Orbit Post Sitemap
HYPE fell about 10% in a week, and the market has once again equated "large unstakes" with "selling immediately." Don't rush, there's still one step between these two movements. According to Hyperliquid's official documentation, transferring from a staking account back to a spot account requires a 7-day queue; Entering the queue means liquidity is being released, but it does not necessarily mean it will be sold after it arrives. Simply put, uncollateral data is more like a potential supply warning, not a sell order that has already been executed. To truly assess the pressure, we still need to look at on-chain transfers, exchange inflows, and spot trading volume after queue expiration. If only the price falls without subsequent funding path verification, attributing everything to the release would be too fast. $HYPE #HYPE遭大额解押减持, a 10% drop in one week is for mechanism and market observation only, and does not constitute investment advice.When was the last time you bought a SanDisk USB drive or SD card? Most likely, it was a long time ago—after all, phones no longer have SIM cards. But if you search the US stock market and search for the code SNDK, you might be shocked: since it was spun off from Western Digital (WDC) in February 2025, its stock price has risen over 5000% since then, with its market value once surpassing $220 billion. The same SanDisk, still the familiar face in USB flash drives, suddenly becomes fiercer than many AI concept stocks? How did the sellers of USB drives become the "AI storage leaders"? First, sort out your background. SanDisk was founded in 1988 and acquired by Western Digital for $19 billion in 2016, and has since been tied to Western Digital's mechanical hard drive (HDD) business—a slow, stable business that made steady profits, a flash memory business with a fierce, costly cycle and AI-related flash memory, all forced under one roof, with valuations suppressed for years. In 2022, activist Elliott entered the market to pressure the market, and with the storage industry crash in 2023 (NAND prices dropped 60-70% in a year), Western Digital couldn't hold on any longer. On February 24, 2025, it officially spun off its flash memory business, independently listing SNDK on Nasdaq, leaving the mechanical hard drive sector to defend itself. The split itself is just the beginning; the real catalyst is AI retelling the storage logic. For the past two years, the market has been focused on GPUs, and it won't be until 2025-2026 that people realize: large model training and inference are not realBTC spot ETFs ended a four-day streak of net outflows, with a net inflow of $32.1 million yesterday. After the FOMC kept rates unchanged, funds did not continue to flee; instead, they made small returns. Although the volume is small, the change in direction is worth watching. The previous consecutive outflows were mostly avoidance before the rate meeting. Yesterday, US stocks plunged and BTC was still relatively strong. Has crypto really bottomed out?
$BTC $AEON — Heavy selling has created extreme volatility.
Price is down 16.18% near 0.09618, with around $12.03M in volume. A confirmed support bounce is needed before considering recovery.
EP: 0.090–0.095
TP1: 0.103
TP2: 0.112
SL: 0.084#Fed3Dissents #MSFTCutsCapex #AIStoryDiverges Rebound ≠ reversal, $ETH surged 4%, $QQQ was dazzlingly green, and the market was waiting—whoever showed weakness first would set today's tone.
Look at the numbers
$BTC 65,283 +1.45% $ETH 1,952 +4.14%
$QQQ -1.12% $SPY +0.10% $IBIT -0.82%
$DXY -0.15% $GLD +0.10%
Hormuz and crude oil are still adding variables to inflation expectations, US Treasury yields and the shadow of Fed tightening continue to weigh on valuations, and the dollar isn't a backdrop—just a quick adjustment of the exchange rate line can disrupt the rhythm of $QQQ$SPY. It's not surprising which switch gets triggered in today's market $MSFT $MU $SNDK $MSFT $AMZN $META $GOOGL
$ETH Elasticity is clearly stronger than $BTC, short-term risk appetite is rising, but $QQQ is sinking downward, and money is shrinking into defense. $IBIT Weaker than spot $BTC; if ETFs weaken, it means the spot market isn't as strong; $DXY If it breathes a little easier, risk assets can catch their breath, but once tightened, it quickly turns hostile; $GLD Still quietly rising, safe-haven funds haven't fully withdrawn—don't be fooled by the hype.Last night was explosive outside, and it relates to us
1. The Fed didn't move, interest rates unchanged. The expectation of a rate hike in September has eased—USD is weak, which is good for risk assets.
2. US stocks fell, crude oil surged over 7%, gold at $4123. Money is flowing into hard assets. Will Bitcoin follow this wave? I'm watching.
3. Microsoft rose 8%, Meta fell 8%. Both are AI-related, but different expectation management led to vastly different results. In crypto terms—project teams need skill in painting the picture.
4. Grok 4.5 released, input 2 output $6 per million tokens. The AI track is still competitive, crypto AI projects face considerable pressure.
5. South Korea restricts retail leverage ETFs—global deleveraging is underway, brothers, watch your positions.
Conclusion: Macro is relatively loose, capital is seeking hard assets. Bitcoin's "digital gold" narrative may be repriced.
Discuss in the comments: Is Bitcoin still a copycat for you? I'm holding Bitcoin + AI observation positions, no leverage used. #Fed3Dissents #MSFTCutsCapex #AIStoryDiverges PCE print came out, short positions got wiped out. Microsoft earnings hit the wires, and market direction flipped entirely.
This trade caught me off guard.
My $SKHYNIX short was down -60.08%, taken out by a sharp spike.
I entered around the 900 zone during consolidation. After sleeping through the session, price ripped straight to 1023 and stopped me out cleanly.
It quickly sank in: the PCE release plus Microsoft commentary had completely reset the market’s core narrative.
🔍 What did the PCE data actually signal?
June PCE fell -0.1% month-on-month, the first negative reading since 2020.
Core PCE rose only 0.1% MoM, below the 0.2% consensus forecast.
Inflation cooled visibly.
Meanwhile Q2 GDP printed at 1.5%, well under the 2.1% expectation.
The data mix is nuanced: inflation is easing, economic growth is slowing, yet consumer spending remains resilient with Q2 consumption growth hitting 3.2%.
The market’s immediate takeaway: this is a risk-on catalyst.
That spike on $SKHYNIX unfolded right under this shifting sentiment.
