withdrawn in batches from Coinbase
◦ On-chain cost ≈ $5.46; later fully transferred to CEX at high prices and liquidated, profit +249%
2. Staking Whale (526,000 LPT)
◦ Entry: February–April 2025, bought and staked from Binance
◦ Cost ≈ $5.90; fully redeemed and transferred back to Binance for profit-taking in June 2025, book profit about $2.01 million
3. Coinbase Prime Custody Outflow Whale
◦ November 2025 transferred out 1,051,859 LPT to anonymous 0xabac address, no complete cost record, likely institutional/large holder large transfer, cost cannot be calculated
3. Current Whale Holdings Layered Estimates (2026 Low Range)
A batch of new whales accumulating at the 2026 bottom (price range $0.48–1.2):
• Low-entry whales (first half 2026): cost range $0.50–1.10, these are the main base whales for this cycle
• Old whales (2023–2025): cost generally $4.5–6, most have taken profits and exited during the 2025 rally, few old whales remain on-chain
• Project treasury + staking contracts: tokens from mining/team allocation, almost zero cost, largest potential selling pressure source (staking unlocks, treasury disbursements)
4. Key Risk Points
1. LPT inflation rate as high as 12.19% per year, continuous new coin issuance from staking, large holders’ positions passively increase, cost continuously diluted, long-term selling pressure exists
2. Distinguish two types of large holders:
✅ Trading whales: withdraw from CEX to wallet with clear purchase records, cost estimable
⚠️ Contract/treasury addresses: not bought on secondary market, cost near zero, unlocking is potential dump
3. Current LPT price around $1.4:
◦ Whales entering at 2026 lows: generally floating profits 20%–180%
◦ Old whales from 2023–2025 mostly exited, main on-chain force are new whales entering at 2026 lows
5. Resistance Reference
• First resistance: 1.6–1.8 (profit-taking zone for low-entry whales this cycle)
• Strong resistance: above 2.0, large concentration of low-cost holdings cashing out
Disclaimer: On-chain cost is estimated based on on-chain transfers only, excluding fees, staking rewards, multi-address holdings.
This does not constitute any trading advice. The money Nvidia is going to invest is not for buying shares
Anthropic is discussing an IPO and wants to bring Nvidia in as an anchor investor.
Nvidia is considering investing up to $10 billion.
What is an anchor investor:
They are the group who subscribes before pricing.
Others follow at this price.
How is the $10 billion calculated:
This is the subscription amount, not the market value.
It does not determine how much Anthropic is worth.
It only determines how much stock is locked up before the opening.
No matter how large the IPO is, there must be someone to take it first for there to be a price.
In this case, the one taking it is the chip-selling company.
#财报观察员：甲骨文AI云收入增121%
#SpaceXCFO称有信心实现1000亿美元ARR $NVDA ETH is rushing toward around 2670, but I'm more afraid this is a bull trap.
What I see: Someone has drawn a roadmap that first lures buyers up to about 2670, then crashes down to the 1800 liquidity zone, even down to the 1500 surrender zone.
The timeline points to making a cycle bottom in November–December, then pulling back to around 1600 before talking about going above 3250.
Right now, ETH is still hovering around 2460, with CPI heating up and rate hike expectations rising, weekend volatility is more likely to deceive.
I think don’t mistake the rebound for trend confirmation. The roadmap can be used as a reference, but don’t follow fantasies with your position size.
What to do: Watch lightly, write a plan to add positions if it breaks key support; chasing highs aggressively reduces your odds.
Invalidation condition: Volume-backed steady hold above 2700 and sustained recovery of previous highs.
Are you waiting for a pullback to buy in, or do you want to chase this wave of ETH now?
#USCPI MoM acceleration, rate hike expectations rising
#BTC spot ETF continuous outflows
$ETH $BTC $SOLBTC has already surged back to 78,000, so why is ETH still hovering around 2,600?
#BTC现货ETF连续流出
Last night, BTC first dropped then pulled back, with $BTC reclaiming the 78,000 level, but $ETH is still stuck around 2,600. It did follow the rise, but the momentum is clearly weaker. Both are warming up, so why does the leader run first while the second is still lingering at the door?
