Orbit Post Sitemap
Capital rotation is still the entire game in crypto. The easiest trades are buy the uptrend, wait for momentum to slow, then short as capital and attention rotate elsewhere.
$ZEC, $HYPE, $LIT are all recent examples, but this pattern has been running for a long time.
I’m watching traders get mind-gamed by $ETH again. Slight outperformance vs $BTC, while $BTC just had a decent month.
BTC tends to rally in July and fade in August. With the tardfi takeover of crypto now complete, summer months are looking even less attractive.
Strong opinions, loosely held. If you’re buying the trend, great. Just don’t convince yourself price can only go up from here.
Take profits and be ready to flip your view when that momentum stalls.
Most moves are still driven by trend-following flows. That often comes before spot rallies. But the lack of real spot participation is still glaring.
Patience.#苹果第三财季业绩超预期, the stock price plunged sharply after hours
Apple's financial report is actually ironclad evidence that AI chips are eating away at consumer electronics
Apple released its penultimate earnings report before Cook stepped down. Revenue reached 109.4 billion yuan, the highest ever. iPhone sales reached 54.2 billion yuan, a record high. Then it fell 6% in after-hours trading, with its market value evaporating by over $300 billion overnight.
On the surface, it looks like "performance won, expectations lost." But I don't think it's that simple—this is a financial report telling you that "AI is taking over consumer electronics."
Apple's guidance for next quarter is only 9%-11% growth, while the market expects it to be above 12%. The difference of 1-3 percentage points isn't because Apple products are inadequate, but because they can't get enough chips and memory. Cook's exact words are "once-in-a-century" increases in storage costs. Apple's Macs and iPads have already raised prices, while iPhones have not. However, if supply constraints persist, prices will inevitably rise.
Do you know where the chips and memory went? AI data centers.
By 2026, 100% of N3 capacity will be sold out, with 60% taken by AI chip customers. A consumer electronics giant like Apple relies on the remaining AI chip production capacity to produce chips. AI servers are competing for capacity, consumer electronics are waiting for capacity—this is a structural contradiction within the supply chain.
So last night's Apple plunge was on the surface "guidance below expectations," but behind it was AI infrastructure squeezing the entire consumer electronics industry. This isn't Apple's problem—it's the entire industry being repriced by AI.
The same applies to the crypto world. Funds are concentrating on AI computing power, AI chips, and AI-related tokenized stocks. Nvidia, TSMC, and SK Hynix have surged faster than Apple—the market is telling you where your money is going.
I will continue to follow this direction and not rush to buy Apple at the bottom. Wait for signals of easing supply restrictions before making any decisions.
Let's talk in the comments: do you think Apple was wrongly killed or should keep falling? Eric Trump has once again called out: Bitcoin will sooner or later rise to $1 million
This is a bit provocative, but before you hear it, it's best to first look at his interests
1️⃣ He is the co-founder of American Bitcoin
The company currently owns about 89,000 mining machines, with a computing power of 28.1 EH/s. Based on recent estimates of Bitcoin's total network hash rate, it accounts for about 3%.
The higher Bitcoin rises, the more mining revenue, company holdings, and stock valuations may all benefit
2️⃣ His personal wealth is highly tied to the crypto industry
Some media estimates that Eric Trump's current net worth is about $400 million, but this does not mean he personally holds $400 million in cryptocurrency
A large portion of this value comes from American Bitcoin shares and other investments related to the crypto industry. The rise in Bitcoin prices is certainly no small matter for his wealth
3️⃣ He is also a co-founder of World Liberty Financial
This is a DeFi project involving the Trump family, which includes the WLFI token and the USD1 stablecoin. When Eric Trump talks about the crypto industry, his identity is closer to project teams and stakeholders, not just an independent analyst sitting off-site
So saying Bitcoin could reach $1 million doesn't mean this prediction is wrong
But this is by no means an objective prediction without positions
It's fine to listen to opinions, but don't forget to first look at what he's holding. Wherever a person's interests lie, their voice often 👀 becomes louder in that direction. $BTC Actually, I increasingly feel that when investing in a company, what you really need to study is not the product, but the business model.
Because technology can be caught up, business models often determine how much money a company can make in the long run.
Why have I always thought that most storage companies have obvious cycle attributes?
The reason is not only that DRAM, HBM, or NAND are affected by supply and demand, but more importantly, most of them adopt the IDM (Design + Build Integration) model.
From chip design and wafer manufacturing to packaging and testing, almost the entire production process must be completed in-house.
The advantage of this model is control over technology and production capacity, but it comes at a very high cost.
A wafer fab, an EUV lithography machine, and an advanced packaging production line all represent massive investments. Even as the market enters a downturn, fixed expenses such as equipment depreciation, factory maintenance, labor costs, and ongoing capacity expansion will persist and will not decrease due to falling chip prices.
Therefore, when storage prices enter a downward cycle, profits are often quickly squeezed and financial reports fluctuate significantly.
Looking at NVIDIA in the opposite direction.
It does not build its own wafer fab, but instead focuses on chip design, leaving manufacturing to TSMC.
Although this fabless model also bears foundry costs, it doesn't have to bear tens of billions of dollars in manufacturing assets, allowing more funds to be invested in CUDA, NVLink, software ecosystems, and next-generation product development.
So I have always believed that a company's true moat is not just technological leadership, but whether its business model can consistently generate high returns.
Often, investing isn't about which technology is the best, but whose business model is easier to weather cycles. At the end of this round, I switched from shorting ETH to shorting BTC 🔄
Previously, I used ETH as collateral to get WBETH as margin, opening a short perpetual position with the same amount. This method brings three benefits:
1️⃣ Equivalent to selling ETH spot, no leverage, no liquidation
2️⃣ No interest cost on ETH collateral, plus receiving funding from the long side
3️⃣ The exchange only holds a small portion of ETH as margin, reducing the risk of the exchange disappearing or being hacked
After that, I switched to using WBETH as margin, shorting BTC spot at the same conversion ratio.
Why? MSTR might stop buying BTC in To put it bluntly, the market should give Amazon $AMZN and Google $GOOGL a ......
Amazon's earnings report is a hit: AWS grows 37%, and after Google, AI demand is once again being validated
When discussing Google's financial report, I mentioned that the companies to focus on in this round of tech and storage stocks are actually Google and Amazon.
The reason is simple: they are among the world's largest cloud vendors and the most genuine buyers of GPUs, HBMs, server DRAM, enterprise-grade SSDs, and data center network equipment.
Google's financial report has already given the answer: Google Cloud's revenue reached $24.8 billion, an 82% year-on-year increase, proving that enterprise demand for AI training, inference, and cloud services has not weakened. Now Amazon has also handed over its paper, and its data is equally strong.
Amazon $AMZN Q2 revenue reached $200.6 billion, up 20% year-over-year; Operating profit reached $27.5 billion, up 43% year-on-year. Of course, the $62.6 billion net profit includes $53.4 billion in pre-tax non-operating income from Anthropic's investment appreciation, so this figure cannot be fully attributed to Amazon's normal operating earnings.
Even more noteworthy are operating profit and AWS.
AWS revenue reached $42.2 billion this quarter, up 37% year-over-year, marking the fastest growth rate in the past 18 quarters;
AWS's operating profit reached $16.6 billion, continuing a significant increase from $10.2 billion in the same period last year.
Even more astonishing, the backlog of AWS orders has reached $496 billion, compared to just $364 billion in the previous quarter.
Amazon also stated that most of the hash rate has already been pre-ordered by customers for 2027, and a significant portion of capacity will be locked in 2028.
🤔🤔🤔
Hey, here's something interesting?
Previously, the market was worried whether cloud providers had built too many AI data centers and ended up with no customers using them; But now Amazon is telling everyone that even if they raise capital expenditure to $220 billion by 2026, computing power will still be insufficient.
This is very interesting.
Amazon isn't spending money and not knowing where to find customers; rather, customer orders are already scheduled through 2027 and 2028, and the company must continue buying servers, chips, and storage to meet these demands.
AWS grew 37%, with orders reaching $496 billion, indicating that AI investment is beginning to translate into cloud revenue, profit, and future contracts, rather than just capital expenditures in financial statements.
So when you look at Google's and Amazon's financial reports side by side, the answer is already very clear:
Google Cloud grew 82%, AWS grew 37%, and both companies stated that their current computing power supply still cannot fully meet customer demand.
This shows that enterprise AI has not stopped, and cloud providers are not preparing to suddenly cut investments. On the contrary, training, inference, agents, and enterprise AI applications continue to rapidly increase demand for data centers.
So where all this money ultimately ends up isn't hard to imagine.
GPUs handle computation, HBMs feed data quickly to GPUs, server DRAM handles models and execution tasks, and enterprise-grade SSDs store training data, model parameters, Checkpoints, and increasing amounts of inference data.
Moreover, Amazon even directly mentioned that the cost of purchasing storage chips is one of the main reasons for the company's capital expenditure increase.
For the storage industry chain, I believe this is a very direct benefit.
Recently, storage stocks such as Micron $MU, SanDisk $SNDK, and SK Hynix $SKHY have experienced a very clear correction.
Previously, the market was concerned about AI capital spending peaking, storage prices falling from high levels, and cloud vendors suddenly reducing purchases;
But what Google and Amazon's earnings reports tell us is that customer demand remains so strong that computing power is insufficient. Even with a recent rebound, Micron still pulled back about 25% in the past month, and the storage sector's expectations and leverage have clearly been released.
Of course, demand validation does not mean storage stocks will only rise and not fall from today.
In the short term, we still need to look at US Treasuries, market sentiment, past earnings expectations, and storage prices, but at least the most important fundamental question has been answered:
Cloud vendors are still buying, and not enough.
Therefore, I believe Amazon's financial report is not only positive for Amazon itself but also another faith recharge for the entire AI infrastructure and storage industry chain, following Google and Microsoft.
The market has been asking: With so much money invested in AI, is anyone actually using it?
