Orbit Post Sitemap
Yushu will be listed on the STAR Market on August 19, 2026, with an issue price of ¥150.80 per share, corresponding to a market capitalization of approximately ¥60.993 billion, and is expected to raise about ¥6.099 billion.
Why am I bearish in the short term?
The core reason is just one sentence:
Good company + high expectations + high valuation = very likely to become a good company with a bad price.
Yushu's net profit excluding non-recurring items in 2025 is about ¥591 million. Based on the issue market value of ¥60.993 billion, the corresponding price-to-earnings ratio at issuance is about 219 times.
This means the market has already priced in:
"The future of robotics will definitely explode, Yushu will definitely be the leader, and future profits will grow rapidly."
The problem is:
These futures have already been bought in advance by the price.
⸻
More dangerously: profit growth in 2026 does not keep up
This is my main concern.
First half of 2026:
* Revenue about ¥1.152 billion
* Year-on-year growth 48.54%
* Net profit excluding non-recurring items about ¥244 million
* Year-on-year decline 19.34% Bitcoin broke above $64,000, but do not rush to celebrate yet. 📊
ETH is below 1900, XRP is under $1, and BNB and DOGE remain flat.
Price is still stuck in the $62,000–$65,000 range below key moving averages.
Heavy unlocks arrive August 20 for KAITO ($9-11M) and LayerZero ZRO ($19M).
Do not mistake this solo move for a full market breakout. Observe first.🟢 #CRYPTO CASH FLOW net inflow of +$35.1M in the latest session | BTC: +$137.3M (08/17) 📊 TODAY ▪️ Total latest session: +$35.1M (net inflow) ▪️ Stablecoin: +$35.1M ▪️ Same-day ETF: n/a — Latest ETF is T-1, not combined with the same day 📊 LATEST ETF (08/17) ▪️ BTC: +$137.3M (08/17) ▪️ ETH: +$5.0M (08/17) ▪️ Altcoin: +$0 (08/17) ▪️ Cumulative ETF in window: +$860.5M 📊 WINDOW 07/29-08/18/2026 · 21 sessions ▪️ ETF: +$860.5M ▪️ Stablecoin: −$1.83B ▪️ Total net cash flow: −$970.5M Now calmly analyze
Where are the AI opportunities in the next phase?
Recently, I researched the upstream and downstream industrial chains and related companies of indium phosphide, CCL, supercapacitors, heat dissipation, and MLCC.
I feel that materials will be the biggest bottleneck for AI in the future. The more core products move upstream, the more they are controlled by overseas companies, especially Japanese companies, and the longer the expansion cycles.
For example, the evolution in the optical field from optical modules to optical chips to indium phosphide materials, and the upstream materials of PCBs are similar. Although PCB expansion cycles are fast, the verification threshold for upstream materials is high.
The ABF film is even more difficult to replace across the entire industry chain from process to equipment to materials. In the heat dissipation field, graphene, diamond, and diamond copper continuously iterate and upgrade existing materials to improve heat dissipation performance.
The expansion of supercapacitors is greatly restricted by the supply of porous carbon materials from Japan's Kuraray materials. The upstream release films of MLCC have very high thresholds for high-end release films, with a single production line investment of several hundred million yuan,
and it takes more than one or two years to put into production and installation. Price increases in the future are highly probable. A batch of companies related to this will grow into major stocks.
Another visible opportunity is at the system level. Previously, one company handled PCB and optical modules, which was relatively simple to iterate. Later, CPO, liquid cooling, power, and TOKEN factories are all system-level comprehensive capabilities.
Companies emerging in this area will have great opportunities. I will elaborate on this field another time.
No need to worry too much about macro and market trends; focus more on segmented industries and company performance progress. This might be the best $BTC $BTC chart you'll see today. Take a close look at what happened near all previous all-time highs.
Bitcoin repeatedly breaks through previous ATHs but then sharply reverses, liquidating traders who had bet on further gains.
In other words, Bitcoin's price history is full of traps and false breakouts.
$ETH has consistently repeated this pattern for years, eventually causing many investors to completely give up on the market. And whenever the market tries to push higher, there's already a large liquidity pool waiting below.
As long as the market hasn't fully cleansed excessive bullish sentiment and leverage, forming sustained upward momentum becomes much more difficult.
This pattern has repeatedly appeared throughout Bitcoin's history, and it is likely to continue. Because ultimately, what drives the market is liquidity, position sizing, and human behavior.#高盛称美联储9月加息可能性非常低 Xiaomi's official Q2 2026 earnings announcement is now live, with the management call scheduled for 7:30 PM tonight. Once the numbers came out, the overall signal from the financial report was quite clear. Xiaomi's Q2 revenue was 108.92 billion yuan, down 6.1% year-on-year but up 9.9% quarter-on-quarter, below the market consensus of about 112.2 billion yuan; Adjusted net profit was 6.22 billion yuan, down 42.6% year-on-year, also below the market expectation of about 6.6 billion yuan. Cumulative revenue for the first half of the year was 208.06 billion yuan, down 8.4% year-on-year, and adjusted net profit was 12.29 billion yuan, down 42.8% year-on-year. Looking at the headline alone, this is a report with a significantly lower base than the same period last year, with particularly obvious profit pressure. Breaking down the business structure, Xiaomi is undergoing a significant revenue restructuring: smartphones and domestic IoT are shrinking, automobiles are rapidly expanding their revenue share, the internet continues to provide high gross cash flow, AI is entering commercial revenue, and R&D and automotive investments keep short-term profits suppressed. The most noteworthy aspect of this financial report is that Xiaomi has gradually shifted from a company that relies solely on smartphone sales to a stage where its profit structure is jointly determined by smartphones, AIoT, internet, automobiles, and AI. In the second quarter, smartphone × AIoT revenue was 84.03 billion yuan, accounting for 77.1% of total revenue, compared to 81.7% in the same period last year; Smart electric vehicles,$BTC US Treasury market suddenly sounds the alarm
The yield on the 30-year US Treasury rose above 5.31%, reaching about 5.33% intraday, hitting the highest level since 2007
This is not ordinary volatility; global capital is repricing the long-term risks of the US
The most contradictory part of the market now is: the dollar is weakening, short-term rate hike expectations are cooling, but long-term Treasuries continue to be sold off
Why is no one willing to buy long-term bonds at low yields?
Several major pressures are at play: the US fiscal deficit continues to widen, the scale of Treasury issuance remains high, tech giants are increasing financing for AI capital expenditures, and inflation concerns arise as oil prices climb back above $90
In other words, the market may not be too worried about the Fed continuing to raise rates in the short term, but is increasingly concerned about fiscal, inflation, and debt supply issues over the coming decades
This is the real reason why long-term rates and short-term expectations are diverging
For the market, 5.31% is not an ordinary number
If long-term rates keep rising, it will directly increase funding costs across the financial system and depress valuations of future cash flows. High-valuation tech stocks, growth stocks, and liquidity-sensitive assets like BTC will all feel the pressure
Especially now that oil prices and Treasury yields are rising in tandem, meaning the market is worried about both weakening growth and a resurgence of inflation
If the 30-year yield continues to push higher, BTC could face a more intense short-term volatility
Going forward, don’t just focus on the coin price; the US Treasury yield spike might be the real variable determining the direction of risk assets $BTC retail investors are holding tonight's $BTC chart where it pulled back above 64,000 overnight and asking me: Is it going to reverse? Should I switch to long? I didn't move a muscle. Let me point out the real boss for you — tonight the 30-year US Treasury yield surged to 5.31%, a 19-year high. The hand pressing down on all overvalued assets is interest rates, not the minor fluctuations in the crypto price chart. The longer interest rates stay high, the harder it is for risk asset valuations to rise, and crypto is just the last link in this chain with the least pricing power. A short squeeze doesn't change this big picture. After trading for a while, you'll understand that the direction depends on the water level, not the waves. Don't be fooled by a short squeeze spike into catching the falling knife — chasing longs at this level has the worst odds. #财报观察员：小米即将发布财报，你更看好哪条业务线？ #财报观察员：Xiaomi is about to release its financial report, which business line do you favor more? After-hours, Xiaomi officially revealed its Q2 financial report. Let's first look at the consensus expectations from institutions: Estimated total revenue of ¥108.8 billion, a slight year-on-year decline of 6%; adjusted net profit of about ¥6 billion. The three major sectors—smartphones, car manufacturing, and AIoT—each have their strengths and weaknesses, with vastly different potentials. 1. Smartphone main business: sales shrink, but the high-end strategy is a great success In Q1, shipments were 33.8 million units, a sharp year-on-year drop of 19%, with sales clearly under pressure. But the highlights are highly valuable: the average selling price (ASP) of phones rose sharply by 8.2%, reaching a historic high of ¥1310. Actively cutting low-end models and fully focusing on the high-end market achieved "volume down, price up," steadily restoring profitability. If the storage chip price surge eases in Q3, the smartphone gross margin will further improve. 2. Xiaomi Auto: the biggest growth dark horse, getting closer to breakeven In Q2, SU7 deliveries reached 104,200 units, with a gross margin hitting 20.1%; operating loss narrowed significantly from ¥3.1 billion in Q1 to ¥2.06 billion. Scale effects continue to materialize, costs are continuously diluted, and the turning point to profitability is visible to the naked eye. With continued new car volume growth, automobiles will become a long-term core engine for Xiaomi's revenue growth. 3. AIoT smart home: first to recover, experiencing a strong rebound The 618 shopping festival strongly boosted sales, with Q2 IoT revenue surging 28% quarter-on-quarter to ¥31.6 billion. Demand for large appliances and smart home sectors has clearly warmed up, stabilizing Xiaomi's ecosystem foundation. Market outlook: Q3 highlights far exceed those of Q2 Who understands! It clearly looks like it's about to break out, but ends up just grinding in place.