💻 The bigger wildcard: Microsoft’s capex cut triggered a violent reaction.
The real market mover was Microsoft earnings.
The company revised its FY2027 capital expenditure guidance down from $190B to $175B.
Its stock surged more than 8% after hours.
Why the euphoria?
For months, investors feared AI capital burn would overwhelm big tech balance sheets.
This signal carries layered meaning:
The AI thesis remains intact, but the “spend at all costs for growth” logic is weakening.
Markets are starting to reward disciplined cost control and returns over unrestrained capex expansion.
For memory chips, this is a mixed signal.
Near-term bullish: AI capex persists, just at a slower pace; demand is not collapsing outright.
Long-term bearish risk: if other tech giants follow Microsoft’s lead and slash spending, memory demand expectations will face repric
🎯 Outlook on $SKHYNIX ahead
Near-term bias leans upward.
Cooler PCE, Microsoft’s positive surprise, plus oversold technical conditions support the bounce.
1023 may not mark the immediate top A sharp critique of Washi and its policies, a new era of imagination and criticism
Using tonight's data as a way to roast Wash, I happened to be discussing Wash's policies with @TiezhuCrypto Boss during the AMA at @binancezh Square, and the exchange of ideas brought new perspectives
A buff for start: this part of my content is mostly personal speculation and cannot be taken as a mainstream reference, so it might be a bit subjective:
1. Compared to Powell, Wash is more like a politician. From his remarks, Powell seems more like a scholar of integrity—one serving politics, the other serving the economy. This statement may carry a lot of subjective tone
2. Wash's downplay of the Fed's forward-looking ability and current June inflation data is a move to overturn the entire Fed framework. From expectation management to data sources, Wash's denial is a denial of the previous Fed system, indicating his ambition
3. No rate hikes in July, and no in September. Rate hikes are a high-pressure test for the US economy. Walsh doesn't dare to play this way, because a single rate hike isn't important; what matters is the return of the rate hike cycle, which will put severe pressure on the US capital market. This is a red line, not easy to cross.
4. Walsh emphasizes that the high interest rate environment and limited easing are meant to limit capital liquidity, guiding funds within controllable limits to liquidity dividend zones—U.S. stocks, or U.S. capital—to support AI's high valuation. This is a political task, and both Republicans and Democrats need such a Fed chair
5. Walsh is not cutting rates simply because there isn't enough time. He has just taken office and needs time to control the Fed and initiate reforms. If rates are cut now, the market's data will still be the same as before. But what he wants most is to convince the market that he will form a working group in the future
Imagine when the market no longer trusts the Fed's statements but only the data, and this data comes from Walsh's working group, who decides how interest rates are adjusted?
6,@TiezhuCrypto Boss Tiezhu reminds everyone that the members of the working group have some degree of connection with Wash, so clearly, this working group has Wash's attributes
When internal conflicts cannot be changed, external forces are chosen to break through. Walsh does exactly that, using external working group data to change the current Federal Reserve officials' stance and thereby guide macroeconomic control. This is Walsh's goal.
7. As for the Fed under Warsh's leadership, I can't judge the final outcome yet; we'll have to wait for the market and time. But clearly, Walsh's actions aren't for the economy—it's for himself, and for his political capital. Once he establishes his own system, the next president—whether Democratic or Republican—won't make much difference, and he will be in an undefeated position.
Regarding our impact, I believe that before the working group is formed and data is released, it will be difficult for both the market and the economy to change the strategy of not cutting rates. Unless inflation continues to decline (which is difficult), employment declines significantly, and there is a risk of economic recession, all of which may have to endure the torment brought by Walsh during this period
I believe the key turning point will come after Trump's midterm elections. When the midterms are settled and political factors stabilize, Wash completes data collection, establishes a new data portfolio, and with external forces, can start a new Washi era. Under limited bearish easing and limited liquidity, the market's survival environment will be relatively harsher!
Especially since Walsh can weaken the Fed's expectations management, keeping the market anchored in economic data, which increases market volatility caused by short- and medium-term economic data. Meanwhile, the speeches that need attention at subsequent policy meetings can be gradually ignored—I almost fell asleep listening to it yesterday! #美联储三票主张加息, tonight's PCE is a new highlight 🚨 The Prisoner's Dilemma of Global AI Stocks
The U.S., China, and South Korea's tech sectors are beginning to resemble a classic prisoner's dilemma.
Everyone is acting in their own short-term interest...
...and that may be making things worse for everyone.
🇺🇸 U.S. tech giants continue pouring massive capital into AI, worried about losing their competitive edge.
🇰🇷 South Korean investors remain heavily exposed to semiconductor names like Samsung and SK Hynix, with many also using leveraged products to amplify returns.
🇨🇳 Chinese markets face cautious sentiment, where investors often rush to reduce exposure before others do.
The result?
Instead of reinforcing confidence across the AI supply chain, uncertainty encourages defensive positioning and fuels broader selling pressure.
In theory, sustained AI investment could support the entire ecosystem.
In practice, fear, leverage, and positioning can create a negative feedback loop where everyone tries to protect themselves at the same time.
That's why, in the short term, market psychology can outweigh strong long-term fundamentals.
Patience and disciplined risk management matter just as much as believing in the AI story.