Now, AWS answers this question with 37% growth, $496 billion in orders, and still insufficient computing power.
Demand remains, orders are there.
Now, storage stocks have just undergone another relatively substantial correction. I believe that at this stage, rather than continuing to panic about whether the AI market is over, it's better to return to earnings reports and see if cloud providers are still buying.
At least Amazon's answer is very clear.
This is a major positive development; there are no potential risks ahead, and preparations are underway for a new round of takeoff......The storage giants stage a "double kill" between long and short! Smart money retreating, are the bears gathering?
A rebound is not a reversal; chasing highs is destined to be a guardhousehold—smart money is voting with its feet.
News: Samsung's Q2 profit surged 1814% year-on-year, setting a new record, with both volume and price in storage business rising; Executives have clearly stated that the shortage will continue into 2028, with an even larger gap in 2027, with 60%-70% of capacity locked in long-term contracts. Global storage sales in July reached a new high of $74.6 billion. But after the surge, smart money collectively shorted, the funding rates of five major stocks turned negative, and whales reduced their positions on a large scale.
$SNDK After a violent rebound from 972 to 1433, it quickly pulled back. The 4-hour RSI resonated across three lines (74.41/61.83/52.22), indicating momentum exhaustion after being overbought.
$MU After rebounding from 706 to 916, the market is under pressure in sync, with RSI (73.76/61.26/53.76) also showing bearish divergence signals.
Smart money movements: both bulls and bears are running
The rates for the five major targets all turned negative, with SKHX, SNDK, and MU 8-hour rates reaching -0.2544%, -0.0242%, and -0.0032% respectively. The three storage giants' nominal open interest dropped 16.6%, SKHX shrank by 29.5%—43 million-dollar long positions collectively reduced by $175 million, and 34 addresses were directly liquidated! Overall, million-dollar addresses saw a net increase of $161 million and a net short position of $271 million in the past 24 hours—both bulls and bears are retreating, but who's swimming naked?
Operational Approach
Short positions: Enter short positions near SNDK rebound 1430-1450, MU rebound near 910-930🔍 On-chain perspective: $GIGGLE's "charity narrative" and whale harvesters
GIGGLE is a BNB Chain token that combines "charity" with "meme," automatically donated to Giggle Academy founded by CZ through transfer fees. Below is a breakdown of the current long-short logic from on-chain data.
---
🧱 Resistance level: The trapped front high is eyeing the market aggressively
The primary resistance is near the previous high—this was the previous stage high, where a large amount of trapped interest has accumulated. The second pressure is the area where large holders hold the cost concentrated. On-chain data shows that over 80% of holders in this range are profitable, with a strong desire to take profits. The third resistance is in the area where the Fibonacci extension overlaps with the trendline, and the daily downtrend line also runs here. Under double pressure, increased volume is needed to break through.
🛡️ Support Level: Layer of defense but caution against breaking through
The first line of defense is the short-term moving average group (MA30 and MA60), currently running in a lower area, serving as an important cost line for the medium-term trend. Further down is the historically dense transaction zone, where cumulative turnover rates are high, accumulating a large amount of buying demand. Dynamic support should focus on the intraday low and the range where trading volume suddenly expands—if a certain price level has a sudden increase in volume, it suggests strong support.
---
🐋 On-chain market player movements: whale position building costs are highly valuable for reference
There have been important whale position building signals on-chain—BSC addresses have continuously increased positions in the past, with cumulative position costs at relatively low prices. Currently, the floating profit is extremely generous, corresponding to a very high return rate. The cost of building a position at this address can serve as an important supporting reference. Another whale immediately withdrew 1.2 million USDT from the exchange and bought it on-chain after Binance announced its listing.
However, some whales chose to take profits—some addresses deposited large amounts of GIGGLE to exchanges, and selling them would yield nearly $700,000 in profit. The divergence between bulls and bears is obvious.
---
📈 Positive factors
· CZ endorsement effect: $GIGGLE Academy was founded by Binance founder CZ and has benefited over one million children. CZ has publicly stated support for all meme coins and may buy or sell one or two tokens in the coming weeks to test new features.
· CEX listing expectations: OKX has listed GIGGLE perpetual contracts, Binance previously announced spot listings, and the liquidity foundation is solid.
· Deflationary mechanism: Tokenomics includes a burn mechanism and a fee donation model.
📉 Bearish factors
· Fixed monthly selling pressure: At the end of each month, Giggle Academy exchanges or sells most of its donated tokens for BNB to fund operations. CZ made it clear that this is "routine" and not used to support prices.
· Whale profit-taking: Some whales have accumulated and chosen to deposit on exchanges to sell their stocks.
· Meme coins with high volatility: Historical highs once reached extremely high prices, followed by significant corrections. Currently, prices are still far from historical highs, with heavy pressure from trapped positions above.
---
🎯 On-chain analyst conclusion
$GIGGLE currently sits in a delicate balance between "charity narratives" and "whale harvesting." The whale's position building costs are relatively low, with generous floating profits—meaning the main players have ample profit potential and may choose to cash out at any time. At the end of each month, Giggle Academy's fixed selling pressure is like the sword of Damocles hanging overhead. It is recommended to closely monitor the unusual movements of on-chain whale addresses and the monthly sell-off window, staying alert until the direction becomes clear. #PCE环比转负, GDP growth slowed to 1.5% #美股加密标的承压, currency price fluctuations affected financial reports, #美伊报复循环加速 oil prices rose 20% this month Rumor gossip:
MMs in the crypto world are no longer satisfied with being bulls for altcoins; they're already studying how to operate in US stocks.
The main reasons are as follows:
Retail investors in altcoins have basically all been cut off
The main thing is, there are no counterparts. Without counterparts, even if you pull them up, it's useless!
Exchanges have also improved risk controls for MMs, making it easy to trigger these controls and prevent funds from withdrawing!Among today's top 10 Binance contract trading volumes, besides SanDisk, SK Hynix, and Micron, there are also leveraged ETFs that go 3x long on semiconductors and 3x long on the Korean stock market, and can then open up to 50x on top of these 3x ETFs...
Meanwhile, Aster's Hynix contract is even deeper than many altcoin contracts, and the official company says they've been optimizing liquidity in this area recently.AI infrastructure hasn't stopped—is this storage wave a rebound or a reversal?
Damn, today we're not talking about virtual things, let's just talk about storage.
The market has risen like this, and it's not speculative speculation. Many people stared at the candlesticks of Micron and Hynix, thinking it was another familiar "semiconductor deception" storyline. But this time, the logic at the bottom really changed.
Google's ledger doesn't count profits, it's 'provisions'
After the market closed yesterday, when Google's earnings report came out, my first reaction wasn't to look at cloud business growth—an 82% year-on-year increase was expected. What really lifted my spirits was the Capex (capital expenditure) guidance hidden in the footnotes—it was raised again.
Do you know what that means?
It means that those Wall Street analysts who used to complain daily about "AI burns too much money, no returns are visible" are like fart when faced with real business needs. The $24.8 billion in cloud business revenue shows that customers are truly paying for computing power, not just listening to PPTs to brag.
When a giant makes money while still daring to invest heavily in infrastructure, for the upstream supply chain, there are only three words: keep going.
Stop using the old calendar system for phone cycles to judge storage
I've told many friends that storage used to be the "tag on consumer electronics"—if phones didn't sell, DRAM would drop; PC clears inventory, NAND is doomed. But that's all an old story.
Now, if you look at a data center's BOM (bill of materials), besides the GPU money-eating beast, the second most important item is storage:
· To train a large model, HBM must be sufficiently managed; otherwise, the GPU can only wait to feed data, which is a pure waste of electricity;
· For inference services, the durability and capacity of enterprise-grade SSDs directly determine response speed, and customers don't have the patience to wait in circles;
· All those user-uploaded videos, chat logs, and industry fine-tuning data rely on large-capacity hard drives to be fed in, and the numbers only increase.
Storage nowadays has become a damn "consumable" for AI infrastructure.
As long as Google, Microsoft, Amazon, and Meta are still building data centers, and those big models are still burning money to buy cards, storage orders will never stop. What does this have to do with the phone replacement cycle?
The earlier drop was because the market treated "spending money" as a problem
Recently, deposit stocks pulled back, dropping so much that even their own mom wouldn't recognize them. A group of price investors around me are all shouting "The cycle has peaked" and "Valuations are too expensive."
But at the time, I had only one view: what they feared wasn't storage price drops, but cloud providers not spending money.
Think about it: if big companies cut Capex, it means AI infrastructure will come to a halt, and storage will naturally become excess capacity. But Google's signal this time is very clear: I have plenty of money, so I want to keep expanding.
Once the market realizes that spending this money can turn into cloud revenue and AI gross profit, then the previous decline becomes pure "emotional mistake."
Stay calm, don't get carried away after a couple of days' increase
Of course, the words are blunt, but the reasoning is not unfounded; the cold water that needs to be poured still has to be poured on it.
Storage is the king of cycles by nature—when it rises, it doesn't care about it; when it falls, it doesn't care about it. Right now, the market is trading "sentiment repair," which means filling in the pitfalls that were previously smashed.
But can we keep pushing forward from here on? Keep an eye on these three idiots:
1. HBM production scheduling: This is hard currency, so if supply exceeds supply, keep pumping;
2. DRAM contract price: The results of negotiations at the end of Q3 are not to be deceived;
3. Next quarter's Capex for other cloud providers: Google alone doesn't have the final say; it depends on whether AWS and Azure follow suit.
One last sentence
The essence of this round of storage market is not some oversold chip stocks rebounding.
Essentially, global data centers are being rewritten by AI, and storage is that paper filled with data.
As long as computing power keeps expanding, so does data; Data keeps expanding, and storage will never be enough.
As for the stock price? That's an emotional matter. But the orders are truly on the table.
Whether you do it or not is up to you.