The back-and-forth tug-of-war between bulls and bears has completely drained my mindset.
$ETH is really frustrating; after several attempts, it can't break through and just drops. The 1900-1920 range is heavily suppressed. Every rebound sees capital outflow. If you can't push through, just come down, okay? I'm waiting with my short position.
$BTC still opens high and closes low. The chart looks strong but there's no incremental capital support. It can't break the resistance level, nor can it break the support level. So it just stays put obediently.
Currently, neither of the two major coins has a clear direction. It's very frustrating. I'll hold this position and see if Ethereum can break downwards.
#BTC沉睡供应创新高，稀缺性再受关注 #高盛称美联储9月加息可能性非常低 Hold on until Thursday
I just hope it doesn't rise too sharply in these two days 📉
I did some research on the Federal Reserve myself
It feels like this meeting's minutes will most likely still lean hawkish
Although there was no rate hike last time
The vote changed from 12-0 to 9-3
Three people directly supported a rate hike
This division is already quite clear
Plus, inflation hasn't completely come down yet
As long as the minutes continue to emphasize inflation risks
The dollar and US Treasury yields might rise again
Definitely not comfortable for crypto
But the hawkish expectations might have been partially priced in already
Thursday might not necessarily see a dump right after the release
$ETH is still hovering around 1900
1950 remains the resistance I'm watching
If the minutes are hawkish
And 1880 doesn't hold, there's a chance to continue downward
So I just want to hold on until the news lands
I dare not add positions recklessly anymore
$BEAT dropped hard again today
The selling pressure after unlocking clearly hasn't been fully absorbed
Let's see if it can hold around 0.25 first
Only with volume can there be a potential oversold rebound
Without volume, I really don't dare to hope for a second wave
$SNDK is ridiculously strong instead
After last week's big surge, it pushed up near 1780 again
The two-month consolidation range has been broken upwards
Might test 2000 later
But it rose so fast
I'm afraid to chase now and end up at the peak
Hope the Federal Reserve doesn't suddenly turn dovish
Also hope ETH doesn't lift me out before Thursday
Let me quietly wait for the result
#财报观察员：小米即将发布财报，你更看好哪条业务线？
#30年期美债收益率创2007年以来新高 ComponentNews classifies server $MLCC as in a "severe" shortage phase, ETNews reports:
- According to DigiKey's shipment data, the delivery time for some high-capacity MLCCs from Samsung has reached about 40 weeks.
Earlier this year, the widely reported delivery time was about 20 weeks, so the AI server MLCC bottleneck continues to widen.
- Murata was about 24 weeks in June.
About 30 weeks in July. Now some have reached about 36 weeks.
This is interesting:
"New capacity expansion postponed from Q4 2026 to 2027"
It does not specify where the expansion is... maybe Murata? But if the capacity expansion is reportedly shelved, the bottleneck should tighten in the short term.
Delivery time is a good way to track demand imbalance. The 30-year U.S. Treasury yield has risen to 5.29%, hitting a multi-year high. Many media outlets and KOL experts have issued warnings, but many people actually don’t understand the relationship between Treasury bonds and the market. I’ll explain it simply in the most straightforward, no-nonsense plain language.
1. What is yield?
It’s the interest investors demand when lending money to the U.S. government. When yields rise, it usually means Treasuries are being sold and prices are falling.
2. What’s the difference between the long end and the short end?
The short end mainly reflects whether the Fed is raising or cutting rates in the near term; the long end reflects inflation, fiscal deficits, and debt risks over many years. So, short-end yields can fall while long-end yields don’t necessarily follow.
3. What does a rise in the long end indicate?
Investors are more cautious about the U.S. long-term outlook and demand higher returns. The market’s concern may no longer be the next Fed meeting but inflation, deficits, and interest burden over the coming years.
4. What impact does this have on the market?
Long-term U.S. Treasury yields are the benchmark for global asset pricing. When they rise, financing costs increase, stock valuations are pressured downward, the dollar becomes more attractive, and emerging markets, stocks, and cryptocurrencies face headwinds.
5. Where is the risk?
Debt increases → more bond issuance → yields rise → interest expenses increase → market worries more about debt → continued selling of Treasuries.
If this cycle continues, it could lead to further rises in long-term rates and put pressure on stocks, cryptocurrencies, and global markets.
Although this doesn’t mean the U.S. economy or the dollar will immediately have problems, market trading is fundamentally about expectations. The next major market rally may not be $BTC rising alone, nor $ETH rising alone, but rather capital shifting from "safe-haven allocation" to "on-chain finance".
The strongest phase in the crypto market is often not just BTC finishing its rise, nor ETH suddenly exploding alone, but a continuous transmission formed by the capital flow path. Step one: with macro environment improvement, capital first buys $BTC because it is the easiest for institutions to understand and best suited as an entry point to the crypto market. Step two: after BTC stabilizes, capital begins to seek higher elasticity and richer narratives, thus flowing to $ETH. Step three: if ETH strengthens, on-chain finance, DeFi, L2, RWA, and stablecoin ecosystems will be reactivated.
Currently, the market is still between steps one and two. BTC is waiting for macro confirmation around $64,000, and ETH is waiting for capital rotation around $1,900. The Federal Reserve, Jackson Hole, the Trump White House crypto meeting, stablecoin regulation, SEC rule progress, and ETF flows are all deciding whether this path can be realized.
If the Federal Reserve releases easing space, BTC is very likely to benefit first. Because BTC resembles a macro asset the most, liquidity improvement, a weaker dollar, and lower real interest rates will all make its digital gold narrative smoother. ETF funds will also more easily return to BTC because BTC is the clearest institutional entry. At this stage, the market is buying certainty and liquidity.
But if only BTC rises and ETH does not follow, the crypto rally is incomplete. A true on-chain bull market requires ETH participation. Because ETH represents the application layer, stablecoin settlement, DeFi yields, RWA, and smart contract ecosystem. BTC can bring money into crypto, but ETH determines whether this money enters on-chain financial activities. If ETH fails to stand up for a long time, it means capital is still stuck in safe-haven allocation rather than risk appetite expansion.
This is also why BTC and ETH should not be judged solely by who rises more. BTC strength indicates crypto is treated as asset allocation; ETH strength indicates crypto is treated as a financial system. If both are strong together, it means the market is not only willing to buy digital gold but also willing to believe in the on-chain economy. The former provides the foundation for the crypto market, the latter provides vitality.
The key now is whether ETF funds and macro data can cooperate. BTC needs continuous inflows to confirm institutional buying, ETH needs on-chain activity and yield logic to confirm revaluation. If BTC stabilizes around $64,000 and ETH can regain $1,900 and break upward, then the market will start discussing capital rotation. Otherwise, the rally is still just a BTC-led defensive rebound.
The next major rally may not start from altcoins, nor necessarily be ignited by Meme. It is more likely to start with traditional capital buying BTC first, then confirmed by ETH reactivating on-chain finance. BTC is responsible for opening the door, ETH is responsible for proving there is real economic activity behind the door.