$SKHY $MU $NVDA
#Fed3Dissents #MSFTCutsCapex #AIStoryDiverges
$BTC $ETH $SNDK Tonight's PCE data didn't bring much to the crypto market. As I analyzed in my previous post, the profit-making opportunities are in US stocks. US semiconductor technology saw a midday rebound, and this rebound exceeded expectations. Why are US stocks rising so fiercely? Mainstream Coin $BTC $ETH Shows No Significant Volatility? The FOMC and PCE did not exceed expectations and are basically in line with institutional forecasts. Sentiment has already been priced in. Tonight's US stock market credit goes to $MSFT Microsoft. Microsoft's earnings report exceeded expectations, driving a comprehensive recovery in AI and chips. Microsoft's Q4 revenue, cloud Azure, and AI business growth all far exceeded Wall Street expectations. At the same time, the full-year AI capital expenditure was cut to prove that AI investment can truly generate profits. The market panic over "burning cash but not profits" was alleviated. The stock surged over 10% intraday, with capital flowing back into the semiconductor sector. Micron, SK Hynix ADR, Nvidia, and memory chip all rebounded violently. The Philadelphia Semiconductor Index surged 7%, directly driving the Nasdaq higher. Tonight's PCE data results combined with Q2 GDP growth of 1.5% were below expectations. Economic weakness + easing inflation The market bets on a rebound in September rate cuts. US Treasury yields fell across the board. Valuation pressure on high-valuation tech stocks eased, and funds aggressively bottom-fished growth stocks. However, note that this rebound is an oversold rally + positive earnings reports. After digesting, there will be some pullback. Expected on Friday? #微软逆势下调资本开支, up 8.5% in after-hours #财报观察员: Microsoft Cloud revenue surpasses 100 billion, Meta underperforms in AICurrently, the US is rising and falling as wildly as with altcoins, making this the most suitable market for 😅 retail investors to liquidate positions
Amid the frenzy of storing bull returns, $META is still plummeting
I'm curious how long upstream vendors can keep celebrating if downstream customers run out of money to invest in data centersA $1.6 billion order can't save the stock price; SpaceX is giving the market one of the most important lessons: strong fundamentals do not guarantee a forever reasonable valuation.
Recently, the SpaceX situation has been quite interesting.
The U.S. Space Force awarded SpaceX a huge $1.6 billion contract for 18 Falcon 9 launch missions, which in itself is definitely not bad news.
If you only look at the company's fundamentals, SpaceX remains one of the strongest players in the global commercial space sector.
Rocket launch business is stable, Starlink continues to expand, and there is huge potential in space internet, satellite communications, and more in the future.
But the market's reaction is very realistic:
Good news comes out, but the stock price does not rise; instead, it continues to adjust.
What this reflects is not whether SpaceX has value, but that investors are starting to reassess the valuation.
My view is:
SpaceX may be a great company, but now the market is re-answering a question—how much is a great company really worth?
In recent years, the capital market has been willing to give SpaceX very high expectations.
Investors are not just buying the current rocket business but are betting on the space economy of the next decade.
This logic is very similar to the previous AI market.
The market prices the future in advance.
When everyone believes an industry will explode, valuations often run ahead of performance.
But the problem is, as expectations rise, any imperfection becomes pressure on the stock price.
The pressures SpaceX faces now mainly come from several aspects.
First, valuation pressure.
The larger a company's future growth potential, the higher the price the market assigns, but this means future higher growth must be continuously delivered, or the valuation will be readjusted.
Second, capital and share pressure.
Currently, the market is focused on SpaceX's stock price pullback, short selling ratio, and subsequent restricted stock releases, all of which affect short-term trading sentiment.
Third, changes in the market environment.
Many high-valuation assets have experienced similar situations this year:
Company business hasn't deteriorated significantly, but stock prices start to return to reasonable levels.
Because the market has moved from the "storytelling phase" to the "profit verification phase."
So I won't simply think:
SpaceX dropping 50% is an opportunity.
Nor will I think:
A stock price decline means the company is failing.
What really matters is looking at the next few quarters:
Can Starlink's revenue growth continue?
Can commercial launch orders convert into profits?
When will the massive investments start generating cash flow?
If these metrics continue to be delivered, then the current adjustment may just be the market finding a reasonable price again.
But if future growth falls short of expectations, valuation pressure will continue to be released.
This situation actually serves as a reminder to all investors:
Don't ignore price just because you like a company's products.
Tesla is like this, AI companies are like this, and SpaceX is no different.
Good companies are always worth attention, but good investment opportunities often appear when the market starts doubting them.
The current SpaceX is more like a renewed contest between "value" and "price."
#SpaceX获$1.6B美军合同，股价暴跌引两派争议 $SPCX $NMR — Buyers are rebuilding strength.
Price is up 3.69% near 8.539, with around $108.49K in volume. A clean breakout could continue the rally.
EP: 8.25–8.45
TP1: 8.90
TP2: 9.35
SL: 7.90Technology sees the dawn
Yesterday, Federal Reserve Chair Powell's failed press conference shifted blame to the market for causing trouble, leading to a sharp drop in U.S. stocks and bonds. Technology stocks plunged at the close but then saw an important narrative-driven rescue after hours.
Microsoft's earnings report showed cloud business growth of 40%, EPS growth of 23%, and more importantly, profit growth covered interest and depreciation, with positive cash flow maintained. They also boldly lowered capital expenditures, gaining market approval. This finally provides a "monetization of AI" template in the AI narrative, no longer just upstream procurement expansion. Microsoft surged 15% tonight.
Tonight, U.S. June PCE dropped from 4.1% to 3.7%, month-over-month -0.1%, again confirming the previous CPI peak, consistent with my earlier June peak judgment. U.S. Q2 GDP grew 1.5% quarter-over-quarter, below the expected 2%, with consumption slowing from 0.9% growth to 0.3%. Inflation and the economy are both slowing; excluding oil price effects, the U.S. could fully cut rates this year.
Overseas markets warmed up, with the Nasdaq rising 2.5%, which is also very positive for domestic tech sectors tomorrow. Tonight's important meeting tone is also relatively positive. The Q2 economic tone dropped the word "stable," shifting from "counter-cyclical + cross-cycle adjustment" to just "counter-cyclical adjustment," indicating recognition of difficulties and that further measures will be taken.
Overall, this helps a comprehensive economic recovery. Next, consumer leaders, infrastructure, and industrial metals can be focused on as defensive sectors.
Increased rate cut expectations also drove gold sharply higher. Gold and tech rebounding simultaneously is rare; gold previously moved inversely to AI. The fact that gold remains strong despite expected declines is a strong signal and shows that the Fed losing credibility benefits gold, which is worth watching.
Bitcoin also rose above 64,000 tonight, with logic similar to gold; both have performed well recently.
Overall, the strong rally in overseas tech tonight is very encouraging. The key is that the fundamental narrative may turn around. Next, it depends on whether Amazon and Apple can continue the momentum tonight.