---
(Disclaimer: The above content is purely nonsense from industry logic and does not constitute any buy or sell advice. If you lose money, don't contact me; if you profit, don't share with me.) ）On the morning of July 28, at 10:11 a.m. – the market experienced a night of destruction with $6.8 billion liquidated, 164,535 accounts to zero, mainly the long side was wiped out 📉. South Korea activated SIDECAR, suspended automatic sales, a period of deleveraging and risk is ongoing.
SNDK is down 20% from yesterday's peak. The semiconductor and storage groups, which used to rise sharply in May-June, have now wiped out all June profits. The bottom catcher is crying silently. 🚽
SOL yesterday entered a small buy order, there is a chance to replenish. The strict strategy is still maintained. Web3 has fluctuated slightly for nearly a month now compared to$MMT Today's rally was strong, jumping from 0.20 straight to 0.389, a 72% increase in one day. Volume is 699 million, and turnover is 265 million—this level supports the rapid price increase.
The liquidation data is quite interesting: in 24 hours, 3.01 million USD, short positions 2.48 million USD, and long positions 530,000 USD. 1.23 million USD liquidated in 1 hour, short positions 1.21 million USD. This wave of rally has wiped out all the shorts, with short liquidations more than four times that of the bulls.
On the daily chart, the price rose 112% on the 7th and 140% on the 30th, clearly showing a one-sided upward trend. This approach often means the bears are repeatedly liquidated, with every bullish candlestick forcing the bears to exit.
At this level, you need to be cautious of risks when chasing in, since the short-term gains are already significant. But if the bears are still being cleared, there may still be room above. This kind of demon coin moves at a fast pace, so control your position well when participating.
Is anyone in the car in the comment section? Share your entry location. Strategy's Q2 results may look rough on the income statement, but the direct impact on Bitcoin appears limited.
The company reported an $8.2B quarterly loss, while its BTC holdings grew by roughly 11% during the same period. The reported loss is primarily an accounting markdown—not evidence of spot Bitcoin being sold.
For the market, the key takeaway is that this doesn't automatically translate into additional BTC supply.
$BTC #SoftPCEStrongDemand #AMZNMissesButRallies #交易之声: Your experience deserves to be heard
If I had to pick just one signal, I believe it most would be NUPL (Net Unrealized Profit) falling below zero.
This indicator is the most reliable North Star in my many years of trading career. It is not a price indicator but a human nature indicator, directly measuring the greed and fear of the entire market.
The calculation for NUPL is simple: (Market Cap - Realized Market Cap) / Market Cap. When this ratio falls below zero, it means that all Bitcoin holders across the network are in a state of paper loss. This isn't a loss for just a few people, but for everyone—from newcomers to veterans with three years of experience, from retail investors to institutions, none are spared.
Why is it more reliable than other signals? Because prices can be faked, candlesticks can be drawn, but the realized on-chain market cap is real. The recorded market capitalization is the price of each Bitcoin at the last on-chain transfer, reflecting the cost of real money in the token. When market capitalization falls below realized market value, it indicates the market has fallen below the average cost line of all historical buyers. This is not a technical support level; it is the support line of economic reality. If it falls further, it will be a systemic trampling.
I have experienced three moments when NUPL < zero: January 2015, December 2018, and November 2022, each time at the bottom of the major cycle. In 2018, NUPL hovered around -0.25 for two whole months, with Bitcoin grinding from $3,200 to $6,500, then kicking off the rampant bull market of 2020-2021. After FTX's collapse in November 2022, NUPL once dropped to -0.18. At that time, the group was silent, Weibo didn't have crypto trending topics, and I invested in that range for eight months, with an average cost of $18,000.
What is the deeper meaning of NUPL<0? It means the weak hand has completely left the field. Those who should cut losses have been cut, those who should quit the industry have withdrawn, and those who should have been liquidated have exploded. Those who stayed inside were either the low-cost ancient giant whales or believers who would rather die than sell. When selling pressure dries up, even a small amount of buying can drive the price back. This is the moment when the spring is at its tightest. But it is not a cure-all. NUPL<0 only tells you this might be the bottom range; if not, it will rise tomorrow. The bottom can be ground for three months or a year. So I use it as an anchor for long-term position management: start regular investing at NUPL <0, buying more as it falls; when NUPL turns positive and enters the frenzy zone (>0.5), start taking profits in batches.
Over the years, I've seen too many different arguments in the crypto world. But NUPL tells me: human nature has never changed; the pendulum of greed and fear always swings. When the entire internet is losing money, opportunities arise. This is not technical analysis; it is a belief in human nature.
So, if I can only ride one indicator through the next bear market, I choose NUPL.📊 On-chain perspective: $MMT Long-short battles under major force control
MMT has recently experienced a textbook-level trend in major players. As a CLMM DEX protocol token based on the Sui chain, MMT surged violently from around 0.19 to 0.30 before retreating, currently consolidating at high levels. Below is a breakdown of the market from on-chain data.
---
🧱 Resistance level: 0.3000 is the life-or-death line
The 0.30 integer level is a cluster of trapped positions at the previous high, with heavy selling pressure above. After the 1-hour candlestick hit resistance at 0.3001, it pulled back, with RSI surging to 84, entering overbought dulling. Sell orders are extremely thick, with order book depth imbalance of -17.48%.
🛡️ Support points: layered defenses but all with hidden dangers
The first line of defense is at 0.2588 (recent pullback low); The second level, 0.2187, is mid-range support; The most critical point is the MA7 moving average at 0.2085. The 0.23 level below serves as main cost support, and a break below could trigger a cascade. 0.21 There is a limit buy order with a large amount of smart money posted below.
---
🐋 On-chain market maker movement: Dumping is in progress
On-chain data reveals clear signals that major players are selling off:
· The largest position address (11.2%) placed consecutive sell orders within the 0.25-0.26 range, with a total of 3 cancellations—a typical false tray trading technique.
· Whale address 0x7b3a... F9E2 transferred 2.1 million tokens to exchanges (costing about 0.19), with unrealized profits exceeding 30%.
· Another address transferred 1.3 million tokens to an unknown wallet in three installments within 5 minutes, suspected of off-exchange selling.
· In the past 24 hours, six addresses had large transfers exceeding 500,000 tokens, with exchanges seeing a net outflow of about 3.4 million tokens, but 70% came from two newly added black hole addresses—this is creating the illusion of replacement.
The 24-hour turnover rate reached 45%, with prices surging and pulling back before closing with a long upper shadow, a typical structure of pushing up shipments.
---
📈 Positive factors
· Upbit ranks third in trading volume, with a 24-hour transaction volume of about $34.05 million, drawing significant market attention.
· Top-tier institutional endorsements: Coinbase Ventures, OKX Ventures, Binance are participating.
· Q4 Staking Module Expectations: Official TG released a rocket countdown, hinting that the market views this as positive news and has taken an early lead.
· Technical bullish structure: The 4H MACD histogram is still expanding, with the price holding above the EMA20.
📉 Bearish factors
· On August 4, 4.93 million $MMT (about $927,000) were unlocked, and the ecological allocation unlock will increase selling pressure.
· The 1H RSI is in the overbought range of 80-90, the 1H MACD histogram is starting to shorten, and bullish momentum is narrowing.
· Market depth imbalance: Bid/Ask ratio 0.89, selling pressure slightly favors.
· A turnover rate of 45% combined with a long upper shadow is a typical stage top signal.
---
🎯 On-chain analyst conclusion
$MMT Currently, the chip distribution phase is under strong control by major players. 0.30 is the dividing line between bulls and bears; if it breaks through and stabilizes with increased volume, it could start a new round of gains; If the breakout is delayed, with ongoing on-chain sales and the dual pressure of August unlocking, a pullback to 0.26-0.27 or even deeper will be highly likely. Currently, there are many liquidity traps, and the risk of chasing at higher prices is extremely high. #PCE环比转负, GDP growth slowed to 1.5% #美股加密标的承压, currency price fluctuations affected financial reports, #美伊报复循环加速 oil prices rose 20% this month South Korea's capital market has reached a historic moment as financial regulators have officially tightened policies for leveraged ETFs on individual stocks. The results were immediate, even exceeding the expectations of many market participants. According to data from the Korea Exchange after the market closed that day, on the first day of the new regulations, the total turnover of leveraged and inverse ETFs for 16 individual stocks across the market was only 3.3071 trillion KRW. How bleak is this number? For comparison: just the day before, this figure was still 12.4485 trillion Korean won. Overnight, turnover plummeted by 75.3%. $SNDK Looking at longer-term data, the average daily turnover in July so far is 12.27 trillion KRW. On the first day of the new regulations, the results were less than a fraction of this average. The regulators' determination to cool down an overheated market is clearly conveyed through this set of data. Even narrowing the scope and excluding inverse products, looking only at 14 purely individual stock leveraged ETFs, the situation is not much better. Their turnover fell directly from 6.9354 trillion won the previous day to 2.4686 trillion won, a drop of 64.4%. $MU The market responded most directly to the new regulations with real money and voting. Short-term funds previously active in leveraged ETFs seem to have collectively waited or exited on the very first day. This regulator-led "cooling down" campaign delivered a dramatic report card on its first day. As for how the market will adapt to the new rules going forward, that remains to be seen.Even if Bitcoin rises, if trading volume and open interest do not follow, that rise is more likely a selective circulation of funds rather than real demand. How can we distinguish whether today's green candle is genuine demand or a temporary concentration on certain assets? The key facts confirmed from the original text are as follows. Some assets including Bitcoin have risen, but trading volume is weak and open interest (OI) is decreasing. This is interpreted as existing positions being closed out rather than new positions being entered. The assets leading the rise are JELLYJELLY, OPG, SLX, LAB, BSB, ALLO, CHIP, MEME, EDEN, HUMA, ZKP, METIS, etc. Meanwhile, BEAT, EDGE, COAI, TRUMP, RAVE, SPACE, SOPH, IP, AVNT, ZAMA, OFC, PIEVERSE, VIRTUAL, ACU, H, MEGA remained in a consolidation range. What this scene implies is not an expansion of risk appetite across the entire market, but liquidity flowing into some strong momentum assets.There is no direct formula for rising oil prices and BTC, but the transmission chain is worth watching: higher energy prices→ inflationary pressures may rebound→ rate cut expectations are being disrupted→ risk asset valuations are under pressure. Of course, a short-term surge in oil prices does not necessarily mean core inflation will rebound. What really matters is how long the price hikes will last and whether they will spread to transportation and services. Don't mechanically short coins just because oil prices rise; it's 😅😅😅 truly intoxicating. This is for market observation only and does not constitute investment advice. #美伊报复循环加速, oil prices rose 20% $BTC $ETH monthly Every bull and bear cycle in Bitcoin inevitably undergoes a deep reshuffle.