Reference sources (do not copy into the main text): Barron’s, Investor’s Business Daily, Investopedia, CoinDesk, Investing.com ETH market page, Blockport ETH ETF monthly fund flows.——When the world's largest asset management companies settle on the same Chain, it ceases to be just a tool for the crypto world and becomes the underlying pipeline of the financial world. In 1968, the U.S. securities market was on the brink of collapse. Millions of paper stock certificates had to be manually transported, verified, and stamped every day, forcing the New York Stock Exchange to shorten trading hours to catch up with the backlog of paperwork. That year, Wall Street created DTCC—the Depository Trust & Clearing Corporation—to centralize the registration, delivery, and clearing of all securities into one system. No one thought this was glamorous. But over the next half-century, DTCC handled 99% of U.S. securities transactions worldwide, with a daily settlement volume exceeding $2.3 trillion. It is the most unassuming yet indispensable infrastructure in the entire financial world. In August 2026, a similar story is unfolding on Ethereum—except this time, what is being cleared is not paper certificates but digitized real-world assets. $44.7 Billion in Options According to a report by ChainDD on August 17, the tokenized real-world asset (RWA) market has expanded from a niche experiment to $44.7 billion in three years. This number itself is not shocking. What is remarkable is the foundation beneath it: Ethereum remains the largest RWA settlement layer. Although public chains like BNB Chain, Solana, and XRP Ledger each have their own layouts, Ethereum firmly holds the institutional preferred position with over half of the market share. Three years ago, RWA was just a proof of concept "moving government bonds onto the Chain." Today, it already covers U.S. government bonds, money market funds, private credi💾 Why SanDisk can keep its balance sheet so clean $SNDK $GPS $SPCX SanDisk's most easily underestimated competitive advantage right now might be its manufacturing system operated jointly with Kioxia for over twenty years. This system allows SanDisk to simultaneously obtain NAND original factory technology, production capacity, and cost control capabilities, while dispersing the heaviest factory buildings, clean rooms, equipment financing, and depreciation assets of many wafer fabs into Kioxia and Flash Ventures. The result is a very unique financial picture. SanDisk is clearly one of the world's largest NAND manufacturers, with fiscal year 2026 revenue reaching $20.248 billion. As of July 3, 2026, the net book value of factory equipment on its consolidated balance sheet is only $674 million. During the same period, cash is $4.762 billion, long-term debt has been reduced to zero, and Flash Ventures-related notes receivable and equity investments total only $678 million. The company's total assets are about $22.5 billion, with the PP&E recorded on SanDisk's own books not even reaching $700 million. This figure is quite extraordinary in the storage industry. Micron is a typical IDM, owning a large amount of factory buildings, clean rooms, lithography, etching, and deposition equipment. Micron's capital expenditure for fiscal year 2025 reached $13.8 billion, and the latest forecast for fiscal year 2026 PP&E capital expenditure will exceed 2.5$BTC has regained the $64,200 mark, while $ETH is around $1,901, and $SOL is still hovering around $75. Currently, it seems that funds continue to favor highly liquid core assets, rather than the entire crypto market moving into risk-on simultaneously. What truly deserves attention is the macro side: the yield on the US 30-year Treasury note rose to about 5.33%, the highest level since 2007; The 10-year yield is also close to 4.74%. Even though market expectations for another rate hike in September have dropped to about 37%, long-term yields continue to put pressure on risk assets. Meanwhile, oil prices climbed back above $90 per barrel, with Middle East tensions and inflation concerns further pushing up term premiums. In other words, the core question facing the market is no longer just "will the Fed raise interest rates," but how high long-term financing costs can go. Therefore, whether $BTC can hold the $64K is only the first step. To confirm a genuine return of market risk appetite, we need to see $ETH, $SOL, and broader altcoins break through with simultaneous volume increases. #XiaomiEarningsWatch #30YYieldHits2007High #SanDiskLongTermDeals #BTC #ETH #SOL #Crypto#财报观察员：小米即将发布财报，你更看好哪条业务线？
My answer is clear: Automotive is the future, AIoT is the foundation, and smartphones are the fading halo.
Xiaomi's Q2 single-quarter revenue exceeded 108.9 billion yuan again, but adjusted net profit dropped sharply by 42.6% year-on-year to 6.2 billion yuan.
Smartphones: Q2 revenue was 42.1 billion yuan, shipments declined year-on-year but the global average selling price surged 25.9% year-on-year to 1351 yuan, showing initial success in premiumization. However, how long the "volume down, price up" trend can last is questionable—ASP rose but revenue still fell, indicating volume dropped faster than price increased, and smartphone gross margin is suppressed around 8%.
Automotive: This is the most noteworthy part of Q2. Revenue was 24.9 billion yuan, deliveries reached 104,199 units, and the new generation SU7 ranked first in pure electric sedan sales above 200,000 yuan in the first half of the year. Operating loss was 2.6 billion yuan but gross margin has reached 19.2%, showing scale effects are taking hold. Once the automotive business achieves single-quarter profitability, the market will reassess Xiaomi.
AIoT: Overseas contributions stand out, internet service revenue is 9 billion yuan with a gross margin as high as 76.8%, and 767 million monthly active users form Xiaomi's deepest moat.
My judgment is that automotive is the variable that will determine Xiaomi's valuation ceiling for the next five years. Multiple institutions maintain a "buy" rating, believing a fundamental turning point will come in the second half of the year. When automotive turns profitable, the market will apply a completely different valuation logic to Xiaomi—from "burning money to tell stories" to "making money and accounting."
I watch all three lines, but the one most worth betting on is automotive. Recently, the Strait of Hormuz has once again fallen into a state of effective closure. This US-Iran conflict, which erupted at the end of February 2026, has lasted for over 170 days. Although there were brief memorandums of understanding and limited navigation attempts in between, Iran has clearly stated that the strait will not truly return to normal commercial navigation until the US meets a series of conditions including lifting the maritime blockade, removing sanctions, and unfreezing assets. About one-fifth of global oil trade originally depended on this chokepoint waterway, but now vessel traffic has dropped to single-digit percentages of pre-war levels. War risk insurance premiums have soared to 30 times the usual rate, and Brent crude oil prices have climbed back above $90. The sharp rise in energy costs has directly pushed up global inflation expectations, while the US Treasury's massive debt issuance has compounded this, causing US Treasury yields to rise across the board. The 30-year Treasury yield once touched its highest level since 2007, and the 10-year yield is also approaching multi-year peaks. The bond market sell-off is transmitting to the stock market, putting risk assets under repricing pressure.
Against this macro backdrop, the probability of a sharp plunge in US stocks after tonight's opening has significantly increased. Historical experience shows that when energy shocks and rising interest rates occur simultaneously, growth stocks and high-valuation tech stocks often bear the brunt first. The semiconductor sector, as the core beneficiary of this AI rally, has already accumulated huge gains, and its valuation elasticity has correspondingly amplified downside risks. Especially those memory chip manufacturers highly dependent on global supply chains and terminal demand prosperity are more vulnerable to capital withdrawal when risk appetite sharply declines. Once the market enters a risk-off mode, funds tend to prioritize selling liquid and previously high-gain targets, creating a stampede effect.
Based on the above logical chain, I recommend focusing on shorting SK Hynix opportunities. As a global leader in HBM high-bandwidth memory, Hynix's stock price has experienced multiple-fold increases amid the AI server demand boom, with its market value once surpassing Samsung to become Korea's largest. However, the current high oil prices may push up data center operating costs, while the high-interest-rate environment will suppress corporate capital expenditure willingness, potentially slowing AI infrastructure expansion. Coupled with weakening overall US stock sentiment, Hynix's US ADR and related derivatives are likely to become concentrated targets for shorts. Whether through direct shorting, using inverse ETFs, or leveraging futures and options tools, a relatively favorable window seems to have emerged timing-wise. Of course, short-term volatility is intense, so strict position management and stop-loss discipline must be observed.
It is especially important to emphasize that geopolitical situations can dramatically turn at any time. Once the US and Iran reach a substantive agreement again and truly restore strait navigation, oil prices may fall and risk appetite recover, quickly reversing the current logic. Therefore, shorting operations are more suitable as tactical trades rather than long-term strategic holdings. Meanwhile, the semiconductor industry itself still has strong long-term fundamental support, and AI demand will not disappear overnight. Investment decisions must be combined with one's own risk tolerance; blind following should be avoided. The market is always full of uncertainty, and tonight's plunge expectation may also be interrupted by unexpected positive news. Staying calm and thinking independently is key to navigating cycles.
Finally, a reminder: the above analysis only represents my personal observation and deduction of the current macro and market environment and does not constitute any investment advice. Financial markets carry very high risks, past performance does not represent future results, please make decisions cautiously based on your own situation and consult professional advisors if necessary. Brothers, the market changes in an instant; may we all protect our principal amid volatility and seize the opportunities that truly belong to us.
$SNDK
$SKHYNIX
#30年期美债收益率创2007年以来新高 After the sharp rise of OKB, a 4.4% correction occurred within 24 hours, breaking the $100 psychological support level. Has the market moved beyond chasing news to a phase of re-evaluating actual capital flows and structural value? OKB fell from $107 the previous day to the current $99.55, breaking the $100 level. Considering that the asset surged from $47 to $142 over the past 7 days, this correction strongly reflects profit-taking after the positive news of the permanent suspension of coin issuance and the fixed total supply of 21 million tokens was already fully priced in. In fact, the timing of the news announcement almost coincided with the peak price formation, which can be interpreted as a 'sell on fact' pattern. However, technical signals do not yet warn of a trend reversal. The RSI dropped from 89 to 76 but remains in the overbought zone, which can be seen as a process of digesting overheated short-term buying pressure. The 30-day gain of +22% and 7-day gain of +5% indicate that the medium-term uptrend is still intact In the past, people bought stocks; now they buy Crypto stocks.
Here's a recent obvious change: Previously, if you wanted exposure to $BTC, you basically had to go to an exchange or manage your own wallet; now more and more people are buying Strategy directly through their US stock accounts. Strategy currently holds over 840,000 BTC, which accounts for more than 4% of the circulating supply. This scale basically turns BTC into a stock—buying one share of MSTR is like buying a leveraged BTC exposure, with custody, compliance, and liquidity all ready-made.
$ETH is following the same path. BitMine has started putting ETH into its treasury—not just hoarding it, but adding a layer of corporate operational leverage to ETH. Buying its stock means you gain from both ETH price fluctuations and company operations; the volatility is greater than spot, but the entry method is just like buying ordinary stocks.
Another data point: In the past 5 days, among the 25 most liquid US stocks, at least 5 are directly related to Crypto—Strategy, BitMine, Robinhood, Coinbase, Circle, and Telegram all made the list. Traditional market money is already voting with its feet.