#加密财库分化：买币还是回购？
The difference is still capital. This round for BTC is supported more firmly by spot buying; when it dips, someone buys, and when it reclaims the level, others chase. Although ETH has rebounded, ETH/BTC hasn’t truly strengthened yet, indicating that more money is clustered in BTC rather than broadly expanding out. Especially with thin weekend liquidity, it’s easy to follow the rise, but the real challenge is whether anyone continues buying after a pullback. Just because it hasn’t risen much doesn’t mean it’s due for a catch-up; without capital signaling, cheap can stay cheap.
Going forward, as long as $BTC doesn’t break below 78,000 on a pullback, the strength remains; the real start signal for $ETH isn’t a 1% or 2% rise, but a volume-backed hold above 2,600, along with ETH/BTC starting to rise. If BTC turns back and ETH falls first, that’s not a catch-up rally, it’s confirmation of weakness. One watches the key level, the other tries to ignite itself — the rhythms are completely different.
Strong coins have buyers on pullbacks; weak coins have no chasers on rebounds. Don’t mistake “not rising yet” for “next in line.” The market never hands out red envelopes by queue order.OKX this time has added JP225 and ZHONGJI into stock perpetuals, which I think is more interesting than simply adding a few more US stock tickers. One is the Japanese stock index direction, the other is the Hong Kong stock AI optical module supply chain. Launching both on the same day sends a very clear signal: the stock perpetual line is no longer just focused on major US stocks, it is expanding towards Asian assets. Let me first clarify the rules. OKX's announcement states USDT-margined stock perpetuals; JP225/USDT will start on September 9, 2026, at 04:00 UTC, and ZHONGJI/USDT at 06:00 UTC; trading covers web, app, and API. This is not buying Nikkei index constituent stocks, nor holding ZHONGJI stock, and there are no dividends or voting rights like shareholder benefits. What you trade is the contract price movement, with margin, funding rate, and liquidation logic all following the perpetual contract model. This new launch is interesting because finally, in the Asian time zone, there are stock perpetuals that are more relevant. Behind JP225 is the sentiment of the Japanese stock market; the Bank of Japan, the yen, semiconductors, and export chains all influence how traders price it. ZHONGJI leans more towards the AI hardware narrative; the market will speculate on optical modules, computing infrastructure, and Hong Kong stock sentiment together. For OKX Planet users, this feels less like "just another stock ticker" and more like bringing traditional market intraday sentiment into the crypto market. But don't think it will be too smooth. OKX stock perpetuals support 24/7 trading Account Position Divergence Radar
Account direction reflects sentiment, and position weight reflects strength. This set specifically looks for places where the two do not align.
$BEAT All accounts and top accounts are biased long, but the top position size is biased short. The number of accounts and position weight are not on the same side. Price falls and positions reduce, risk exposure is contracting, so it cannot be directly written as new shorts. Next, watch whether the top position size turns long; otherwise, even if there are more long-biased accounts, it is only a numerical advantage.
$DOGE The number of accounts consistently leans long, but the top position ratio remains below 1, so the numerical advantage has not turned into a top position advantage. Price and positions both retreat, reducing position pressure is releasing, but which side is exiting cannot be confirmed by this data alone. The top position ratio needs to repair toward 1 to consider that position weight has started to catch up with account sentiment.
$SUI Different account metrics stand on different sides; currently, treat it as divergence without amplifying any particular proportion. A 15-minute rise with position reduction looks more like short covering or overall withdrawal driving it; new longs have not yet been confirmed. What is currently lacking is consistency; continue to watch whether the divergence expands or begins to narrow. Oracle's earnings report brings the two keywords "AI surge" and "burn rate anxiety" to the forefront simultaneously.
---
1. First, look at the most explosive numbers
OCI cloud infrastructure revenue reached $7.4 billion, up 121% year-over-year. What does this mean? Last quarter's growth rate was 93%, the quarter before that was even lower, and now it's accelerating again. Overall cloud business revenue was $11.6 billion, up 62% year-over-year. The company's total revenue was $19.35 billion, up 30% year-over-year, fully exceeding market expectations.