In 2011, 2013-15, 2017-18, and 2021-22, each bear market experienced pullbacks of around 70%-90%.
But interestingly,
Every time the market thinks "BTC is finished," it always hits new highs.
History does not simply repeat itself, but human nature always repeats.
Going crazy when it rises, despairing when it falls.
The current question is:
Will the 2025 pullback mark the start of a bear market, or a shakeout within a bull market?
Let's wait and see—the market will give its answer.The earnings reports of six tech giants are all out, with only Nvidia left, expected in August.
Currently, the results of these companies are quite interesting: Microsoft and Amazon surged; Google, Tesla, Meta, and Apple declined.
Amazon has the highest revenue, Google’s cloud business grows the fastest, but the stock price reactions are completely different.
The market’s attitude toward AI investment has changed. A year ago, as long as a giant announced increased AI investment, the stock price would rise first. Now the market is starting to check the books: Has the money spent actually turned into revenue and cash flow?
So this round of earnings reports, on the surface, is about performance, but in reality, the market is regrading the growth quality of these six companies. In this lesson, we’ll score them: two top performers, two somewhat unfairly treated, and two with their own troubles.
Two that rose: Money was spent, and it proved to be profitable
Microsoft: AI has started generating rent
Revenue $90 billion, up 18% year-over-year; Azure grew 43%, Microsoft Cloud grew 27%. More importantly, next quarter Azure guidance is about 45%, commercial contract liabilities $678 billion, up 84% year-over-year.
The logic behind Microsoft’s rise can be summed up in one sentence: they dare to spend money and can prove that the money spent is making money. Enterprises want to use AI, so they buy Azure computing power, databases, Microsoft 365, Copilot—Microsoft isn’t just selling a model, but repeatedly charging across the entire enterprise AI chain. Plus, capital expenditure and cash flow expectations are not as bad as the market feared, so investors are willing to give a higher valuation. After the earnings report, the stock rose over 15%, adding nearly $450 billion in market value in one day.
Amazon: AWS accelerates, proving computing power is truly being bought
Sales up 20%, operating profit up 43%; AWS revenue $42.2 billion, up 37% year-over-year, the fastest in over four years; advertising also rose 26%, reaching $19.8 billion.
What was the market’s biggest fear before? That Amazon would spend over $200 billion a year building data centers, only to have customers taken by Microsoft and Google. This time AWS suddenly accelerated, telling everyone: the new computing power is indeed being purchased, and supply can’t keep up with demand. So even with an annual investment plan mentioning $220 billion, the market accepted it. The stock rose over 12% pre-market after the earnings.
Spending big money, Amazon and Google’s stock prices moved in opposite directions. The difference is: one proved returns with growth, the other exposed greater cash flow pressure.
Two somewhat unfairly treated: Great performance, but burning cash too fast
Google: Did very well, but the parent was scared by the bill
Looking at the results alone, Google might be the best among the six: revenue $119.8 billion, up 24%; search up 17%; Google Cloud surged 82%, cloud business operating profit $8.8 billion.
But the stock price fell. The reason is capital expenditure was raised again: from $180-$190 billion in 2026 to $195-$205 billion, with a clear increase expected in 2027, and about $5.9 billion cash consumed in Q2 already.
The market’s thinking is easy to understand: Cloud grew 82%, yet cash flow is so tight, how much more investment will be needed? Plus, Google’s stock had already risen a lot with high expectations, so "good" is no longer enough; it has to be "better than the most optimistic expectations." Once spending is raised, investors take profits first.
Meta: Advertising is strong, but cash flow is being rapidly eaten by AI
Previously discussed Meta separately, here’s a brief recap. Revenue $60.8 billion, up 28%, ad impressions up 14%, price per ad up 12%, solid fundamentals. But costs rose 55%, EPS $6.18, below the expected $7.22; most strikingly, free cash flow dropped from $8.55 billion to $784 million, shrinking by about 90%.
The money earned from advertising is being quickly consumed by AI infrastructure. And new businesses like AI assistants and computing power rentals don’t yet have clear revenue scale. The market wants to see a clear monetization path before it’s willing to add back valuation.
Two with their own troubles
Tesla: Selling more cars, but earning less per vehicle
Delivered over 480,000 vehicles in Q2, deployed 13.5 GWh of energy storage, sales are back. But to boost volume, they continued price cuts and discounts, vehicle revenue per unit dropped from $45,345 to $42,730, operating profit only about $400 million, profit margin 1.4%, significantly below expectations.
Tesla’s current high valuation relies on Robotaxi, FSD, and Optimus robots, not car sales. These new businesses are not rolling out fast enough, so the market can only look back at the car business — and car profits are declining, AI requires heavy investment, so the stock can only fall.
Apple: Beat expectations this quarter, but the market focuses on next quarter
Revenue $109.4 billion, up 16%; EPS $2.02, up 29%; iPhone revenue $54.25 billion, up 22%. The quarterly results are not bad, but problems lie in two areas.
First, about two percentage points of gross margin this quarter came from tariff refunds, and EPS includes about $0.11 of this "extra" income — meaning part of profit growth is not from core business improvement. Second, next quarter guidance: revenue growth 9%-11%, below market expectation of 12%, and management mentioned tight supply of high-end chips, possibly limiting iPhone, Mac, and iPad shipments.
Demand remains, but supply can’t meet it, so that demand temporarily doesn’t convert into revenue. Plus, Apple’s stock has risen a lot this year with a high valuation, so slightly missing guidance leads to capital pulling out first. The stock fell about 7% pre-market after earnings.
Microsoft and Amazon: Spend a lot, but earn faster, so they rise. Google and Meta: Strong revenue, but cash burn is too large, valuation is suppressed. Apple and Tesla: One is stuck on next quarter guidance and supply, the other on profit margin.
Looking at earnings reports is never just about how much revenue grew, but how actual results compare to market expectations. Microsoft proved AI can already generate orders and cash flow; Amazon proved there is demand for expanded computing power; Google and Meta have equally excellent growth, but investment scale exceeds market’s psychological tolerance.
#财报观察员：亚马逊指引不及预期，股价却反涨9% BTC weekly volatility hits a two-year low, weekly chart may face a 10% level reversal, downside risk heating up in August.
Bitcoin has been rebounding along with those risk assets, once surging past $65,000 intraday, up 10.52% cumulatively this month. Several Wall Street traders said assets exiting around $63,000 are particularly strong, while around $67,000 there is a pile of selling pressure from those looking to take profits.
Although US tech stocks surged, crypto hasn't really followed. Right now, new USD liquidity is lacking, and capital inflow is still quite distant. Fortunately, ETFs are showing some improvement; the US spot Bitcoin ETF had a net inflow of $233 million on Thursday, and this week finally turned back to a net inflow of $203.8 million. July's total is about $438 million, offering hope to end the large outflows seen in the previous two months.
Today is the closing date for July monthly options, with the battlefield centered around $64,000. Bitcoin options expiring today total 149,000 contracts. Our biggest pain point is exactly $64,000, with a notional value of $9.6 billion. The month-end settlement combined with the monthly close is locking the price tightly around $64,000. Everyone is basically waiting and no one dares to take heavy directional bets. Weekly volatility has dropped to a two-year low, near the weekly 200 moving average. The 200 EMA and bull market support band are also converging. A decisive weekly-level move of about 10% may come next. One more note: Bitcoin's past three bear market cycles all had bearish August candles, closing down an average of 14.8%. According to historical patterns and the current bearish technical outlook, the probability of a decline in August is quite high.
$BTC $BTC #Microsoft's single-day market value increase nears 450 billion, setting a US stock record
I am Conan.
Microsoft surged over 15% on Thursday, with its market value increasing by about $450 billion in one day, surpassing Nvidia's record.
Azure's growth rate hit 43%, accelerating from the previous quarter, with annualized revenue breaking 100 billion for the first time.
Next quarter guidance is at 45%, far exceeding market expectations.
---
The market's biggest fear is AI investment going to waste, and Microsoft has answered that once and for all.
Commercial remaining performance obligations soared to 678 billion, an 84% year-over-year surge—
contracted but unrecognized revenue enough to cover more than two years.
Nadella said data center capacity will double within two years, with new lease contracts in Q2 alone exceeding 130 billion.
Microsoft's current Capex is not burning money; it's locking future money away in a drawer.
---
Even more impressive, every $1 spent on infrastructure can leverage $1.5 in cloud revenue.
Once this positive feedback is priced in, valuation logic will have to be rewritten.
Whether the 45% growth rate can be maintained next quarter is the only standard to test the quality.
But at this moment, Microsoft has proven that AI can truly monetize—
OpenAI alone is expected to contribute over $24 billion in revenue to Microsoft this year.
---
Transmission to BTC
Computing power demand explosion → storage shortage → tech stock sentiment recovery.
BTC, as a high Beta asset, benefits simultaneously, touching 65k intraday.
Conan has finished speaking. Reflect on it.
$BTC $ETH $APT has extremely low liquidity, and Aave's on-chain funds are also going to be withdrawn from Aptos, it will still drop Is Apple also unable to hold out? Despite earnings beating expectations, the stock price plunged 7%. 🫤
#苹果第三财季业绩超预期, the stock price plunged sharply after hours
Apple has given the market a lesson this time,
Following SK Hynix, it once again perfectly demonstrates that good financial reports don't necessarily mean stock prices will rise.