So my judgment is that the next wave of Crypto capital inflow may not be through exchanges but rather through the stock market. Money doesn’t need to know on-chain operations to get BTC and ETH exposure; this path is very smooth for institutions and large funds.
This is just my personal market observation and does not constitute investment advice. DYOR.The market is currently very fragmented: Bitcoin is surging, while most major altcoins are either stagnant or weakening, and a broad rally has not yet arrived. Meanwhile, a major shift is happening at the industry’s foundation, with many Bitcoin mining companies reallocating hash power and electricity resources to AI. 📊 Market status: typical structural divergence ✅ $BTC stands above 64,000, up over 1% intraday. But awkwardly: $ETH has fallen below 1900, $XRP dropped below $1, $BNB and $DOGE are basically flat, and $SOL is oscillating around $76. Only $HYPE remains strong, with a 7-day gain of 7.5%, continuing to outperform the market**. Current situation: capital is concentrated flowing into BTC, altcoins cannot attract liquidity, representing a typical bloodsucking market, not a full reversal. Technical signals to watch: Bitcoin has failed to hold above the 50-day moving average for 4 consecutive days, and the price remains below the 200-week moving average. Analyst view: The huge $62,000–$65,000 range has not been broken; without an effective breakout, it remains a consolidation pattern, and medium-term bearish pressure has not completely disappeared. ⚡ Major industry transformation: miners cut 21% hash power, shifting to AI Publicly listed Bitcoin miners have cut mining hash power by 21% over the past three quarters. Mining profits are thinning, while AI hash power hosting yields higher returns, causing many mining farms and electricity resources to shift from Bitcoin mining to AI infrastructure. While miners are fleeing the mining track, AI crypto tokens are developing independent trends: AI platform Venice’s annualized revenue exceeds $100 million, $VVV single-day upIf the SafePal order leak topic is ultimately confirmed to be true, the harm is not only to privacy
but also strikes a blow to the core psychological contract of hardware wallets.
Users buy hardware wallets precisely because they fear exchanges, hot wallets, on-chain phishing, and private key risks. If purchase records, contact information, addresses, and other off-chain data are compromised, users will suddenly realize: assets are safe on-chain, but that doesn't mean people are safe offline.
This is very painful.
Wallet manufacturers often say they don't touch private keys or custody assets, which is of course important. But hardware products cannot avoid logistics, after-sales service, and order systems; any link left unattended or poorly managed for too long can become an entry point for attackers.
Crypto security cannot focus only on mnemonic phrases.
True security is that even others don't know you have a wallet.
#SafePal订单泄露，隐私保护待完善 BTC's breakout with high volume does look like the market is about to reopen, but if you only look at BTC rising and ignore counterfeit trading volume, it's easy to fall into the illusion of capital creation. A true comprehensive bull market is not just about BTC rising, but about funds spilling over from mainstream coins to altcoins, creating sector rotation and profit-making effects. If BTC rises but other cryptocurrencies are slow to follow, it indicates that market risk appetite remains weak and funds are simply clustering in safe-haven assets. 📊Looking at the 24-hour market, BTC trading volume has clearly expanded, indicating that some funds have indeed participated in the breakout, but the overall trading volume of the altcoin sector has not kept pace. Many low-priced coins appear to be rebounding but are actually only slightly following the gains, with trading volume still below average. Under this structure, ordinary investors are most likely to misjudge: they think a BTC breakout is the start of a bull market, so they rush to buy low-level altcoins, only to end up buying weak stocks with low volume. Why do we say there are traps here? Because capital can be used with a small amount of capital to drive BTC to break out, generate market heat, then gradually distribute strong targets at high levels, or continue to flow into a few certain assets. Once BTC breaks out and there is no counterfeit rotation to take over, the market can easily shift from a "strong breakout" to a "false breakout." Especially near key resistance levels, market sentiment is most intense and prone to intense shakeouts. ———— historical experience, healthy breakout rallies usually have three characteristics: First, when the main currency breaks out, volume increases significantly; Second, after a breakthrough, it pulls back without breaking the key position; Third, hot sectors continue to spreadThe yield on 30-year U.S. Treasury notes its highest since 2007!!
This round of rally was driven by three core factors: First, Middle East tensions pushed up oil prices, the market repriced long-term inflation, and expectations for rate cuts continued to be delayed. Second, the U.S. fiscal deficit remains high, with massive Treasury bonds continuously injected, resulting in an oversupply of long-term bonds. Overseas central banks keep reducing their holdings of U.S. Treasuries, and the market needs higher yields to take over. Third, AI companies are issuing large-scale bonds to expand computing power, while corporate and government bonds compete for market funds, further pushing up long-term interest rates.
In the crypto market, risk-free returns exceed 5%, significantly increasing the opportunity cost of capital allocation to crypto assets, exerting medium- to long-term pressure on BTC and ETH. Institutional funds prefer to sit back and hold stable bond yields, with less willingness to enter highly volatile crypto assets. If yields continue to rise in the short term, it is highly likely that risk asset markets will continue to be suppressed; Only when yields turn to decline will the possibilities for risk assets reopen.
This article is only a market review and does not constitute any investment advice. #30年期美债收益率创2007年以来新高 #闪迪收涨逾8%, long-term agreements draw attention #财报观察员: Xiaomi is about to release its financial report. Which business line do you favor most? $BTC $ETH $SNDK ETH's Supercycle and the Current Short-Term Market As of the evening of August 18 Beijing time, ETH price is fluctuating around $1890–$1900, with a 24-hour main trading range roughly between $1885–$1916, and a market cap of about $228 billion. This position looks quite awkward compared to the past year; in 2025, ETH once hit a historical high of about $4946, and now it has retraced about 62% from that peak. The price performance is far behind the long-term expectations built around ETFs, corporate treasuries, stablecoins, and RWA at that time. Meanwhile, stablecoins on the Ethereum network remain close to $150 billion in scale, over one-third of ETH has already been staked, the US ETH ETFs still show net inflows, and corporate treasuries hold millions of ETH. 📉📈 So today’s discussion on ETH’s supercycle: the price has undergone a very deep compression, but the institutionalization of fundamentals continues to advance. My judgment is that ETH already meets the conditions for a long-term revaluation that spans beyond the traditional four-year crypto cycle, but this supercycle is still in the fundamental preparation stage and has not yet been confirmed on the price level. The short-term $1850–$2000 range, this $150 band, is the market pricing in this matter. The supercycle mentioned here refers to ETH’s demand sources gradually expanding from a single crypto market to ETFs.#财报观察员：小米即将发布财报，你更看好哪条业务线？
Tonight at 19:30, Xiaomi will release its Q2 and mid-year results. If I could only pick one, I choose automobiles: phones determine the performance floor, AIoT increases user stickiness, but automobiles determine the valuation ceiling.
On August 17, the SU7 series surpassed 500,000 cumulative deliveries in just 28.5 months. Demand has been validated; now the real question is: the more cars sold, can more money be retained?
Last quarter, Xiaomi delivered 80,856 vehicles, with smart cars, AI, and other new business revenues totaling ¥19.86 billion, a segment gross margin of 20.1%, but an operating loss of ¥3.1 billion. It resembles a newly exploded public blockchain: a strong user base doesn’t equal a proven business model; only when subsidies end and it can still make money is it truly the second curve.
So before the earnings report, I only allocate one-third of my planned position. If deliveries grow quarter-on-quarter this season, segment gross margin holds around 20%, and losses narrow significantly, I will add in two tranches after the report; if only sales look good but ASP and gross margin continue to fall, I won’t chase.
The most valuable time for the second curve is not when growth is fastest, but when it no longer needs blood transfusions from the first curve. Is the TRUMP family going to open a bank???
It's not a traditional commercial bank that can attract deposits and issue loans, but a trust bank license. So far, it has only received preliminary conditional approval and has not completed the final review to officially open.
The main entity is World Liberty Financial, a crypto company deeply invested in by the Trump family, applying to establish World Liberty Trust Company, a trust bank. This license cannot accept deposits or issue loans. Its core purpose is for its own stablecoin USD1: to operate independently from third-party custodians, directly responsible for the issuance and redemption of USD1, reserve asset management, and institutional digital asset custody.
This development has a clear impact on crypto market sentiment: once finalized, it means the stablecoin sector obtains formal federal-level financial qualifications, which is positive for the entire crypto industry sentiment and indirectly boosts risk appetite for BTC and ETH; in the short term, it will directly affect the $WLFI token first.
However, at this stage, it is only a phased permit and still requires fulfilling multiple regulatory conditions such as capital adequacy and internal control mechanisms. There are uncertainties in approval, so it should not be treated as an immediate positive for trading.
This article is only a market review and does not constitute any investment advice. #30年期美债收益率创2007年以来新高 #OKX预言家第二季正式上线 $BTC $ETH $SNDK #Strategy sold $334 million worth of stock last week, increasing its dollar reserves
The leader has something to say
Strategy sold $334 million worth of stock last week, did not buy Bitcoin, and increased its dollar reserves to about $4.8 billion. Saylor said stock buybacks are not a priority, but if MSTR shows a significant discount relative to net asset value, buybacks will be considered. He hopes STRC will return to around the $100 face value.