The above is only personal opinion, not investment advice. Please be aware of risks. I just closed my position, and now I want to fight back
First, post your trade: open long at 915.97, flat at 954.99, 15x segregated position, +56.31%. This order was made from panic profits, sold during a noisy period, no problem.
But now I want to be honest: at the 1018 position, I've entered the market again.
Why change your mind? Let's talk from the picture:
First, market signals provide confidence.
Looking at the candlestick chart, the previous two large bearish candles dropped from 1100 to 885, and panic buying was completely wiped out. But look at today's trend: the price has broken above 1,018, MA5 983 and MA10 950 are all under its feet, and the short-term moving averages are starting to diverge upward. This rebound is no coincidence; it's the result of capital buying real money. MA120 is holding below 1100, but there is only about 8% of the gap from now. Once it breaks through, it will be a new upward relay.
Second, SK Hynix's fundamentals are fundamentally sound.
When institutions say "performance falls short of expectations," I just laugh. Revenue and profit hit record highs, AI memory is in short supply—how is this a negative sign? This is clearly a financial report wash, washing out unstable chips and then moving lightly to continue rising. If general storage drops, so be it. HBM orders are scheduled until next year, and that's the core logic.
Third, tonight's PCE may actually be a catalyst.
The Fed's three votes advocating rate hikes are true, but the market has already reacted early. From 1100 to 885, all the negative news that should have been digested has been absorbed. As long as tonight's PCE isn't particularly outrageous, all the negative news is positive. If the data leans dovish, take off immediately; Even if it's hawkish, even if it drops to 980, it's still a handout—not risk, but opportunity.
So here's what I'm doing now:
I've already entered in batches around 1000, with positions lighter than the previous trade, but stop-loss at 970, take-profit at 1080-1100. The price-to-loss ratio is close to 3:1, so even if there's a pullback at this level, I wouldn't be worried.
The logic is simple:
· If it drops to 885, you won't buy; if it drops to 915, I buy
· If it rises to 1018, you wouldn't dare to buy, but I do
· Because the fundamentals haven't changed; what has changed is only market sentiment
When others panic, I am greedy; when others hesitate, I increase my position. For SK Hynix this round, I expect a new high.
$SKHYNIX
#美联储三票主张加息, tonight's PCE is a new highlight
#海力士业绩创纪录但不及预期, storage stocks experienced sharp fluctuations 🚨 Breaking news! Trump publicly stated that Bitcoin has irreplaceable core value
He bluntly stated that if the U.S. does not seize the blockchain track, other countries will soon take the lead.
Don't mistake politicians for hyping crypto out of technological optimism; in reality, global debt is rampant and traditional financial systems are riddled with loopholes.
PCE inflation data, endless money printing by countries continuously diluting wealth, and governments targeting decentralized assets like $BTC are essentially a reluctant choice when traditional finance reaches a bottleneck.
Market Perspective: Inflation data and political news are only short-term disruptions; the core remains whether funds can use this wave of policy expectations to break out of a one-sided rally. #美联储三票主张加息, tonight's PCE is a new highlight ---
**4 rose and 1 fell, averaging an increase of 8.9%, with total turnover of $140 million. **
Opening OKX's tokenized US stock page, the first impression was: it has risen quite well.
**But look closely—what's going on so far? **
---
XSNDK (3x short NASDAQ) :**+19.77%**
XSOXL (3x long semiconductor) :**+13.04%**
XSKHY (ARK Innovation ETF) :**+10%**
XSPCX (PC Industry) :**+1.77%**
**XSPY (S&P 500): -0.01%**
---
## What is the market doing?
**The market is not moving, and extreme stocks are going wild. **
This is not risk-on; this is **hedging + speculation. **
The S&P 500 is stagnant→ indicating someone is on the tray
But NASDAQ's short selling tool rose 20% → indicating the true sentiment of tech stocks is **panic**
**Betting on both sides, hedgeing against each other—this is institutions fighting, retail investors acting as cannon fodder. **
---
## Where is the capital flowing?
The transaction structure explains everything:
- **XSPY turnover $84 million**, with only 2% volatility
- **XSNDK turnover $30.75 million**, volatility 28%
- **XSOXL traded $8.75 million**, volatility 31%
**Liquidity in the main market is shrinking, and liquidity in extreme products is exploding. **
Smart money isn't buying the market; they're ** short tech stocks and long safe-haven assets. **
XSKHY rose 10% → funds are moving from the index to ARK innovation
This isn't bottom-fishing; it's ** funds searching for direction. **
---
## What role does BTC play in this game?
BTC $64,630，+0.78%
Tech stocks are in chaos, yet BTC has not fallen.
**This is no coincidence. **
When you see XSNDK up 20%, XSOX up 13%, and XSKHY up 10%.
BTC only rose 0.78%
**It is following the rise, but not following the trend. **
**Some people are using BTC as a safe haven—not buying BTC to hedge, but not investing BTC. **
---
There's a line on Toutiao:
> "Bitcoin seesaws as hit to risk sentiment and surging oil offsets Fed rate hold"
**Crude Oil Surges, Risk Sentiment Plunges, Fed Holds Stead**
Three negative factors, BTC only fluctuated.
**This indicates that someone is buying at the bottom of BTC. **
---
**But don't get too happy just yet. **
Fear index 28, so funding rates are likely negative
This is not a signal of a bull market restart
This is the ** market searching for a bottom, but not yet. **
---
## The Last Words
While Wall Street is crashing tech stocks, institutions are hedging, and retail investors are chasing gains and selling lows
BTC believers are still posting 🚀 in the group
**This is either the fooliest optimism or the smartest bottom-fishing. **
**The answer will be revealed next week. **
---
**Holding Discipline: XSNDK long 3x 5x position, entered at $1222, TP $1344, SL $1160. **
**Don't copy the trade, don't blame me if you lose. **Four asset categories are simultaneously halved by 54%, and it's no coincidence! Who are the real hard assets 🔥?