Third quarter revenue was $109.4 billion, up 16% year-over-year; Earnings per share were $2.02, higher than the market expectation of $1.89. iPhone revenue grew nearly 22%, Mac grew 29%, services grew 12%, and Greater China also grew 22%.
The data looks impressive, but $XAAPL dropped from around $340 to $315, hitting a low of $306.
Where did the problem lie?
Apple's unexpected performance this time was partly due to tariff refunds "helping."
The company disclosed that tariff refunds contributed about 2 percentage points to gross margin and increased earnings per share by $0.11. Excluding this part, Apple's actual earnings per share were about $1.91, just two cents higher than the market expectation of $1.89.
So this financial report did win, but without the headline, it didn't seem to win by as much.
The market is more focused on the next quarter.
Apple expects revenue growth of 9%–11% next quarter, a significant slowdown compared to this quarter's 16% growth. Before the earnings report, the stock price was already close to $340, with the market buying in early on the iPhone recovery, AI upgrades, and service business growth. By the time the actual handover comes, "meeting expectations" is no longer enough; expectations must be further raised.
Apple is also under heavy pressure on AI. This quarter's R&D expenses reached $11.7 billion, a 32% year-on-year increase, showing that Apple is indeed spending money to catch up; But the financial report still shows no independent income from AI. Whether the new Siri AI can encourage users to upgrade devices or increase subscriptions is unavailable in some countries either, so for now, it's still a story, not money on the books.
This is also why Apple and Microsoft's earnings diverged after their earnings reports. Microsoft can already prove with Azure revenue that AI is making money, while Apple can only prove it is still investing.
This financial report is also related to the recent storage market.
Rising memory prices bring profits to SanDisk, Micron, and SK Hynix, but cost to Apple. Apple had previously raised prices for some Macs and iPads due to memory shortages. This time, tariff refunds helped Apple boost its gross margin, but refunds won't be available every quarter. If storage and advanced chip prices continue to rise, future profit margins will be squeezed.
Therefore, the surge in storage stocks and Apple's decline are not in conflict. Upstream is profiting from price increases, while downstream is starting to worry about costs.
In the short term, $306 is the support level where panic buying appeared last night; For the rebound, the first target is $325, but the real major pressure remains near $340 before the earnings report.
My judgment is that Apple's business hasn't collapsed; this time, the drop is due to valuations and overly high expectations. To pull the stock price back, Apple needs to provide a more direct answer: how much revenue AI can actually generate, and how much profit will rising costs take away.
The same applies to the crypto world. The market is no longer satisfied with "good data"; capital is starting to ask where profits come from and how much longer they can grow.
$XAAPL $BTC #PCE环比转负, GDP growth slowed to 1.5%
Dual signals landed! PCE turned negative month-on-month and GDP was only 1.5%, intensifying the Fed's policy dilemma
On the evening of July 30th Beijing time, the U.S. Department of Commerce simultaneously released major economic data: the June PCE price index unexpectedly turned negative month-on-month, and the annualized GDP growth rate in the second quarter slowed sharply to 1.5%. This set of contrasting data completely disrupted market expectations for the Federal Reserve's policy in the second half of the year.
Inflation saw a phased cooldown: overall PCE fell 0.1% month-on-month in June, marking the first negative growth since the pandemic in 2020, dropping year-on-year to 3.7%, mainly due to a temporary Middle East ceasefire pushing oil prices lower and energy items lowering price readings. Core PCE, excluding energy and food, rose 0.1% month-on-month, below the market expectation of 0.2%, and slightly down to 3.3% year-on-year. However, it still falls significantly short of the Fed's 2% inflation target. Stickiness in endogenous services inflation has not been resolved, and this cooling is geopolitical and not a turning point in the trend.
Growth data appeared weaker, with second-quarter GDP annualized at 1.5%, below the expected 2.1% and the previous value of 2.1%. The drags are concentrated in surging imports, inventory reduction, and government spending cuts; Excluding disturbances, private final consumption grew by 3.9%, the highest since early 2023. Household consumption and AI enterprise investment continue to support the economic fundamentals, indicating a structural slowdown characterized by "weak outside, strong inside," not economic slowdown.
After the data was released, market sentiment quickly recovered. The CME rate tool showed the probability of a rate hike in September fell from 82% to 59%, Nasdaq futures and semiconductor sectors surged sharply, while US dollar and US Treasury yields briefly declined. However, internal divisions within the Fed remain sharp. At previous policy meetings, 9 voted to stabilize and 3 to support rate hikes. Hawkish members worry that if inflation rebounds, subsequent tightening will be more costly.
Currently, the Fed is caught in a typical dilemma: a slight cooling in inflation temporarily eases pressure for aggressive rate hikes, but core prices remain high; GDP slowdown opens up room for easing, but domestic demand remains resilient, and excessive rate cuts may reignite inflation. In the short term, growth stocks and cryptocurrencies are positive, but in the medium to long term, close attention should be paid to oil prices and August inflation data. If the energy rebound pushes prices higher, the risk of the Federal Reserve restarting rate hikes cannot be ignored, and market volatility risks remain elevated
$BTC $ETH "AMZN Guidance Misses + Negative Cash Flow, Reversing 9%: The Market Is Not Buying Earnings, It's Jassy's Comment 'Not Enough Before 2027'"
$AMZN Tonight, "expectation management" took it to the extreme.
Q2 numbers exploded: revenue 200.6 billion, up +20% year-on-year; AWS 42.2 billion, up 37% year-on-year (fastest in 18 quarters); EPS of 5.75, beating expectations of 1.82.
But if you flip to the lower half, it's full of spoilers: Q3 guidance is 197–202 billion, median 199.5 billion, below the market 203.9 billion; Capex increased from 200 billion to 220 billion; TTM free cash flow turned negative by -7.6 billion.
According to the old script, this should have dropped 5%, but after hours, it was +9%.
Why? The market didn't even look at that "guidance miss"—it was listening to Jassy's call: "Even if we add 220 billion, 2026 capacity is still insufficient; 2027 is the same, and demand in 2028 is already significant." ”
Translated: Don't ask about break-even time, first ask if production can be enough. AWS annualized $169 billion, AI business $25 billion annualized with triple-digit runs, and backlog nearly $500 billion. With supply constraints ≠ demand collapsed, Capex isn't just burning money but grabbing tickets.
Google's similar move last week (raising Capex + turning cash flow negative) fell, Microsoft (maintaining Capex increased), while Amazon rose in between—the difference isn't in numbers; at AWS, 37% has flipped back the "AI investment return rate" scale.
Old investors pour water: That 53.4 billion net profit includes Anthropic's valuation gains, and the operating profit of 27.5 billion is real money; The price increase in memory has also pushed up some CapEx, not purely driven by demand. Packaging and pricing "cost-driven scarcity" and "demand-driven scarcity" is the sugarcoating that should be peeled off the most among these 9%. $AMZN
#财报观察员: Amazon's guidance falls short of expectations, yet stock price rises 9% Compared to $OKB, Bitcoin needs to fall even further
July 31, 2026.
The progress of this bear market is relatively slow.
In the last bear market, Bitcoin $BTC fell to a relative bottom in June 2022, and in this bear market, Bitcoin has only dropped by half.
Only OKB fell sharply.
At the start of this bear market, it suddenly dropped by 77%. This makes the mid-stage and even the later stages of the bear market quite resilient to declines.
Many coins will be halved in the second half of the year, and OKB might just add a double bottom.
In bear markets, people fall quickly; in the later stage, they add a double bottom and get out of the bear market. Those who fall slowly in a bear market will continue to fall in the second half of the year.
Bitcoin falls short of falling and lacks sufficient time.
OKB falls enough but not long enough in a bear market.
If it were like the previous cycle, dropping more than 65% in June, then Bitcoin would only need to fill the double bottom by year-end.
Unfortunately, there are no ifs.
Currently, only OKB fits this 'if' scenario.
－－－－－－－－－－－－－
(07/30,26) Reading notes and reflections:
There are no perfect assets, only good quality and affordable assets.
You can't say Bitcoin will always be better than platform coins; both good and bad are temporary; the most important thing is the pricing.
Only when quality assets fall to good positions can you achieve good investment returns.
In the short term, prices are driven by sentiment, but sentiment cannot always dominate the market; ultimately, prices must revert to their means.
Don't expect an investment to always yield high returns.
Focus your energy on the buying phase; as long as your purchase price is low enough, you don't need to worry too much about selling.
The vast majority of mistakes occur when chasing gains at high levels.July 31: Micron Stock Analysis: Behind the AI Storage Frenzy, the market may have overlooked cyclical risks
Micron's current stock price is around $130, and recently it has become a hot topic due to rising demand for AI storage.
Many investors believe Micron is the new winner in the AI wave.
The logic is simple:
AI data centers require massive storage, HBM demand is growing, storage prices are rising, and Micron's profits have improved.
But I believe the market may have already traded too much in advance for expectations.
Micron's biggest problem is that it remains a cyclical company.
The biggest feature of the semiconductor storage industry is its very obvious profit fluctuations.
During industry downturns, the company significantly cut capital expenditures;
After the industry recovered, prices rose;
Improved profits will attract increased production capacity.
This cycle has persisted for decades.
The market's higher valuation of Micron now largely stems from the AI narrative.
But the problem is that the biggest beneficiaries of AI are still concentrated in the core supply chains of GPUs, high-end packaging, and HBM.
Storage demand is growing, but whether profits can sustain high growth over the long term still needs further verification.
If AI investment growth slows or storage supply recovers, the market may reassess Micron's value.
My view is biased.
Not because the company is doing poorly, but because current prices already reflect a lot of optimistic expectations.
For cyclical stocks, the most dangerous times are often when the industry looks its best $MU Google's financial report doesn't focus on revenue in the storage sector, but on a "run"
Many people are watching Google's revenue grow by 24% and Cloud Cloud's 82%, thinking 'Wow AI is amazing.' But the veteran in storage has only one number in mind: Capex has been raised again.