Previously, Strategy's rhythm was to raise funds to buy coins, and the market was used to it. Now it is starting to focus on cash reserves, preferred stock stability, and capital structure adjustments. The question is whether this is to save bullets for the next round of buying or if the company's continuous BTC buying role is weakening.
This has a neutral impact on Bitcoin. The $4.8 billion cash reserve can be converted into buying power at any time, but currently, there are indeed no new purchases. Strategy has shifted from a one-way long position to flexible allocation; the market can no longer treat it as a perpetual bull that only buys and never sells.
The SPCX base position continues its pattern, with floating profits between 110 and above 150 being substantial. Bitcoin continues to wait for direction, avoiding chasing after missing out.
The above analysis is time-sensitive; stop losses must be set on positions. Good luck. $BTC $ETH $SNDK Former OpenAI researcher Daniel Kokotajlo said a few days ago that AI is quickly learning skills that only top hackers used to have, and in the future, stealing from crypto wallets might become a fully automated assembly line.
I originally thought it was just another "AI is scary" prophecy. But after looking at this month's events together, I realized the tense was wrong — it's not in the future, it's already happening.
First, a term: agent framework. It's the open-source foundation that lets "self-working AI" run, anyone can download and use it.
This month, I saw the names of two frameworks twice — once for good news, once for bad news.
Good news: On August 6, MetaMask (the most widely used Ethereum wallet, that little fox) launched Agent Wallet, letting AI help you manage your money. The officially supported frameworks include Hermes and OpenClaw.
Bad news: On August 12, Israeli company Dream revealed that someone used Hermes and OpenClaw to break into 21 government systems in Taiwan over four days, cracked open 85 accounts, and stole more than 2,500 personnel records.
The same open-source tools. You use them to transfer USDC, others use them to hack government doors. And the AI writes the attacks itself, learning new tricks if blocked and continuing.
I think ordinary people can still do a few things: use multisig for large amounts (requiring several keys to move money), add a sufficiently long passphrase to wallets, and if you really want AI to manage money, first set strict single-transaction and daily limits. When the yield on the U.S. 30-year Treasury bond surged strongly to 5.31%, it marked the highest peak in the market since the eve of the 2007 financial crisis. The current turmoil in the bond market has long surpassed mere interest rate fluctuations; its essence is a vote of no confidence from the global capital markets in the U.S. government's fiscal integrity and long-term commitment capacity.
If all the focus is blamed solely on central bank monetary policy, it clearly deviates from the truth. Current employment and inflation indicators are gradually easing, and market expectations for short-term rate hikes are cooling down, yet long-term bond yields continue to soar against the trend. This fully demonstrates that the catalyst driving long-term rates is not short-term policy expectations but a deeper fiscal structural imbalance—the continuous expansion of the federal fiscal deficit, the flood-like issuance of government bonds into the market, and the significant deterioration in the willingness of major global buyers to absorb them.
The Treasury recently auctioned $25 billion of 30-year bonds, with the final winning yield rising to 5.216%, not only setting a new cost high since 2001 but also seeing the bid-to-cover ratio drop to a low level of 2.39 times. The weak auction results forced primary dealers to absorb the excess supply, while overseas central banks and large institutional investors continued to reduce their holdings. With the budget office revising deficit estimates upward, debt interest snowballing sharply, and tech giants issuing large amounts of bonds to fund AI data centers, chips, and power facilities, combined with energy price volatility, these factors together weave a perfect storm severely damaging the long-term bond market.
Under these circumstances, the yield curve showsThe market experienced significant volatility at the start of the first half of 2026, with Bitcoin's movement notably converging after entering summer. Spot trading volume and realized volatility have declined to the low levels of this cycle, and the market's reaction to various news shocks continues to weaken. On the options front, the 1-month implied volatility has dropped to 32%, significantly below the 42% annual average, reflecting market expectations for short-term price stability within a narrow range.
Looking back historically, Bitcoin's sustained low volatility state has phase characteristics. The last time volume and volatility bottomed simultaneously was in the summer of 2023, during which a bear market bottom structure was formed. At that time, the market also maintained a long period of range-bound oscillation, awaiting external catalysts; after sufficient clearing of existing selling pressure, Bitcoin broke out of the consolidation range and rose about 50% cumulatively from early October to the end of the year.
Current potential driving factors differ from those in 2023, but the market environment shares commonalities: low trading activity and continued compression of volatility, with participants awaiting the implementation of macro and regulatory policies. It should be noted that low volatility does not inherently constitute a bullish signal. However, the historical experience of 2023 is valuable as a reference; prolonged sideways phases often precede significant market turning points. The ongoing consolidation phase may indicate that the clearing of short positions is nearing completion. Compared to historical data, Bitcoin is currently in an oversold range.
Core inference: The market is currently in a phase of temporary consolidation. Historical data shows that long-term sideways markets often precede trend reversals, with the autumn 2023 market serving as an important reference.
Chart 1: Bitcoin spot trading volume and volatility have declined synchronously, with market structure highly similar to 2023
$BTC $ETH
#30年期美债收益率创2007年以来新高
#高盛称美联储9月加息可能性非常低
#BitMine增持至581.5万枚ETH，质押率约87% $BTC BITCOIN IS AT $64,155, BACK INSIDE THE 2021 ATH ZONE. IN 2022 THAT WAS A BEAR TRAP.
Old all time highs do not act as resistance forever. They become the box where the last sellers get flushed out.
2022: price fell into the 2018 ATH range, printed the trap at $15,500, and never traded there again.
2026: same box, same structure, four years later.
The pattern's projection off this structure is $150K.
$58K weekly close and I'm wrong.$ETH institutions quietly injected 2.85 billion last week, while BNB was neglected
Where the money is, love follows. Last week, the US spot ETH ETF saw a net inflow of 2.85 billion USD, setting a record — during the same period, BTC ETF only had 550 million, not even a fraction. Institutions' words can be deceptive, but their wallets are not; they are moving from BNB to ETH.
1. The capital structure is improving
The ETH/BTC exchange rate continues to strengthen, indicating this is not just retail hype but big money rotating positions. The price has climbed back above 1900; although the weekly chart is still red, the sustainability of inflows is far more important than daily price swings.
2. Fundamentals are also improving
The Ethereum Foundation just released the early testnet for Glamsterdam. If account abstraction is ultimately implemented, the barrier for L1 to run applications directly will significantly drop — this adds weight to long-term value, not just empty promises.
If 1900 holds, the next target is the 2000 USD level. The only risk is the Fed minutes on Wednesday; if they turn hawkish, the entire market will get hit. For a coin that institutions are aggressively buying, I always hold a high opinion.$OKB touched the $100 mark earlier this week before pulling back to the high-$90s, and the reaction has been split between "topping out" and "just catching its breath." The fundamentals underneath the token support the second read more than the first. The Supply Story Is Real Last year, OKX executed a one-time burn of roughly 65.3 million $OKB — worth tens of millions of dollars at the time — fixing total supply at a hard cap of 21 million tokens. That's not marketing language; it's on-chain fa#财报观察员：小米即将发布财报，你更看好哪条业务线？
I'm betting on automobiles. SU7 delivery volume is ramping up faster than expected, and the blockbuster effect on brand momentum is something phones can't provide. As long as the gross margin isn't too ugly, automobiles will be Xiaomi's core narrative for the next three years. Phones are the foundation, but the second growth curve is most likely in cars. Can gold continue to be bullish??? (August 18)
The medium to long-term bullish logic still holds, but short-term blind chasing is not advisable.
The core support for the bulls: recent weakening of U.S. economic data, the market lowering Fed rate hike expectations, and pressure on the dollar; continuous gold purchases by global central banks, and long-term risk support from Middle East geopolitical conflicts. At the same time, gold and BTC have emotional linkage, occasionally strengthening together during phases of rising risk appetite.
Short-term risks are also prominent: after a rapid rally, gold has accumulated a large amount of profit-taking positions. If U.S. inflation and employment data again exceed expectations and the Fed takes a hawkish stance, gold prices can easily pull back quickly.
From the crypto market perspective, gold's strength only drives overall market risk appetite, indirectly providing emotional support to BTC and ETH, but it will not directly change the trend of mainstream coins themselves.
Being bullish can be considered as a medium to long-term allocation strategy; short-term chasing carries high risk and requires waiting for a pullback or new catalyst signals before making a judgment.
This article is only a market review and does not constitute any investment advice. #30年期美债收益率创2007年以来新高 #黄金站上4430美元，期权资金转向看涨 $BTC $ETH $XAU At the moment the market plunged, I watched OKB slide from 107 to 99, and actually felt relieved. Have you ever felt that price has finally fallen sharply, but the price has finally been honest? Last week, the price jumped from 47 to 142, and this parabolic rally itself carries a sense of "overdrawing the future." When the news of the permanent shutdown of minting and the lockdown of 21 million total supply officially landed, the market actually began to realize profits—this wasn't negative news, but a typical scenario of all positive news being exhausted. Today it fell 4.4%, with the RSI falling from 89 to 76. Overbought pressure is being digested, and the trend itself hasn't been damaged yet. But what I want to say is not that number, but the real shift in capital preferences. Let's start with the hard logic. ICE (NYSE's parent company) joined OKX's board of directors, X Layer's RWA market share exceeds 80%, and xStocks' trading volume surpasses Solana's—these are the foundations for OKB's stability, not something that can be shaken by short-term candlestick charts. 30-day +22% and 7-day +5% gains indicate that mid-term funds are still in the market, but short-term profit-taking positions need to be turnover. I have drawn the key price levels very clearly: - Below 98 is the 24-hour low, 95 is the lifeline, 90 is the golden pit - Above are 104, 107, 110; only strength rebounds if it breaks above 110. If it doesn't break 95, a pullback is an opportunity to add positions; Falling below 90 is the real warning sign. Break through 110, target 115 to 124. Where are the risks? The market is always trading ahead of expectations. If ICE cooperation lacks substance,#30-year US Treasury yield hits highest since 2007
The 30-year US Treasury yield has directly reached a level unseen since 2007, and this signal cannot be ignored. Risk-free returns firmly stand above 5%, so what does this mean?