This is not a slight beat, but a crushing beat.
More attention-worthy is the delivery side. In a single quarter, 850MW of data center capacity was added, over 300,000 GPUs went online, delivery volume was nearly three times that of the previous quarter, and GPU utilization reached 97.9%.
What does nearly 98% utilization mean? Computing power is basically snapped up by customers as soon as it is deployed. The CFO directly used the word "acceleration" to describe this quarter during the earnings call.
Demand is even more exaggerated. Remaining Performance Obligations (RPO) surged to $664 billion, a new historical high, with $26 billion added in a single quarter. This quarter, new AI cloud contracts signed exceeded $30 billion.
The CFO clearly stated that about half of the RPO will convert into revenue within the next 36 months. To translate: this is not just a promise, but a real revenue pipeline of cold hard cash for the next three years.
2. But on the other hand, we cannot ignore the other side
Capital expenditures are ramping up aggressively. Q1 capex reached $28.45 billion; after deducting customer prepayments, the company's net spending was still $17.97 billion. Free cash flow? Deep underwater, with a single quarter negative $5.4 billion.
For the entire FY2026, capital expenditures total $55.7 billion, with free cash flow negative $23.7 billion for the year. The company plans to increase FY2027 capex to $90-95 billion.
JPMorgan has already downgraded Oracle's rating, with the core concern being "thin margins cannot support high leverage." S&P also downgraded Oracle's credit rating from BBB to BBB- in July, citing three main reasons: long investment payback period, high uncertainty in the AI industry, and high customer concentration risk.
Speaking of customer concentration, this may be the biggest hidden risk. OpenAI accounts for about half of Oracle's RPO. This means if OpenAI's financing pace falters or the industry's computing power supply landscape changes, Oracle's order backlog will immediately come under pressure.
3. The competitive landscape is heating up rapidly
Microsoft Azure's annual revenue surpassed $100 billion for the first time, up 41% year-over-year; Google Cloud's quarterly revenue was $24.8 billion, up 82% year-over-year, with a backlog of $514 billion. AWS is also active, with quarterly revenue of $42.2 billion and 37% growth.
The three giants are all locking in long-term computing power contracts with leading AI labs. Anthropic has committed to spending over $100 billion on AWS over the next decade, and Microsoft has secured about $30 billion in Azure computing power purchase commitments.
The entire industry's capital expenditures are astronomical. The four major North American cloud providers combined Q2 capex was $171.2 billion, up 78.6% year-over-year, with an expected full-year total exceeding $732.5 billion. The nine major cloud providers' combined capital expenditures are expected to grow about 90% year-over-year, surpassing $886.7 billion.
This is not just Oracle's big bet; the entire AI infrastructure sector is collectively doubling down. The problem is, Oracle's scale is far smaller than Microsoft and Google, so if a price war breaks out, its ability to withstand pressure is questionable.
4. What is the market hesitating about?
After the earnings release, Oracle's stock price rose nearly 10% in after-hours trading, but the next day, after opening up 8.5%, it quickly fell back and turned negative multiple times during the session.
Both bulls and bears can find ammunition in this earnings report. Bulls say: OCI growth 121%, RPO $664 billion, demand far exceeds supply, customers are lining up to send money. Bears say: free cash flow is deeply negative, capital expenditures continue to expand, profits are eaten up by depreciation, can this game continue?
Before the earnings release, the market's core anxiety was "Is Oracle building ahead of demand?" After the report, this anxiety turned into "Demand is indeed strong, but can this money ultimately be earned back?"
5. Core judgment
Oracle's transformation direction is no longer in dispute. The leap from a traditional database software company to the world's fourth-largest cloud infrastructure service provider is underway. AI computing power leasing is currently one of the most certain demand growth areas in the tech industry, and OCI's growth in this sector leads all major players.
But from an investment logic perspective, this earnings report presents a typical high-risk, high-reward model:
· Upside logic: massive RPO backlog + accelerated computing power delivery + customer prepayments easing pressure; if AI demand continues to exceed expectations, Oracle could be one of the biggest beneficiaries of this AI infrastructure wave.