What does adjusting Capex mean? This means Google thinks if they don't spend money now to grab hardware, they'll have to wait in line next year.
There is a harsh reality in today's AI infrastructure: GPUs can be replaced (AMD, self-developed), but HBM and DRAM have no alternatives. HBM is controlled by SK Hynix and Samsung, with limited capacity. If Google orders one more car, Microsoft loses that car. This isn't procurement—it's a supply chain squeeze.
The consequence of a run is that storage factories are not only selling goods but also picking customers—whoever offers more stable long-term orders and higher prices will get the goods. So Google's Capex hike is essentially telling storage manufacturers: "Put me first, money isn't the problem." ”
As for SSDs and NAND, those are byproducts. When the server is full, training data, inference logs, and video corpora are stored—all at the petabyte level. The more cloud customers there are, the thicker the cold data stacks, and storage consumption becomes a bottomless pit.
But there's a hidden trap: the biggest fear during price hikes is placing repeated orders. Nowadays, everyone is scrambling for goods, and actual demand may be inflated. If a cloud company suddenly claims AI revenue falls short of expectations or cuts Capex, inventory will backfire. So this round of rebound is fueled by "panic over buying stock," not "performance delivered."
My view can be summed up in two sentences:
· In the short term, the momentum of Capex upward adjustments remains, and the storage sentiment won't die off anytime soon.
· In the mid-term, keep a close eye on the pace of HBM capacity release. If SK Hynix's new plant starts production in Q1 next year and the shortage eases, then the peak of this cycle will be reached.
To put it bluntly, this round of storage rally isn't about how powerful AI is, but about how long the giants can compete for the next step. The day the false start ends is when the real disagreement begins.
？$YGG As a long-established Web3 token, few people pay attention to this coin. Recently, the team disbanded its core department and established related departments, mainly focusing on AI game creation. Every time the market drops $ETH and then rebounds, YGG has always been very strong. Today, I saw YGG still at a historic low, and I think it could earn several times the profit in a week.Single-currency contract movements
$SNDK Futures trading starts to accelerate, with price, position, rate, and active orders all viewed together.
Prices rise but positions shrink; 15M readings +0.81%/-1.02%, understood as a rebound after reduced positions. Active buying accounts for 56.9%. Buying back can push prices higher, but to go further, new buying demand will follow.$AEON As mentioned, don't go long. For spot traders, wait for a light position around 0.05 to buy a light. Around 0.03, you can buy a bit heavily. Don't buy below 0.02. This level is also hard to reach. The delisting threshold is the threshold. If the spot position gets stuck in a year, it will be lifted in at most a year, so there should be decent returnsBiggest mistake right now? Calling every green candle a bull run.
This isn’t a broad rally. It’s selective rotation. Money is piling into a handful of names while most alts are still stuck at the lows. Don’t let the index fool you.
💰 Seeing clear inflows: $BTC, $JELLYJELLY, $OPG, $SLX, $LAB, $BSB, $ALLO, $CHIP
👀 On watch: $MEME, $EDEN, $HUMA, $ZKP, $METIS
🏆 Market backbone: $BTC — liquidity anchor | $ETH — ETF/institution favorite | $SOL — high-beta leader | $TAO, $WLD — AI core | $HYPE — risk appetite gauge | $DOGE, $ZEC — retail sentiment
📉 Still weak: $BEAT, $EDGE, $COAI, $TRUMP, $RAVE, $SPACE, $SOPH, $IP, $AVNT, $ZAMA, $OFC, $PIEVERSE, $VIRTUAL, $ACU, $H, $MEGA
Long-term winners don’t chase pump noise. They follow the flow before the crowd notices, wait for confirmation, and protect capital.
Core play: track liquidity, wait for signals, hold your line. Patience gets paid. FOMO doesn’t.
Not financial advice. DYOR.
#DailyOrbit @OKX Orbit
#SoftPCEStrongDemand
#AMZNMissesButRallies The ninth market-wide circuit breaker of the year was suddenly triggered, with South Korea's Finance Minister Koo Yoon-chul bowing in the National Assembly to apologize. In this wave of Korean stock liquidations triggered by liquidations triggered by liquidations of Samsung and SK Hynix's single stocks with double leverage ETFs, we are facing an extremely absurd split: on one side, SK Hynix's Q2 operating profit soared 557%, setting a record high, while Samsung's net profit exceeded expectations and loudly proclaimed strong demand for AI server HBM chips in the second half of the year; On the other hand, the secondary market is mercilessly slaughtered by valuations, with countless retail investors who firmly believe in fundamental leveraged ETFs losing everything in liquidated positions. While industrial orders loudly announce prosperity, secondary market prices are fiercely crushing valuations. If you ask me, in this battle between the physical world and financial screens, which side of the signal is more worth trusting with real money? My answer is extremely clear: I firmly believe in strong demand signals from the industry side, and the crash in the secondary market is merely financial noise created by lever stomping and liquidity vacuums. 1. The wave of leveraged liquidations has distorted fundamentals into a withdrawal machine. We need to clarify the timeline of this round of sharp declines to see the truth. At the end of May this year, South Korea boldly launched a double-leveraged ETF tracking a single tech giant to boost retail trading. At that time, AI was at the peak of its popularity, and countless retail investors, driven by greed, used these leveraged tools to frantically overbuy Samsung Electronics and SK Hynix. But they overlook one fundamental chip rule: when high-leverage assets fall short of expectations during a tightening cycle lacking liquidity,#美股加密标的承压, coin price fluctuations impact earnings $Coinbase diversifying revenue sources beyond Bitcoin trading
Despite the company reporting strong fundamentals, Coinbase's stock price fell about 7% in after-hours trading, indicating investors remain cautious about the broader market environment rather than questioning the company's specific execution capabilities. The exchange successfully reduced its reliance on spot Bitcoin trading, with 88% of its net revenue now coming from other sources, including a prediction markets division whose quarterly revenue doubled quarter-over-quarter, with annualized returns exceeding $100 million. Additionally, Coinbase holds a record 10.3% share of global crypto trading volume and holds over 30% of all circulating USDC.
For investors, these data confirm that Coinbase has effectively executed its "Everything Exchange" strategy to stabilize returns amid Bitcoin price fluctuations, although the immediate stock price reaction suggests the market will wait for these new businesses to achieve sustained profitability before revising the stock's valuation.I just saw a major piece of news: the probability of the Federal Reserve raising interest rates again this year has soared to 69%! The entire crypto market is already short on cash, with all the capital drained by US stocks and AI. If a rate hike really hits in the second half of the year, liquidity will likely tighten even further. So from the perspective of risk assets, the crypto world is indeed uncomfortable in the short term. Rising expectations for rate hikes mean higher US interest rates and higher funding costs, putting pressure on leverage and speculative sentiment in the market. Highly volatile assets like Bitcoin and Ethereum are often the most sensitive to liquidity. At the slightest disturbance, they first cut valuations, then contracts, and finally fundamental screening. However, there's no need to sentence people to death just by seeing a "69% probability of rate hikes." This data itself comes from the prediction market and reflects more of sentiment and position games, not that the Fed has already made a decision. And what truly determines the market trend are the upcoming inflation, employment, earnings season, and speeches by Fed officials. If economic data weakens, rate hike expectations may quickly decline, and risk assets may instead rebound. In the short term, the more stable approach is: don't easily add high-leverage or blindly bottom-fish when negative news ferments. Prioritize key support for $BTC/$ETH, stablecoin inflows and outflows, and whether the profit-making effect in US tech stocks and AI sectors has faded. To put it plainly, now is not a "sure winter" but a "cold wave warning." The real danger is not the rate hikes themselves, but the market lacking liquidity on one side and piling up large amounts of high-leverage positions on the other. As long as leverage is cleared and expectations are highUS chip stocks collectively ignited the market, silver quietly followed suit, but what really kept me awake wasn't the rise and fall, but the subtle shift in the position structure. Have you ever wondered how much expectation has already been stuffed into the price when everyone is staring at the same positive narrative? Tonight's market was actually quite straightforward. US stocks strengthened across the board, with the memory chip sector leading. SanDisk, SK Hynix, and Micron all surged simultaneously, and the demand narrative for AI infrastructure heated up again. Silver, as a dual-attribute product of industrial + precious metals that I have been tracking for a long time, is also rising as risk appetite warms. On the surface, it seems like the chain of "AI driving storage, storage driving silver" is fermenting. But if we only look at the surface, it's easy to overlook a more crucial factor: how the derivatives market is pricing this rally. My observation is that in this round of rallying, changes in option skew and forward curve patterns are more worth pondering than the spot price itself. When memory chips and silver both strengthen, the implied volatility of call options in the market rises noticeably faster than the bearish ones, indicating that funds are not simply chasing gains, but are using derivatives structures to express a "fear of missing out" mentality. This structure often means the market hasn't reached its most crowded yet, but it also reminds us that once expectations are fully priced in, pullbacks can become very elastic. Because of this, my operation today was actually simpler. All Martingale orders took profits as planned, with no tracking of any unusual coins or relaxation of position management just because the account hit new highs. An 89.39% historical win rate isn't built on luckDaily data is out: BTC stands above 65,118, up 2.14%, while ETH shares the same price at 1,928. The Nasdaq surged 2.8%, the storage sector rebounded violently, SanDisk rose over 17%, and SK Hynix surged 24% to 25%. The market is verifying all previous judgments.
BTC is approaching the short liquidation zone, and the direction is about to be decided
BTC is currently at 65,118, with a large number of 50x and 100x short liquidation orders gathering in the 65,400 to 66,000 range above. If it breaks above 65,500, short closing will amplify upward momentum. Short liquidations above are about $400 million, while long positions below are about $420 million, with both sides nearly even. Long and short positions are in the same direction, but short positions have higher leverage, so once a breakout occurs, the stamping speed will be faster.