For something like $BTC, which has zero coupon and is purely driven by expectations, it naturally gets discounted — holding for a year is basically a bet that someone will buy at a higher price, as cash flow is zero. Over this year, Bitcoin’s maximum drawdown was 46%, while gold, which also tells an inflation-hedging story, actually rose 33%. In turbulent times, tangible hard currency is king, and in this regard, Bitcoin still falls short for now.
Capital flows are also very honest: crypto ETFs saw a net outflow of $90.55 million in a single week, clearly money is moving.
$SETH is even worse off. Staking yields look like 4% annually, which seems okay, but with 5.3% US Treasuries available, the opportunity cost is obviously fake. Many hedge funds are already arbitraging: lending out ETH to stake for 4%, then buying US Treasuries to lock in 5.3%, grabbing risk-free interest rate spreads. Money is always the smartest; it goes where certainty is higher.
$SNDK jumped from 1528 to 1786 in five trading days, nearly 17%, with a long-term supply, 80% gross margin, and AI storage boom — the story sounds great. But storage is cyclical; supply and demand depend on expansion rhythms of Samsung, Micron, and Hynix. SanDisk is a lone seedling and can’t turn the tide alone.
Summary: As long as the 30-year US Treasury yield doesn’t turn back, don’t expect major systemic rallies in crypto and cyclical growth. Next, watch the Fed meeting minutes for dovish clues and geopolitical developments A giant holding nearly one billion worth of Bitcoin turns to bet on AI chips
The Wall Street quant giant, recently revealed to have nearly one billion dollars worth of Bitcoin ETFs on its books, did something puzzling again this week.
Jane Street led a $700 million investment into an AI chip startup called Etched, pushing its valuation directly to $21 billion. Even more surreal, this is not just a financial investment—they also signed on as Etched's first customer, and the chips have already started shipping.
Just a few weeks ago, Jane Street disclosed in SEC filings that it holds about $990 million in Bitcoin spot ETFs, equivalent to over 15,000 coins, with more than $800 million invested in BlackRock's IBIT. This veteran institution, which makes its living from market making and arbitrage, holds real Bitcoin on its books and has now become the first buyer of AI chips.
At the end of the day, Jane Street has never been purely a crypto player nor purely an AI player. It is one of the world's largest market makers; money flows wherever there is liquidity and price difference. The Bitcoin ETF gives it a compliant crypto exposure, and the dedicated chips might help reduce its own trading computational costs. Both investments make sense when viewed separately.
We tend to think of crypto and AI as two separate, unrelated circles—one focused on on-chain belief, the other on computational arms races. But money flows without regard to narrative boundaries. The timing is even more intriguing: this month, Jane Street just reported its first-ever monthly loss, losing about $15 billion in July alone due to AI hedge funds and Asian equities, marking its first monthly loss in a decade. While money was bleeding, it placed its bets on a specialized chip not yet fully validated by the market.
Interestingly, specialized chips like Etched’s, which hardcode large model inference into hardware, have their biggest potential buyers among the most compute-intensive users. High-frequency trading and quant firms are already major GPU consumers and understand better than anyone how expensive inference costs are. Jane Street is both a major buyer of Bitcoin ETFs and the first customer of these chips, making this connection smoother than it appears on the surface.
However, when the lead investor also becomes the first buyer, the signal is somewhat subtle. Is it a genuine belief in the chip’s capabilities, or just a way to support insiders and prop up valuation? Outsiders find it hard to tell at a glance. Etched is betting on inference costs continuing to fall; the $21 billion valuation is impressive, but many predecessors in the specialized chip space have failed. Holding Bitcoin ETFs in one hand and AI chip orders in the other, is this cross-sector money an early strategic move or just a way to create a new story for themselves? How long is the rotation cycle of altcoin sectors???
In a stagnant market, the short-term hot sector speculation cycle mostly lasts 3-7 days, while in a bull market with incremental funds, it can last 2-4 weeks.
Currently, market liquidity is relatively weak, and the rotation speed has clearly accelerated. For example, recently: first, the BEAT meme coin rally lasted 4 days before quickly cooling off; then funds briefly switched to storage mapping SNDK; in the last 2 days, speculative capital suddenly surged $GPS pulse, with the hot sectors switching at increasingly shorter intervals.
Long-term narrative sectors have longer cycles. Sectors with industrial stories like AI storage and AI Agent can extend rotation cycles to 2-4 weeks, but short-term meme coins will still intermittently divert funds.
In a bull market phase with ample incremental funds, a mainstream narrative sector can often last more than a month, with multiple targets within the sector rising in turn; once the market turns into a stagnant game, speculative capital quickly enters and exits, and sector rallies are often just discovered by retail investors when the heat is already near its end.
The core factor determining the speed of rotation is total market liquidity. When BTC and ETH stabilize and off-exchange funds enter, sector sustainability will lengthen; when the market fluctuates and funds are insufficient, one-day or three-day hot sectors become the norm.
This article is only a market review and does not constitute any investment advice. #财报观察员：小米即将发布财报，你更看好哪条业务线？ #30年期美债收益率创2007年以来新高 #闪迪收涨逾8%，长期协议受关注 $BTC $ETH $SNDK Uniswap founder says AMM will take over financial markets
The founder of Uniswap is painting big pictures again, but this time the vision seems somewhat real. He said that AMM, the automated market maker mechanism, could become the core engine of financial markets in the future. Even stock ETFs on-chain would no longer be quoted in USD but traded directly between assets. This sounds bold, but on closer thought, it makes some sense.
Here’s the contrast. We’re used to everything being priced in USD, buying ETH by looking at its price against USDT. But once stock ETFs move on-chain, the market might shift from all assets priced against USD to direct swaps between related assets. This lowers holding risks and market-making costs, giving AMM a chance to penetrate the heart of traditional finance. It’s not crypto catering to Wall Street; it’s Wall Street’s trading methods being reconstructed on-chain.
For practical implications, in the short term, this means the DeFi market will expand. Stocks going on-chain means on-chain liquidity must absorb traditional assets, increasing opportunities for market makers and liquidity providers. The logic of LP business will also upgrade. This narrative will repeatedly fuel topics for DeFi tokens in market cycles, but don’t treat it as an immediate benefit—slow adoption has always been an old problem on-chain.
Some might say the on-chain depth is just a puddle compared to the NYSE. That’s true, but historically, every financial infrastructure replacement started from a small edge pool. Electronic trading was also mocked by floor traders for a long time.
Of course, AMM isn’t omnipotent. Its biggest weakness is slippage when liquidity is insufficient; if the pool is small, a large order can distort prices. Traditional order book matching still has an advantage in depth. So, in the short term, coexistence is more likely than immediate replacement. But the trend is already emerging: stablecoin settlements, RWA on-chain, stock tokenization—once these pieces come together, the variety of on-chain trading will exceed what we can imagine now.
In the long run, I believe in this direction. Asset tokenization is a slowing but irreversible trend. Wallets will no longer just hold coins but an entire financial market. Uniswap’s big vision might not be so far-fetched after all. What do you think—ten years from now, will we buy stocks in brokerage apps or complete exchanges with one click in wallets? This transformation won’t happen overnight, but the direction is already coded in.The real meaning behind $SNDK's surge: The AI market has expanded from GPUs to "how data is stored"
The hottest trend in the US stock market these days isn't ordinary AI software or some small-cap concept stock, but storage stocks being re-embraced by the market. $SNDK surged nearly 9% in a single day, with a cumulative increase exceeding 30% over five trading days. This isn't simply due to NAND price hikes, but because the market suddenly realizes: the bottleneck in AI infrastructure may not only be GPUs but also storage.
In the past, when people looked at AI, the first thing they saw was $NVDA. Because model training requires GPUs, inference also requires GPUs, and the biggest chunk of data center budgets goes to GPUs. Later, the market started focusing on HBM, because no matter how powerful the GPU is, without high-bandwidth memory, performance can't improve. Now NAND and enterprise SSDs are being reconsidered. The reason is very practical: AI models don't just finish after training; after large-scale commercialization, they have to handle inference requests daily, read context, query databases, save user data, cache, and perform retrieval. Data isn't just sitting idle; it is constantly read, written, migrated, and accessed.