Data Highlights:
Bitcoin falls 54% in 268 days | Silver drops 54% in 169 days
SanDisk down 55% in 36 days | SK Hynix down 53% in 34 days
Cycles vary greatly, and the declines are almost uniform, with the underlying logic completely connected.
The cause of the crash is unified
The Federal Reserve's long-term high interest rates, ongoing global liquidity tightening, institutions collectively withdrawing cash, and all leveraged and expected assets with overdraft expectations selling off simultaneously.
1. Semiconductors (SanDisk SNDK, SKHY): Previous AI rally overloaded gains, but positive news materialized collectively to take profits, causing cyclical stocks to crash valuations;
2. Silver XAG: combines industrial + safe-haven attributes; high interest rates suppress interest-free precious metals, weakening the economy dragging down industrial demand, resulting in double pressure;
3. BTC: Global liquidity amplifier, leveraged trading triggered chain liquidations to amplify losses, highly linked to US tech stocks.
Core Soul Question: Who Are the Quality Assets?
Short-term declines are converging, but long-term values are worlds apart:
• Storage chips: A highly cyclical industry, where capacity expansion directly suppresses profits, and the market depends on the AI capital expenditure cycle;
• Silver: Industrial rigid demand but no scarcity barriers; prices are mutually constrained by the US dollar and the real economy;
• BTC: Total supply is permanently constant with no additional issuance, a globally accepted digital hard currency, with the underlying long-term anti-inflation logic unchanged.
Short-term liquidity is indistinguishable, but over a 3-5 year cycle, assets that are scarce and cannot be issued will have a clear gap in resilience and recovery.
Which type of asset do you think is more optimistic about long-term allocation? Let's talk in the comments!
#BTC #白银 #美股半导体 #宏观行情解读
⚠️ This is only an objective interpretation of asset logic, does not constitute any investment or trading advice, and the volatility risks of various asset types are extremely high! #韩股波动剧烈引监管介入, Finance Minister apologizes for leveraged ETFs #美光暴跌后: Is it at the bottom or halfway up the mountain? #交易之声: Your experience deserves to be heard $SNDK $SKHY $BTC $BTC Terry Duffy, Chairman and CEO of CME Group, stated publicly: If U.S. regulators approve crypto perpetual futures, traders will face significant tax and regulatory uncertainty. The core controversy centers on the funding rate mechanism: the funding rate for perpetual contract long-short periodic swaps, as defined by U.S. law, is more consistent with swaps; However, the current CFTC classifies it as futures. Currently, CME has officially initiated legal action against the CFTC approval, and the outcome of this lawsuit will determine the future direction of compliant perpetual contracts nationwide. 1. Simply Breaking Down Core Differences 1. Two Classifications, Vast Differences in Tax Rules ✅ If classified as [futures]: Subject to U.S. Tax Code Section 1256, profits and losses enjoy a mixed rate of 60% long-term capital gains + 40% short-term gains, with a lower tax burden for trading institutions. ⚠️ If determined as a [swap]: no longer enjoying preferential policies, all profits are taxed at the ordinary income tax rate, and institutional transaction costs increase significantly. 2. Completely Different Regulatory Rules Futures and swaps have different regulatory frameworks, clearing requirements, entry thresholds, and information disclosure standards. If the court overturns the existing classification, approved U.S. domestic compliant perpetual businesses will face a huge impact. 3. Current Status: Rules in a Vacuum So far, the IRS has not issued any formal tax guidance for crypto perpetual contracts. Even if the court finalizes the product characterization, the IRS will still need to issue separate documents clarifying the tax details, which is a long-term issueDamn, another project with hundreds of thousands of participants is gone
Yesterday, the grassroots project ODY (Odyssey) went to zero overnight
Before the exit scam, there were over 350,000 token-holding addresses, liquidity once approached $18 million, and 24-hour trading volume was close to $20 million
In the entire BNB Chain, this is not a small project
I checked the contract; it is not fully decentralized as advertised. The contract retains management rights such as Mint (token issuance), governance, whitelist, and tax rate modification
In other words, LP lock-up ≠ no token issuance
If the address with permissions can continuously mint tokens and sell the newly minted tokens into the liquidity pool, they can keep extracting USDT from the pool
The project conducted large-scale token minting and cashing out in the final stage
ODY hardly relies on Twitter or overseas communities; it mainly depends on offline grassroots promotion, community meetings, and referrals among acquaintances
Many people enter grassroots projects usually because: "My friend made money," "Hundreds of thousands have already joined," "Such a big community can't possibly run away"
The biggest enemy in crypto has never been regulation, but these Ponzi schemes disguised under the blockchain banner, repeatedly draining the entire industry's trust dry Tonight, the most eye-catching in the U.S. stock market was not the index, but the storage sector. The Dow rose 0.59%, the Nasdaq gained 2.68%, and the S&P 500 gained 1.28%. Compared to the steady strengthening of the broader market, some storage stocks performed like "sudden takeoffs": SK Hynix rose 17.15%, Micron Technology gained 16.85%. A nearly 17% increase in one day is clearly not just an ordinary rebound. Capital is seeking new outlets for the AI market Some time ago, when the market mentioned artificial intelligence, attention was mostly focused on popular companies like Nvidia, Microsoft, and Meta. But as core asset valuations rise, capital naturally seeks new opportunities along the industry chain. Storage chips stand right in this position. Whether it's AI servers, data centers, or large model training, all rely on high-speed reading and writing of large amounts of data. As computing power demand continues to grow, the market will begin to reassess the value of storage. From this perspective, the sharp rises in SK Hynix and Micron may reflect the expectation that funds are trading a certain expectation: the AI market is not over, but is just beginning to spread from the most crowded core stocks to other segments of the industry chain. However, it should be noted that this judgment is based solely on market performance. A single-day surge may indicate strong capital sentiment, but it cannot alone prove that a new trend has formed. The index is rising, but the market is not celebrating across the board. Another detail worth noting is the performance differences between sectors. The Nasdaq rose 2.68%, clearly outperforming the Dow and S&P 500, indicating capitalRebound ≠ reversal, $ETH surged 4%, $QQQ was dazzlingly green, and the market was waiting—whoever showed weakness first would set today's tone.