· Downside risk: uncontrolled capital expenditures + high customer concentration + intensified industry competition; if AI demand fluctuates, the negative cash flow snowball will grow larger.
One detail worth noting: about 40% of this quarter's capex is covered by customer prepayments, with customer prepayment revenue reaching $11.36 billion this quarter. Customers' willingness to pay upfront itself indicates tight computing power supply and demand. But this also means that if demand reverses, these prepayments correspond to infrastructure projects already underway—non-refundable and unstoppable.
This is Oracle's current situation: if the bet is right, it will be the next trillion-dollar infrastructure giant; if wrong, it will be burdened with tens of billions of dollars in debt and excess computing power assets.
The market and news are all laid out; how to bet depends on how much you believe in AI demand.
#财报观察员：甲骨文AI云收入增121% The 2021 Gas price table can no longer accommodate Ethereum in 2026
EIP-8038 plans to increase the Gas costs of certain state access operations, including SSTORE, SLOAD, cold account access, and some code reads. The last systemic adjustment of related costs dates back to the Berlin upgrade in 2021.
Over five years, Ethereum's state size, node burden, and execution environment have all changed. If the protocol still charges according to the old costs, some on-chain operations may experience an imbalance of "users paying very little while nodes do a lot of work."
This imbalance may not be obvious under normal circumstances but will quickly amplify as block capacity continues to increase. Attackers can concentrate calls to underestimated operations, creating excessive execution pressure with limited Gas.
Therefore, scaling $ETH is not simply about raising the Gas limit but first calibrating the true cost of each type of operation. Only when the price table closely matches resource consumption will increasing block capacity not secretly overdraw node performance.
Gas is Ethereum's internal resource pricing system. When pricing is incorrect, cheap does not mean efficient; it only means someone else is paying for the cost, and ultimately, the one paying is often the degree of decentralization.Roughly $684 million in futures positions were liquidated in 24 hours about $423 million shorts versus $261 million longs. Direct consequence: the tape bounced, but it was a squeeze after a PPI scare and mixed CPI, not a clean regime change. Institutions still pulled ETF money while leverage got cooked. Hot U.S. PPI first shoved Fed hike odds higher and shoved Bitcoin under $77,000, so longs paid first. Then CPI landed mixed hotter monthly core, cooler annual core and the crowd flipped from fThe core variable for the global financial markets next week is undoubtedly the Federal Reserve's policy meeting. Before the results are announced, the market is unlikely to follow a smooth one-way trend; instead, it will be a week of intense volatility driven by news, with repeated spikes and liquidations possible.
The current key price levels for BTC are clear: resistance is concentrated between 79,500 and 81,500, with the previous high at 82,300 repeatedly failing to break through, indicating weakening bullish momentum; the first support is at 77,000, with strong support between 75,500 and 76,000. If the strong support is effectively broken, the correction space will further open, possibly targeting around 73,000.
There are two scenarios in the market. First, if there is a rate hike and hawkish remarks, US Treasury yields will continue to rise, risk assets will come under pressure, BTC may first test around 76,000, altcoins will collectively see valuation cuts, and sector rotation will pause. Second, if there is a rate hike but cautious statements hinting at a possible pause later, after the negative news is priced in, the market may experience a "drop then rally" pattern, with a short-term dip to shake out weak hands followed by capital inflow and a rebound pushing above 80,000.
Currently, BTC is unstable, altcoins find it difficult to strengthen independently, and chasing highs carries significant risk. Next week, leveraged account liquidations may repeatedly occur, and spikes will be very frequent.
In terms of strategy, do not heavily bet on the rate hike outcome in advance. If the rebound does not break 80,000, do not blindly go long; if support at 76,000 shows no sign of stabilization, do not rush to bottom-fish. Spot positions can be accumulated gradually, contracts must reduce leverage, and patiently wait for the meeting results to clarify the direction before following the trend.
$BTC $ETH
#美国CPI环比加速，加息预期升温