This is where $SNDK's story is being rewritten. It used to be a typical storage cyclical stock, rising when phones, computers, and SSDs restocked, and falling when inventories were excessive. Investors were reluctant to give it a high valuation because the storage industry is too cyclical: when prices rise, manufacturers expand production; customers stock up in advance, then destock later. This cycle is too familiar, so the market naturally discounts it. But the AI inference era has brought NAND into new scenarios. Especially with keywords like KV cache, enterprise SSD, and data center flash emerging, $SNDK is no longer just a "consumer electronics storage supplier" but is being positioned as AI inference cost-reduction infrastructure.
This is crucial. The AI training phase competes on peak computing power, while the inference phase competes on cost efficiency. If a model serves hundreds of millions of users daily, the cost per inference becomes a matter of survival. Expensive memory can't be stacked infinitely; low-cost, high-density, scalable flash storage has the opportunity to take on a more important role. Storage is no longer just a backend supporting player but begins to participate in optimizing the AI system's cost structure.
However, this surge in $SNDK shouldn't be seen as blindly positive. The faster it rises, the more scrutinizing the market becomes. Investors will ask: Are long-term supply agreements really locking in profits? Can AI customer demand sustain through 2027 or even 2030? Will NAND price hikes stimulate capacity expansion? If the consumer side is squeezed by high prices, will it drag down overall shipments? These questions remain unresolved.
Therefore, the most important thing to note about $SNDK right now isn't how much it has risen today, but that its identity is shifting. It used to ride the inventory cycle; now it aims to ride the AI data cycle. The most important change for a company isn't necessarily the price moving first, but the market's interpretation of it changing first. $SNDK is now being reinterpreted as the data foundation for the AI inference era, which is the core reason for its recent popularity.
The first phase of AI buys GPUs, the second phase buys HBM, and the third phase starts buying storage. $SNDK's recent rise shows the market is no longer satisfied with just watching the leading AI front-runners but is searching for components in AI infrastructure that haven't been fully priced yet. OP 540 million token airdrop is at risk of being confiscated
If you still hold OP tokens, don’t rush to move them yet. An airdrop of 540 million tokens is facing the risk of confiscation, and its fate is being decided by a fierce governance civil war. Currently, about 9.1 million OP votes support confiscation, while about 4.25 million OP votes oppose it; neither side has enough to make a final decision.
Where’s the problem? Airdrops are originally gifts from the project to the community, but on Optimism, unclaimed tokens might be reclaimed by the protocol. Supporters say the funds should be used to build the ecosystem, while opponents see this as plundering early contributors. It’s the same token, but one side wants to confiscate it and the other wants to distribute it. The argument isn’t about money, but about who has the right to interpret the rules.
From the perspective of memes and altcoins, this is actually an old problem in crypto. Many projects’ governance tokens have voting thresholds so high that retail investors can’t participate, and only a few large addresses truly influence outcomes. OP is just exposing this contradiction, showing that so-called decentralized governance sometimes just shifts power from one company to a few wallets. You think you’re participating in governance, but the decision-making power has long been divided.
Ultimately, this drama damages the credibility of governance tokens as a whole. Investors are already skeptical, and now that even airdrops can be taken back at will, who would dare hold long-term? If a project can’t even uphold the most basic contract spirit, no matter how high the token price jumps in the short term, it’s just a castle in the air.
Some have done the math: if these 540 million tokens are truly confiscated and redistributed, the short-term selling pressure on the market won’t be huge, but the symbolic meaning is very bad. It tells everyone that the governance rights in your hands can be rewritten by the rules at any time. This uncertainty hurts morale more than a 10% price drop. Once community trust cracks, it’s much harder to repair than to pump the price.
In the short term, this internal conflict is a real drag on the token price; capital fears uncertain rules the most. For swing traders, I suggest avoiding this unresolved governance token until the outcome is clear, to avoid being shaken out by the tug-of-war. In the long run, whether the governance mechanism can be fixed will determine if OP is still worth trusting. Are you on the confiscation side or the distribution side? Share your reasons. I think this is much more interesting than guessing the token price. The fear that rules can be rewritten keeps people awake more than price fluctuations.Bitmine increased its ETH holdings by $18.9 million in one week
Tom Lee, this person, never talks about tops or bottoms, but Bitmine in his hands has never stopped buying ETH. The Data Nerd monitored that in the past week Bitmine bought another $18.9 million worth of ETH, plus $30.8 million of BMNR. Calculated, its ETH holdings already account for about 4.81% of the entire network, a proportion that would be intimidating for any institution.
What’s more intriguing is the contrast. Retail investors in the market are still debating when Bitcoin will return to 90,000 and whether ETH will continue to bottom out, while institutions are already investing real money steadily. BMNR is Bitmine’s own stock code; it buys ETH and itself simultaneously, effectively tying the company and ETH more tightly together. This strategy is not about short-term fluctuations but betting on ETH’s long-term position.
Behind Bitmine’s approach is a premise: it believes ETH staking yields can cover the cost of capital. Currently, ETH staking yields about 3% to 4% annually, which doesn’t sound like much, but compared to leaving cash idle in accounts, institutions prefer to earn interest on it. Coupled with narratives like RWA and stablecoin settlements, ETH’s story is richer than just hoarding coins. Another point is that BMNR is originally a proxy token for ETH; institutions buying it means indirectly increasing ETH exposure without directly holding the coin, which is especially friendly for compliant funds.
The most common mistake retail investors make is blindly following institutions when they buy. But you have to think clearly: Bitmine buys on a six-month to one-year cycle; it can afford to lose time, but you may not. Its holdings are publicly transparent on-chain addresses, effectively laying its cards on the table. When it really decides to exit, retail investors often realize it too late. What you learn from institutional moves is the direction, not to mindlessly copy.
For the market, 4.81% of the supply held by one entity means ETH’s chip distribution is more concentrated than it appears. In the short term, as long as these whales don’t dump, there is an invisible floor under ETH. But conversely, if they really reduce holdings one day, the selling pressure won’t be small. In terms of swings, I think ETH is actually a bit more resistant to drops compared to Bitcoin, but don’t treat it as a risk-free asset; institutional positions are a double-edged sword.
I agree with the long-term logic; ETH’s staking and RWA narratives are still growing. The current rhythm of continuous institutional buying is a psychological support for spot holders. But do you think one entity holding nearly 5% of the supply stabilizes the market floor or plants a bomb that could explode anytime? This dual-edged concentration deserves every coin holder to closely monitor the on-chain addresses.#BTC Low Volume Liquidity Trap: Behind the $64K Rebound
Bitcoin strongly reacted at the Monthly Open price, rising to the $64K level. The Monthly Open is a benchmark highly valued by institutions. By analyzing data from major exchanges, we can grasp the structural reasons behind this sharp rise under low volume conditions.
Funding Rates and Position Divergence
At the Monthly Open point, funding rates on major exchanges showed clear divergence.
- Binance, Bybit, OKX, Deribit: Funding rates entered negative territory around -0.00, indicating a clear short position advantage
- HTX: Maintained a positive funding rate of +0.05, with long positions dominant
Rebound Mechanism
Near the $62.7K Monthly Open support line, during price consolidation, funding rates on major exchanges centered around Binance turned negative. This overcrowded short position became the main catalyst for a short squeeze, driving the price upward.
Current Market Conditions and Risks
- Funding Rate Trend: Funding rates on Binance, OKX, Bybit, and Deribit are moving from neutral back toward negative. If this environment persists, further short-term upward pumps may occur.
- Volume and Volatility Risk: This rise lacks solid spot volume support. Insufficient market liquidity can amplify volatility, making price movements more extreme.
- Liquidity Illusion: Recent sharp two-way swings resemble short-term liquidity sweeps rather than a sustainable organic trend.
Key Resistance Level: The $65K level still acts as strong resistance. Caution is advised below this range in a low volume environment.
✏️ Summary
#BTC Monthly Open $62.7K reaction triggered a short squeeze, rebounding to $64K, but spot volume support is insufficient and funding rates have turned negative again. Therefore, attention should be paid to both the $65K resistance breakout and volume recovery. Shorting $16 million but betting Bitcoin won't crash
There's a trader named LucasMeow who recently did something particularly contradictory on Hyperliquid. On one hand, he used 40x full margin to short 100 BTC and 25x full margin to short 5000 ETH, with the two short positions totaling a nominal value of about $15.89 million, currently floating a profit of about $503,000; on the other hand, he went to the prediction market Polymarket and spent nearly $470,000 betting that these coins won't crash at all. In other words, he's betting on a crash with his left hand and betting against a crash with his right hand.
Looking specifically at his Polymarket positions, the logic is even more intriguing. He spent about $104,000 betting that BTC will not fall to any of six low points between $15,000 and $45,000 within the year; another $48,200 betting that ETH won't fall to $800; and the largest bet of $314,500 betting that Satoshi won't move the batch of coins that have been dormant this year. The total cost of these eight prediction positions is about $466,700, roughly 86.5% of the $539,000 margin on his two short positions.
This is quite ironic. A person who has staked his fortune on a crash turns around and spends almost the same amount of principal betting that the crash won't happen. What's even more thought-provoking is that the largest portion of his $460,000+ insurance, $314,500, is placed on Satoshi not moving the coins, indicating that what he truly fears is not the market but that ghost wallet that has been dormant for over a decade and is unpredictable.