Look at the numbers
$BTC 65,283 +1.45% $ETH 1,952 +4.14%
$QQQ -1.12% $SPY +0.10% $IBIT -0.82%
$DXY -0.15% $GLD +0.10%
Hormuz and crude oil are still adding variables to inflation expectations, US Treasury yields and the shadow of Fed tightening continue to weigh on valuations, and the dollar isn't a backdrop—just a quick adjustment of the exchange rate line can disrupt the rhythm of $QQQ$SPY. It's not surprising which switch gets triggered in today's market $MSFT $MU $SNDK $MSFT $AMZN $META $GOOGL
$ETH Elasticity is clearly stronger than $BTC, short-term risk appetite is rising, but $QQQ is sinking downward, and money is shrinking into defense. $IBIT Weaker than spot $BTC; if ETFs weaken, it means the spot market isn't as strong; $DXY If it breathes a little easier, risk assets can catch their breath, but once tightened, it quickly turns hostile; $GLD Still quietly rising, safe-haven funds haven't fully withdrawn—don't be fooled by the hype.U.S. GDP came out at 1.5%, below the expected 2.1%, with macro data tugging between bulls and bears.
A weakening economy weakens risk appetite and temporarily suppresses $BTC; However, economic cooling will lead the market to gamble on Fed easing, with medium- to long-term positive expectations.
U.S. Treasury yields have already risen, cushioning the tightening pressure from economic weakness, so there is no need for excessive panic. Next, focus on future inflation data and Fed officials' speeches to determine the long-term market outlook.2014: Mt. Gox collapses, BTC at $200, bottoming out after 3 weeks.
2018: BitGrail collapsed, BTC at $3,200, bottoming out after 2 weeks.
2022: FTX collapsed, BTC at $16,000, bottoming out after 2 weeks.
2026: BitMEX collapses, BTC $63,000, bottoming out in 2-3 weeks?
Every time, the market says, "This time is different."
Every time, the market is wrong.
The difference is: the market caps of BTC in the first three rounds were $2B, $20B, and $300B respectively. Now it's $1.3T.
Same rules, but on a larger scale. $BTC $ETH $SOL $TSLA $NVDA $SPCXJuly 30, 2026 The Korean stock market in July is experiencing a deeply painful "deleveraging" crisis. The KOSPI index plunged about 40% in a single month, with nearly 700,000 retail investors losing everything in the collapse of leveraged ETFs. What followed was an internal regulatory technical assessment of the "renewed short selling ban"—this scene came just four months after South Korea lifted its 17-month ban on short selling. "Unlocked—crash—shouted—re-lifted"—the Korean stock market seems to have fallen into a self-recurring vicious cycle. *Every time the market fluctuates sharply, short selling is brought up as a "firefighter," but once the fire is extinguished, it has to be reintroduced for internationalization. Is this repetition a stopgap measure, or a deeply rooted "systemic disease"? A "Capricious" Policy History South Korea's attitude toward short selling is arguably the most volatile among major global markets. * 2011 European debt crisis: First comprehensive ban on short selling, lasting several months. * 2020 COVID-19 pandemic: Again fully banned, then gradually restored. * November 2023: Again banned under the pretext of cracking down on "naked short selling," only reinstated at the end of March 2025. * July 2026: Only four months after resumption, discussions about restarting the ban resurfaced. This frequent "policy switch" has led international investors to jokingly call the Korean stock market a "policy testing ground." Each ban is accompanied by the slogan of justice to "protect retail investors," but each lifting is forcedThe biggest risk in South Korea this time is not that Samsung and SK Hynix are not profitable, but that the market is experiencing a liquidity crunch.
When foreign capital withdraws, margin financing positions are forced to close, and programmed trading is sold simultaneously, prices briefly detach from fundamentals and enter a "whoever sells first, lives first" stage.
But every extreme deleveraging in history has led to a redistribution of chips.
Panic is responsible for crashing low prices.
Real capital usually only enters after the stampede ends.Microsoft's big bullish candle after hours, I don't think it's because it will save money, it feels more like the market finally sees AI starting to generate revenue.
OKX 19:31 xMSFT quoted at 429.28 USDT. Quarterly revenue in the earnings report is $90 billion, while the market was originally expecting only $87.62 billion. As for "cutting capital expenditures," don't be misled by the headline: management said the actual investment plan hasn't changed, about $175 billion is more an accounting presentation change related to leasing data centers.
$XMSFT is already very excited now. If it can hold 420 after the open, the sentiment can continue; if it spikes then falls back to 405, I will treat that as after-hours funds exiting first. Like it or not, chasing highs still depends on volume.
#EarningsObserver: Microsoft cloud revenue breaks 100 billion, but Meta's guidance disappoints—AI stories are diverging? #MicrosoftEarnings #USStocks #AI #TechWatch#财报观察员：微软云收入破千亿，Meta却指引拉胯——AI故事分化了？
AI and semiconductors launch a full-scale counterattack: Why did US stocks surge at the open?
Today, US stocks opened significantly stronger, led by the Nasdaq, with capital rapidly flowing back into the AI and semiconductor sectors.
The core driver of this rally is not a sudden dovish turn by the Federal Reserve, but Microsoft’s earnings report reaffirming AI demand.
Microsoft reignites AI trading
Microsoft’s quarterly revenue reached $90 billion, up 18% year-over-year; EPS was $4.74, up 23%; Azure grew 43%.
What truly excited the market was Azure.
The growth rate exceeded expectations, indicating that enterprise demand for cloud computing and AI computing power remains strong. Microsoft has continuously increased investment in data center construction, and the market has long worried about excessive capital expenditure and slow return realization.
But this earnings report proves:
AI investment is not just burning cash; revenue has started to catch up.
After Microsoft’s earnings release, capital not only bought Microsoft shares but also simultaneously lifted Nvidia, AMD, and Broadcom.
The logic is straightforward: cloud providers continue to expand data centers, so there are still orders for GPUs, AI accelerators, and high-speed networking equipment.
Microsoft essentially validated demand for the entire AI computing power industry chain.