The tech stock market has reached a turning point, and delivering on earnings is the real deal
Microsoft rose over 15%, Azure cloud revenue grew 43%, surpassing $100 billion for the first time this fiscal year. Amazon rose about 9%, AWS grew 37%, and capital expenditures were raised to $220 billion. The two giants have proven their solid results to the market that AI is not a bottomless pit that burns money. Meta fell about 8.6%, with future spending commitments approaching $700 billion. The market is using prices to tell you that those who keep dreaming will be abandoned.
The storage sector's short squeeze confirmed previous judgments
SanDisk rose over 17%, and SK Hynix surged 24% to 25%. Choi Tae-won's first personal purchase, expectations of a storage supply shortage, and Microsoft's earnings report as catalysts combine to drive three factors. The 1046.62 long position was pulled out from the deepest panic, with a single-day floating profit exceeding 1500 points. The storage sector is not a rebound, but a short squeeze.
Operationally
Continue holding long positions below 63,000, with stop-loss moved up to 63,500. The above range between 65,400 and 66,000 is the short liquidation zone. If volume breaks through, BTC is expected to test 67,000. If this range is blocked, there may be a short-term pullback, but the medium-term upward direction remains unchanged.
The short squeeze in the storage sector proves that the fundamentals of AI hardware have not collapsed; the AI narrative has reached a watershed moment, not the end. Hold onto your positions, don't let fluctuations scare you away.
。 Think carefully. #PCE环比转负, GDP growth slowed to 1.5% #财报观察员: Amazon's guidance fell short of expectations, but its stock price reversed by 9% #苹果第三财季业绩超预期, with post-hours shares plunging $BTC $ETH The local decline in $SNDK is -42% and the biggest rebound we've had since its inception is +14% (which is very limited)
Yesterday's reaction to the lack of increase + good capital expectations from META/MSFT is not very constructive on memory names.
At this stage, I wouldn't be surprised if we have another leg down due to the severity of the market deterioration.
Not comfortable at all, but I think we'll see a 20%+ recovery soon anyway. (Don't try to time the bottom using leverage unless you're a professional trader.)
As for long-term allocation, my positive outlook on memory hasn't changed, but the PA indicator now will be very different from just the April-June rise.
After this kind of sell-off, the market needs time to digest and won't just reverse in a V shape.
Make sure you have a plan and stick to it... GL$SNDK #SoftPCEStrongDemand A Morgan Stanley-led syndicate is pouring $1.5 billion in structural concrete for a 1.6GW captive power plant and Texas campus—$1.4 billion of which is bridge loans. This isn’t ordinary data center leasing; it’s driving a deep rock-embedded pile for Anthropic’s computing power needs. Google didn’t just sign a purchase order; it’s taking on all the rental and electricity default risks for Anthropic onto its own balance sheet in exchange for 20% equity. This move is called "trading credit for load-bearing walls"—have you ever seen a developer use payment guarantees to exchange for the owner’s property rights?
Breaking down this blueprint: a 1.6GW captive gas power plant means this site doesn’t intend to rely on the municipal grid at all. This is an independent energy supply system, the lifeline of the building complex. Google’s TPU chips are anchored by long-term leases inside this "digital skyscraper." Anthropic is no longer renting a "standard factory" in the cloud; it wants a high-rise tower customized to its own column grid spacing—10GW planned for the long term, essentially drawing a vertical city out of the desert.
Note the key move: Nexus, as the developer, secured bridge loans, meaning it locked in cash flow before the main structure topped out. Google’s role isn’t just a contractor; it’s the general contractor, property operator, and invisible shareholder. It uses an irrevocable letter of credit to pre-cast the "utilization rate" of this data center into certainty. Traditional infrastructure looks at pre-sale certificates; AI infrastructure looks at "lease guarantees," essentially securitizing computing power demand into a long-term promissory note.
Now the tower crane turns to the AI infrastructure template: private credit institutions start risk assessments based on the combined structure of "data center + power plant + chip leasing contracts." Morgan Stanley’s syndicate no longer focuses on debt-to-asset ratios but on the TPU chip depreciation curve and the "building load" measured by Anthropic’s model training burn rate. It’s like doing wind tunnel tests for skyscrapers, except this wind tunnel is simulated by GPU clusters.
The metaphor hidden in $XNFLX is even more chilling: the once streaming giant is now just an empty shell name, while those truly profiting from the "data real estate" dividend are the capital holders with energy assets and chip orders. They swap debt for equity, leverage guarantees to seize control, and coat every default insurance as fireproof paint on load-bearing walls.
As the construction countdown begins, the shadow of the tower crane points to a harsh truth: the ultimate form of AI infrastructure isn’t the server racks behind glass curtain walls but a main beam cast from computing power leases, energy contracts, and financial engineering—this main beam dares to be exposed in storms because it knows the real load-bearing structure is never on the blueprints. #googlebacksaiinfra Market hotspots are quietly shifting, traditional counterfeit narratives lose their appeal, and on-chain tokenized financial asset tracks are attracting incremental attention—specifically, the TradFi sector. $XSOXL, benchmarked against the US 3x Semiconductor ETF, saw a strong rebound after a deep decline, rising over 12% in a single day and becoming a major highlight of today's session. Tokenized stocks connect the secondary US stock market with on-chain funds, enabling 24-hour uninterrupted trading and attracting some cross-market capital that covers both US and crypto markets. However, these leveraged tokens naturally have amplified volatility attributes, with price fluctuations far exceeding those of the underlying asset. In the short term, the market will rely more on oversold recovery, while the medium- to long-term trend will be closely tied to the prosperity of the US semiconductor sector. The TradFi sector is still in its early stages, with overall weak liquidity. In the short term, it is suitable for trading swings, making it difficult to break out of a long-term trend.🚨 Wall Street is celebrating. Bitcoin is barely reacting. That divergence may be the market's biggest story.
Today's macro data reinforced the soft-landing narrative:
📊 PCE came in negative month-over-month.
📈 GDP printed 1.5%.
🛒 Amazon rallied despite weaker guidance.
🇰🇷 KOSPI posted a record single-day gain.
Traditional markets are acting like risk is back on.
Bitcoin isn't.
With $BTC trading around $63.7K, crypto has remained relatively flat while equities continue to push higher.
For years, the assumption was simple:
«If the Nasdaq rallies, Bitcoin follows.»
That relationship appears to be weakening.
Rather than moving in lockstep with stocks, Bitcoin may increasingly be responding to its own fundamentals—liquidity conditions, institutional flows, ETF demand, on-chain activity, and crypto-specific capital rotation.
The next few weeks could be more important than today's headlines.
The key question isn't whether equities can keep climbing.
It's whether fresh capital starts flowing back into Bitcoin.
Watch the liquidity. Follow the flows. Let the market confirm the trend.
Not financial advice. Always do your own research.
$BTC #Bitcoin #Crypto #DailyOrbit #Markets #Investing #OKXOrbit#DailyOrbit 📊 Tokenized Stocks Market Update
Top Performers:
• $XLITE : $729.48 (+2.07%)
• $xTSM: $419.43 (+1.68%)
• $xMETA: $550.47 (+1.35%)
Under Pressure:
• $XMSTR : $95.33 (-2.64%)
• $XAAPL : $309.87 (-0.98%)
Overall Sentiment:
Tech and chipmakers lead the green list today, while high-beta crypto-adjacent equities see a minor pullback. Manage your risk closely!
NFA – Educational purposes only.
#SoftPCEStrongDemand #AMZNMissesButRallies #MSFT450BInADay Complete analysis of Bitcoin (BTC) market performance on 2026.07.31
1. Current Basic Data on the Board
The current price is about $64,800, up 1.5% in 24 hours, with a slight rebound for three consecutive days;
1. Derivatives: 24-hour net liquidation totaled $147 million, short liquidation $92.3 million, short-term short leverage concentrated liquidation, passive buying pushed prices up;
2. Institutional funds: On July 30, spot ETFs ended a four-day streak of net outflows, with a single-day net inflow of $32.1 million, but the five-day average still showed a net outflow of $494 million. Only BlackRock managed the outflow, while the rest continued to lose blood, with significant divergence in incremental funds.
3. Market sentiment: The Fear and Greed Index is 28, indicating extreme fear. Retail investors' trading willingness is low, and no followers follow the trend during the rebound;
4. Technical indicators: RSI 49.7 is neutral, MACD is slightly bearish, and there is no clear one-way trend signal in the short term.
2. Key Support/Resistance Levels
Support range (from strong to weak)
1. $63,300 (50-day moving average, core support level for this round of consolidation; if it falls below it, the rebound logic will fail);
2. $62,500-$63,000 (previously concentrated dip-fishing zone, strong mid-term support);
3. $60,000 (psychological threshold, long-term institutional heavy position).
Pressure range (from strong to weak)
1. $65,000 (short-term first level, accumulated short-term unwinding pressure, multiple rallies failed to hold steady);
2. $67,000 (20-day moving average, medium-term trapped market concentration zone);
3. $71,700 (200-day moving average, the long-term trend dividing line).
3. Bullish Logic (Supporting Short-Term Rebound)
1. Concentrated short leverage liquidations release short-term selling pressure, temporarily easing selling pressure on the market;
2. The Federal Reserve has maintained high interest rates this time and has not implemented any rate hikes, thus avoiding the impact of extreme negative factors;
3. BlackRock ETF continues to flow steadily, with long-term institutions positioning in batches on dips, solid buying in the 60,000-63,000 range;
4. U.S. tech stocks report earnings beyond expectations, with global risk appetite slightly warming and driving a simultaneous recovery in crypto assets;
5. The narrative of scarcity after four years of halving remains effective in the long term, with many long-term holders locked up and circulating tokens continuously decreasing.