What's more awkward is that this insurance is basically useless. Because when he bought it, these events were already high-probability No positions, and even if all eight predictions come true, the maximum profit would only be $50,800. Meanwhile, if his short positions reverse, with BTC rising to a liquidation price of $96,715 and ETH rising to a liquidation price of $2,542, the total loss would be about $5.993 million. That prediction profit wouldn't even cover a fraction of the liquidation loss, only 0.85%.
Cross-platform hedging is not new in the community; veterans often use prediction markets to insure on-chain positions. But in normal play, the insurance scale should match the risk, and this trade is clearly mismatched. Moving back and forth between Hyperliquid and Polymarket has become a standard move for many leveraged players, but LucasMeow's move here looks more like a gambler's consolation prize.
Interestingly, he hasn't stopped yet. On-chain data shows he placed two non-reducing sell orders at $62,277 and $61,500 for BTC, preparing to short an additional 125 BTC, with a nominal amount of about $7.715 million, still firmly bearish.
A person who only spends a fraction on hedging dares to keep adding to his short positions on the path to a crash. With 40x and 25x leverage in place, any decent rebound could wipe him out. Whether he truly sees the trend clearly or is already trapped, only he knows.Claiming to be a fortress giant, hiding 3 trillion in the footnotes
The Hyperion data center that Meta built in Louisiana covers an area roughly equivalent to 1,700 football fields. The lease term only starts counting from 2029, with an initial four-year term and the possibility to extend up to 20 years. Meta also additionally promises to compensate bondholders for any shortfall if it terminates the lease early. Such a huge commitment is not recorded as a single cent of liability on its balance sheet. The reason is simple: the company believes the probability of having to pay this compensation is low.
The Wall Street Journal reviewed the footnotes of the latest securities filings of nine companies. Alphabet, Meta, Microsoft, Amazon, Oracle, Nvidia, Broadcom, SpaceX, and AMD disclosed off-balance-sheet commitments totaling about 3 trillion dollars, the vast majority directly related to AI infrastructure. These nine companies spent a combined 600 billion in capital expenditures over the past year, meaning the amounts hidden in the footnotes are five times the actual cash spent and roughly three times the sum of outstanding leases plus long-term borrowings. Even more striking, this figure was 1.8 trillion just two months ago, rising by 50% in two months.
Breaking it down, one part is 1.2 trillion in leases not yet started, with Meta alone accounting for 347 billion; the other part is 1.9 trillion in purchase commitments, mainly used to pre-lock Nvidia chips and storage capacity. Under current accounting standards, these two categories do not need to be recorded on the balance sheet before delivery or lease payments begin. Legal, but indeed invisible.
There is an even more complex layer called residual value guarantees, totaling about 70 billion. Special Purpose Vehicles (SPVs) first borrow money to buy chips, with repayments relying on cash flows from AI companies' usage contracts; if the AI company stops paying, the chips are subleased or sold to cover debts; any shortfall is covered by the chip manufacturers. Who ultimately bears the risk? Nvidia, Broadcom, and other "shovel sellers." Yet these guarantees are usually not recorded on their books either. Broadcom has pushed this logic to the extreme: the Big Sky project guarantees a 35 billion debt, with Apollo and Blackstone funding the purchase of custom chips leased to Anthropic. This guarantee layer helped secure an investment-grade rating and lowered financing costs. Bank of America estimates Broadcom's AI platform could carry 370 billion in senior debt by mid-2029. Nvidia's Huang Renxun says they provide up to 25% residual value support, assessed case by case.
Rating agencies are already uneasy. Moody's says the main risk is the intense frequency of such transactions in the short term; even if on-paper leverage is not high, it will limit Broadcom's financial flexibility. S&P straightforwardly classifies residual value support as contingent debt obligations, to be included in adjusted debt. A DoubleLine portfolio manager put it bluntly: this is like exploiting a loophole in the system to get preferential treatment from rating agencies. We are entering an era of financial engineering, which is essentially masking financial realities. CreditSights' metaphor is even harsher: this is equivalent to selling a put option, costing almost nothing during boom times, but becoming deadly when severe downturns, client defaults, and hardware price drops occur.
Of course, some are unconcerned. Janus Henderson says triggering residual value support requires extreme circumstances, such as a cliff-like drop in token usage growth, which is not seen at all now. These companies are not trying to hide, but to finance.
I am more concerned about two other details. Alphabet and Amazon's free cash flow has turned negative, with capital expenditures exceeding operating cash inflows, meaning they will need to keep raising money from the market. Meanwhile, China's open-source models are providing comparable performance at much lower prices, and related token prices have dropped over 50% cumulatively in recent weeks. Revenue is uncertain, but the bills are basically irrevocable.
This pile of commitments is not yet a deadly repayment crisis, but more like a deferred depreciation wall and a bill that has not bottomed out. What do you think will ultimately happen to this 3 trillion hidden in the footnotes?A whale's crude oil short position has piled up to eight million dollars
Today, an interesting address appeared on-chain. The data comes from TradingBeats' on-chain monitoring, and every move of this type of wallet is closely watched. This account has been shorting storage stocks for a while; just one short position on SanDisk alone was nearly nine million dollars, yet it still shows a floating loss of over three hundred thousand dollars. Normally, losing money would mean stopping, but it did the opposite and shifted new chips toward crude oil.
Just past midnight today, this address opened more than 27,000 Brent crude oil short contracts on Hyperliquid at an average price just above $88, with a transaction amount of about $2.46 million. Even more aggressively, it placed over 60,000 sell orders near $91, with an order value of $5.67 million. If all these orders are filled, its short position in crude oil would reach $8.2 million.
I reviewed its position details and found this stance somewhat risky. The crude oil short is leveraged 20 times, currently with a floating loss of only $25,000, which doesn’t seem much, but the liquidation price is set at $98.81. Brent crude has risen 3.4% in the past two days, once touching $91. Any slight unrest in the Middle East could push oil prices higher, making this position quite fragile.
What really surprised me was its shorting rhythm. SanDisk, Seagate, and Western Digital all plunged pre-market; it was already short on storage and now is putting bullets into crude oil. An address betting simultaneously on storage and energy going down is rare on-chain. It seems to be betting on the same thing: a collapse in macro demand.
The problem is, the direction it’s betting on conflicts with its account status. The SanDisk short is already floating a loss of over $300,000, yet it stubbornly holds $1.57 million in sell orders at $1821 without reducing. On the crude oil side, it’s adding leverage and stacking up. If oil prices and storage stocks don’t fall but instead rise, losses under 20x leverage will accelerate rapidly. Everyone on-chain is watching this address, wondering if it holds some unknown insider information or is just stubbornly holding on.
In on-chain perpetual markets like Hyperliquid, watching whale addresses place bets has become routine for many. Some follow to copy, others do the opposite; every time this address adds to its position, it’s like publicly livestreaming its cards. But what’s public isn’t just its profits, but also its pressure.
There’s always a type of player in the market who refuses to admit losses and instead doubles down to try to turn things around. This whale’s current stance is either a foresight others don’t understand or the next specimen to be liquidated. What do you think? Does it really have an internal logic, or is it already hostage to its positions? Falling below 49,400 means losses even for holders who would rather die than sell
Yesterday, a piece of data gave me chills down my spine. CryptoQuant analyst Axel Adler Jr did some calculations and found that the cost basis for long-term Bitcoin holders is at $49,400. Currently, Bitcoin is hovering around $64,100, which seems far away, but once the price drops below this number, this group will overall shift from unrealized gains to unrealized losses.
What does long-term holder mean? It refers to those who bought coins and hold on tightly, not selling even when prices fall, known as diamond hands. They hold 16.35 million Bitcoins, just 58,000 less than the historical peak of 16.41 million on July 30. Over the past 78 days, Bitcoin has been oscillating between their cost basis and 1.5 times that amount, with 1.5 times corresponding to $74,100. In simple terms, these people are currently making about a 30% profit, far from a time to celebrate.
The most intriguing aspect is the selling frequency. Data shows that in the past 15 days, there were only two days when the supply from long-term holders decreased, meaning they have barely sold any. Normally, this unwillingness to sell is good news for the price, indicating the strongest holders are holding firm. But the analyst poured cold water on this, saying it only means the supply side is clean, not that a new wave of price increase has started. For prices to really go up, demand and price confirmation are both needed.
Looking from another angle, the current price is only 1.3 times the cost basis, which is actually low in the context of the whole cycle. Historically, every time Bitcoin falls below the long-term holders' cost basis, it’s rarely a soft landing; it’s more like a prelude to a stampede. When it really reaches that point, whether diamond hands continue to hold or are forced to let go is uncertain.
The problem lies in this delicate balance. On one side, diamond hands remain unmoved, and the market can’t find chips to dump; on the other side, they are not invincible, and 49,400 is the accounting dividing line. If it really falls, even the most loyal holders will start calculating their losses, and that kind of emotional contagion is never trivial.
Where are we now? Bitcoin is still hovering above 64,000, with a decent safety margin from that line. But in the past two weeks, macro factors like US Treasury yields hitting new highs and geopolitical tensions have shifted the winds quickly. Do you want to believe diamond hands can hold forever, or do you think this defense line will eventually be tested?