Semiconductor sector rebounds comprehensively
The semiconductor sector trades on the same logic.
Nvidia and AMD benefit from computing power investments; Micron and SK Hynix benefit from HBM demand; Broadcom benefits from growth in network chips and custom AI chips.
As long as tech giants do not significantly cut capital expenditures, it is difficult to be outright bearish on this industry chain.
The only one hurt is Uncle Zuckerberg
Microsoft proves AI investment is translating into revenue, so the market naturally is willing to give it a higher valuation.
But Meta faces the problem that capital expenditures keep rising, AI investment is very aggressive, yet it is difficult in the short term to directly prove how much new profit this money will bring.
So the market’s attitude toward Microsoft is: spending a lot, but earning fast. Toward Uncle Zuckerberg, it is: money is indeed being spent, but returns still need to wait.
Federal Reserve pressure still exists
The Federal Reserve has not released any obvious new positive signals, and the pressure of high interest rates has not disappeared.
It’s just that today the market chose to trade corporate earnings first, temporarily putting macroeconomic negatives aside.
So the core of this rebound is clear:
Microsoft’s performance proves that AI capital expenditure is not just burning money; revenue growth has begun to materialize. The market has bought back AI and semiconductors, and for now, the only one hurt is Uncle Zuckerberg, who is still wildly spending money. $MU $SKHYNIX $XMETA @你的爱播Misa @OKX中文 🐋 鲸鱼观察：谁还在2026年看好这些币？
=> $LINK => $TAO => $SOL
=> $FET => $SNDK
=> $MU
=> $AAVE
=> $ETH
=> $UNI
=> $NEAR $LAB $SOL $SSV $AR $LDO
仍然早期。仍在积累。$GRVT has delivered an explosive debut, surging over 380% in a very short time. After the initial rally, price is cooling off as early profit-taking kicks in. Chasing green candles here carries higher risk. Wait for a healthy pullback or a confirmed breakout before considering a new entry. Patience often beats FOMO in fast-moving markets. 🚀📈 In the hot AI market environment, Microsoft has chosen to tighten capital investment, turning a "negative news landing into a positive" scenario.
The market's core focus has quietly shifted: from competing to pour money into expanding computing power in the past two years, to now prioritizing profitability quality.
Reducing ineffective capital investment can directly improve the profit statement, which is exactly the current preference of capital.
However, risks also exist. Will cutting expenses slow down AI infrastructure deployment? Whether long-term competitiveness is affected still requires continuous monitoring and verification.
$MSFT
#微软逆势下调资本开支，盘后涨8.5% $SNDK is down 42% locally, and its best bounce ever is only +14%. That tells you how weak memory names are right now.
Yesterday's no-move, even with good $META/$MSFT capex headlines, shows there’s zero conviction here. With how badly the market’s deteriorated, another leg down wouldn’t shock me.
I’m not comfortable, but I still expect a 20%+ recovery soon. Just don’t try to lever the bottom unless you’re a pro.
Long-term I’m still bullish on memory, but the playbook is different now. This isn’t April-June anymore. After a selloff this big the market needs to digest — no instant V-shape.
Have a plan. Stick to it. GL
#DailyOrbit #Fed3Dissents
#MSFTCutsCapex @OKX Orbit 📉 [Market Report] Leverage Hell, Blood Flows Like Rivers!
On July 29, $SOXL closed at $91.99, down 69.5% from the 52-week high of $302. During the same period, SOXX fell only 29.1%. On July 30, it rebounded pre-market to around $112.7, but remains deeply mired in a technical bear market. SOXL needs to surge 228% to break even, while SOXX only needs to rebound by 41%. The daily reset mechanism accelerates the erosion of principal during the ongoing decline.
---
📊 [Resistance & Support] $112.7 is the life-or-death line
Pressure level (ceiling above)
🔴 $135: Upper boundary of the recent forced liquidation zone
🔴 $170-$180: Key mid-term resistance zone
🔴 $200: Bull-bear dividing psychological threshold
Support Level (Lower Lifeline)
🟢 $101.8: Intraday low on July 29, a recent "bottom" defense line
🟢 $100: Psychological level of the rounder number; breaking below would trigger a larger wave of stop-losses
🟢 $91.99: Closing price on July 29, historically oversold zone
---
🚀 [Positive Factors] The last stubbornness of the bulls
💎 Strong industry fundamentals: Omdia forecasts China's semiconductor market to surge 92.9% to $812 billion by 2026; SEMI forecasts global equipment sales to increase by 23.2% to $165.9 billion, setting a new record
📈 Giant Results Exceed Expectations: SK Hynix Sets Record Quarterly Operating Profit; Samsung Electronics Semiconductor profits are strong; TSMC raised its capital expenditure guidance to $60-64 billion
🤝 Long-term contract price locking protection: SK Hynix has finalized about 10 long-term supply agreements; Samsung executives predict that supply tightness will persist until 2028
---
💀 [Negative Factors] The deadly weapon of bears
😱 Leverage backlash: SOXL plunged nearly 63-69.5% from its June 22 high. The daily reset mechanism accelerates wear down during a volatile market, requiring a surge—170%-228%.
🇰🇷 Asian storm impact: After SK Hynix's earnings report, the KOSPI index plunged as much as 13% intraday; Goldman Sachs says selling pressure triggered a record wave of trading halts
📉 AI Faith Shaken: Meta's Free Cash Flow Plunges 91%, Triggering Tech Panic; The top three SOXL holdings—Micron, Nvidia, and AMD—continue to face pressure
---
💰 [Performance Guidance] Strong industry vs. stock price divergence
The Semiconductor Industry Association expects the industry size to surpass $1 trillion by 2026. TSMC's Q3 revenue guidance is $44.6–45.8 billion. SOXX attracted $5.4 billion in inflows in July. However, SOXL is designed to track daily 3x returns and is not suitable for long-term holding—it amplifies gains during one-sided rises and accelerates destruction during fluctuations or declines.