4. Core Bearish Logic (Suppressing Major Rally Momentum)
1. Severe hawkish divisions within the Fed, with three members voting in favor of rate hikes. The market prices a 59% chance of a rate hike in September, and a high interest rate environment will divert funds into Treasuries for the long term, suppressing risk asset valuations;
2. Inflation resilience has exceeded expectations, with rebounds in consumer and energy prices, the Federal Reserve rate cut has been delayed, and there is no liquidity easing benefit;
3. The probability of the regulatory bill CLARITY being implemented before the August recess is extremely low, and ongoing policy uncertainty is forcing institutions to increase positions on a large scale;
4. Funds are structurally diverging, with only a single ETF inflow, most institutions continuously reducing positions, and the rebound lacks new capital support, making rallies very easy to pull back;
5. Seasonal historical patterns show that August and September are the traditional bear market downturn window, with the risk of a final deep pullback in historical cycles.
5. Three Types of Market Outlook Scenarios (Probability Ranking)
1. Neutral oscillation (45% highest probability)
The $63,300-$67,000 range is moving sideways and consolidating, with existing funds competing, following US stocks and Fed news with slight fluctuations, no one-sided sharp rises or falls, continuing until the September policy meeting;
2. Weak Pullback (35% Probability)
Unable to hold above the 65,000 resistance level, bullish momentum is exhausted, effectively breaking below the 63,300 support, triggering programmed stop-loss losses, testing the 60,000-62,500 range;
3. Strong rise (20% low probability)
Single-day large ETF net inflows + Federal Reserve signals rate cuts, with volume surging above $67,000, opening upside to challenge above $70,000.
6. Summary and Risk Reminders
1. The current three-day rally is a recovery driven by short liquidation, not a trend reversal. The lack of incremental funds and easing policies is a core positive factor, so the sustainability of the rebound is questionable;
2. The market is highly volatile, the risk of contract margin liquidation is extremely high, and there is no domestic regulation to provide a safety net. If the platform is banned or the market crashes, the principal cannot be recovered;
3. In the short term, the market mainly fluctuates within a range, with 65,000 above and 63,300 below as key dividing lines. There is no clear one-sided trend, making aggressive speculation unsuitable.BTC $63,718, 24h -1.36%, Panic Index 25. **This market isn't cold; it's already completely cold. ** But not all places are cold.
Looking at OKX's tokenized US stock sector: 5 gains, 0 losses, average +18.12%. XSNDK (3x short Nasdaq) +28.85% for one day, XSOXL (3x long for semiconductor) +34.29%. **Total transaction volume $45.9 million, money is being pumped out of BTC and ETH to play Wall Street's leveraged game. ** 🚀
Why? Because BTC has been stuck in the $63K-$66K range for 13 days, with volatility so low it makes you want to sleep. Meanwhile, in the US market, AI narratives + earnings season + Fed expectations swing — it can kill people three times in a day. **Shorting the Nasdaq earns 28% in a day, who would still keep BTC sideways with you? **
Looking at RWA perpetual contract trading volume, it's close to catching up with BTC. Aave plans to shut down 6 blockchain markets and withdraw 50 low-usage tokens. **The death knell for fake has already sounded, and even established DeFi companies are struggling to survive. **
The flow of funds is clear: from BTC/ETH → tokenized US stocks/RWA. This is not a bull market correction, but a structural shift.
**Are you still holding BTC for a breakout? Qian had already run out to find his new father. ** By the time you figure it out, you probably won't even get the chance to collect the body. 🐸After the South Korean stock market plunged 43%, it recovered nearly half of the lost ground in one day
This "V-shaped reversal" taught all traders a lesson
The South Korean stock market has recently staged a textbook-level "V-shaped reversal."
Just a few days ago, the South Korean KOSPI index plummeted more than 43% from its June high, wiping out hundreds of billions of dollars in market value in about 40 days, triggering circuit breakers multiple times and plunging the market into panic.
However, just one day later, the market sentiment suddenly reversed.
The KOSPI surged more than 16% in a single day, marking a historic-level rebound. South Korea's two major heavyweight stocks—Samsung Electronics and SK Hynix—rose approximately 20% and 25% respectively, driving the entire index sharply upward.
What caused the sudden surge? There are several main reasons:
1. U.S. tech earnings far exceeded expectations
Especially Microsoft and other AI leaders reported impressive results, reigniting global AI investment confidence. The optimistic sentiment in the U.S. stock market quickly spread to the Asia-Pacific market.
2. South Korean semiconductor leaders saw massive capital replenishment
Samsung Electronics and SK Hynix, as the global leaders in memory chips, are core components of the AI computing power industry chain. When capital replenished, they became the main forces driving the rebound.
3. Regulators stepped in to stabilize the market
South Korean regulators took measures to stabilize the market regarding high-leverage ETFs, easing the panic caused by the chain liquidation of leveraged products.
4. Short covering + bottom-fishing capital inflow
After continuous plunges, shorts took profits and bottom-fishing capital rushed in, with these two forces jointly pushing the index into a technical surge.
What insights does this offer for the crypto market?
This V-shaped reversal in the South Korean stock market closely mirrors scenarios often seen in the crypto market:
Panic during a crash is real, but the end of panic is often the start of a reversal.
Leverage is a double-edged sword—the high-leverage ETFs in the Korean market amplified both the decline and the rebound. This is exactly the same logic as contract liquidations in the crypto market.
Leading stocks drive the rebound—just as Samsung and SK Hynix led the KOSPI rally, crypto rebounds often start with BTC and ETH.
Narrative-driven capital—AI narratives ignited semiconductor stocks, just like every crypto bull market is supported by a core narrative. 🇨🇳🇺🇸 China stopped buying oil, and this has a much greater effect on lowering prices than anything the US Navy has done
China's crude oil imports have dropped by more than 40% compared to last year, yet the economy has remained operational.
Beijing burned more domestic coal, drawing from its own reserves and relying on its large electric vehicle fleet to fill the demand gap.
Goldman Sachs analysts noted that even as gasoline consumption decreased, traffic congestion remained normal, indicating where all those trips ultimately went.
At the same moment that about one-fifth of global supply is trapped in the Strait of Hormuz, the world's largest buyers were "pulled out" from the market, explaining why the originally predicted price surge never materialized.
This would probably sound unpleasant in Washington.
Part of the U.S. is trying to prove it can still secure maritime routes, but in reality, what supports the global economy is a Chinese-style industrial policy built for China's own reasons.
And this is only temporary. Purchases in July this year have already started to pick up.
$CL $META Q2 2026 Financial Report: Strong Growth Momentum, Pressure on Earnings Quality
Core Information
1. The divergence between revenue growth and net profit decline
Total revenue was $60.8 billion (YoY +28%), but net profit was $15.84 billion (YoY -14%)
Main cause: $2.4 billion in legal costs and $1.18 billion in redundancy and severance pay
This reflects that profit growth is insufficient to offset one-off expense shocks
2. Advertising business competitiveness continues to improve
Family of Apps had 3.6 billion daily active users (YoY +3%), with a solid network effect
Ad impressions +14%, unit price +12% (both volume and price increases), with AI algorithm optimization making a significant contribution
Q3 revenue guidance is $61 to $64 billion, maintaining strong momentum
3. Capital expenditures have increased significantly, and free cash flow has been damaged
The full-year 2026 Capex guidance has been raised to $130–145 billion
Operating margin fell to 31%, with free cash flow of only $784 million
High-intensity capital investment puts significant pressure on short-term cash flow
4. Reality Labs continues to suffer widening losses
Quarterly operating loss of $4.62 billion, with unclear profit paths
Further eroding free cash flow
5. Key risk factors
If the $100 billion Capex does not realize accordingly, long-term ROIC faces downside risks, with potential financial losses due to regulatory review and legal proceedings targeting teenagers
➡️ AI-driven advertising business is driving both volume and price growth, forming competitive barriers, but litigation costs, massive losses in the metaverse, and capital expenditures worth hundreds of billions are putting significant pressure on short-term profit margins and free cash flow. Whether long-term capital returns can be achieved remains to be seen.
#微软单日市值增近4500亿, setting a record for the US stock market Coinbase $COIN released its Q2 earnings report, which fell nearly 7% in after-hours trading.
After reading it, it's actually not a bad idea. In recent years, Coinbase has focused on stablecoins, custody, subscriptions, and prediction markets, always trying to shed the label of "profiting from fees by speculating on coins."
As a result, as crypto trading cooled down, revenue immediately plummeted. Q2 total revenue was $1.22 billion, compared to about $1.29 billion the market was waiting for it. Among them, transaction revenue fell from $756 million last quarter to $599 million. In three months, they lost nearly $157 million.
Even more awkwardly, Coinbase's market share rose from 9.1% to 10.3%. Simply put, more people are trading on Coinbase, but the entire crypto community is not as enthusiastic about trading anymore. The market size shrinks, even if the share increases, it doesn't help, and the final fees are still low.
Subscription and service revenue was $555 million, accounting for 48% of net income, which seems to be holding up half the sky.
But this segment of revenue also fell by 5% quarter-on-quarter. The higher proportion is partly due to faster declines in transaction revenue.
$USDC It's about the same here.
The average $USDC users spent on Coinbase products reached $20 billion, a new high, while stablecoin revenue dropped from $305 million to $292 million. If you put in more money, when interest rates drop, Coinbase will still earn less. The market is indeed rising rapidly, with contract count and revenue more than doubling compared to last quarter, and annualized revenue exceeding $100 million.
But at an annualized 100 million yuan per quarter, that's roughly 25 million yuan. Trading revenue dropped by 157 million yuan in a quarter, and this new amount of money still can't be replenished. So Coinbase's current problems are quite clear. New businesses have been launched, and indeed some people are using them, but the money earned is still not enough.
Adjusted EBITDA is still $208 million, maintaining positive results for 14 consecutive quarters. On paper, it lost $360 million, including about $210 million in cash investment losses and over $52 million in restructuring costs, but the company's daily business hasn't suddenly collapsed. But $COIN has already risen quite a bit, and everyone is buying the story of "making a lot of money without relying on coin trading fees in the future."
Now the story is there, but the money hasn't fully caught up yet. Coinbase wants to change its job and has put all its new business on the market. But as long as $BTC isn't in the market, the financial reports will immediately look bad.