---
🎯 [Wall Street Price Target] Analysts are deeply divided
MarketBeat has set a $SOXL price target of $135.47 based on consensus of its holdings. Third-party platforms predict $269 by 2026. However, in an extremely bearish scenario, it could drop to $65. Currently, $112.7 is at the lower edge of the forecast range, implying about 20% upside to $135. However, caution is needed—the target price of leveraged ETFs is limited in reference value and depends more on the performance of the underlying semiconductor index.
---
⚠️ Triple leverage is a double-edged sword. The industry fundamentals haven't collapsed, but leverage losses are a real black hole. Can the 112.7 yuan hold on? Keep a close eye on the $100 rounder level—holding it is a technical rebound starting point; if not, it will continue to seek a bottom toward $91.99 or even lower. Leverage players, please buckle up! $SOXL #美联储三票主张加息, PCE becomes a new highlight tonight. #微软逆势下调资本开支, up 8.5% in after-hours trading. #财报观察员: Microsoft Cloud revenue surpasses 100 billion, but Meta's guidance is disappointing—Is the AI story diverging? According to the latest market trends (as of US intraday on July 30): • Latest price: Around $1,200 (rebound over +18% from yesterday's close of $1,015.89), intraday high once approached $1,240 • Recent trends: July suffered a heavy blow, more than halved from the late June high of $2,200+ (monthly drop over 50%), with several consecutive days of sharp declines. However, today saw a clear rebound with active trading as the memory sector recovered overall. • Performance this year: Still with staggering cumulative gains (YTD over 300-400%), one-year gain over 2000%, but significant short-term volatility. My independent market analysis (based on current market conditions) 1. Short-term rebound logic: AI data center demand for enterprise-grade SSDs and high-capacity NAND remains strong, with technical recovery after recent sector overselling. Long-term contract agreements are smooth and gross margins are high, with fundamentals not fundamentally deteriorating. 2. Medium- and long-term outlook unchanged: SanDisk is a pure storage player after independence, benefiting from AI infrastructure capital expenditure. Supply-side expansion is physically limited (cleanrooms, equipment, etc.), so tight supply and demand may persist. However, caution is needed regarding industry-wide cyclical fluctuations and valuation correction pressure. 3. Current Opportunities and Risks: • Opportunity: After a correction, valuations will become relatively attractive, making it suitable to watch earnings reports (expected in early August) and AI-related order verification. • Risk: Extremely high volatility; if AI spending growth slows or competition intensifies, pressure may continue. Heavy positions are not recommended to chase gains. PersonalMU surged 15% from underwater today, and I made a purchase
Today, storage stocks rebounded across the board, but what really made me act wasn't SanDisk—it was Micron.
As of my market view, MU had risen about 15.5%, rising intraday from around $714 to around $856, with a fluctuation of nearly 20% from low to high. That morning, he was still struggling underwater, and in the blink of an eye, he blocked the door with the person who had missed the day before.
I also bought a MU observation warehouse.
But I didn't suddenly believe in the storage cycle just because I saw a big bullish candle, nor did I think that after such a big drop, it would definitely hit bottom. What really interests me is: the logic behind MU's rise today is not exactly the same as SanDisk's.
SanDisk had fallen too hard earlier. Today's big bullish candlestick clearly included short covering, disappearance of panic trading, and short-term bottom-fishing. MU certainly has this kind of power, but behind it lies an even more important layer — the market is rediscovering, and AI companies' money is still pouring into data centers.
Microsoft's latest quarterly capital expenditure reached $41 billion, with about two-thirds allocated to short-cycle devices such as GPUs and CPUs; Azure's annual revenue also surpassed $100 billion for the first time.
What does that mean?
A few days ago, the market was still worried whether AI capital spending had already burned too high and whether cloud providers would start hitting the brakes. The answer Microsoft gave was: demand hasn't disappeared, and computing infrastructure continues to expand.
The GPU handles computation, while the storage handles feeding.
The more computing power servers are built, the harder it is to avoid the demands for HBM, DRAM, and data storage. Although MU was recently criticized as a symbol of "too crowded AI transactions," it is also one of the most direct memory chip targets in the U.S. market.
Even more interestingly, MU had fallen about 25% for four consecutive trading days, and at one point fell about 36% throughout July. Samsung just delivered strong storage performance and expected tight storage supply to persist until 2028, but MU initially only rose about 2.3%, indicating the market was unwilling to believe any positive news at the time.
This is the truly subtle part of today's rally.
It's not that the storage industry suddenly improved overnight, but rather that the market was already pessimistic a few days ago: when good news arrives, no one believes it, and as soon as stock prices drop slightly, everyone rushes to sell.
Data shows that retail investors recently sold stocks at the fastest level since the pandemic collapse, while storage stocks like MU, SanDisk, Western Digital, and Seagate together accounted for about 88% of retail net sales that day.
Yesterday, they were lining up to cut losses, and today MU has risen 15% from underwater.
To put it bluntly, those who got beaten today aren't just the bears, but also those who just couldn't stand the volatility and dumped their chips at low levels.
So I bought MU not to bet it would immediately hit a new high, but to observe one thing:
When those who most want to sell have already sold out, how much positive news do storage stocks need before they can rise again?
If today is just short covering, then after a big rally, it's easy for the price to spike and pull back. The trading volume looks alarming, but there's no capital to continue the rally.
But if MU can hold most of its gains and no panic selling reappears during subsequent pullbacks, then the significance of this bullish candlestick is not just "falling too much and then rebounding," but may indicate that funds are starting to reprice AI storage.
SanDisk rose even more sharply today, but I'd rather keep an eye on MU.
Because SanDisk is like an emotional explosion, MU is more like the market answering a question:
AI money is still burning; can storage really be considered the first batch to be abandoned?
I'll buy a bit first and see if today's big bullish candlestick is a temporary surrender by bears, or is it the first day storage stocks return to the table.
$MU The local decline in $SNDK is -42% and the biggest rebound we've had since its inception is +14% (which is very limited)
Yesterday's reaction to the lack of a raise + good capital outlook from META/MSFT is not very based on memory names.
At this point I wouldn't be surprised if we have another leg due to the severity of the deterioration in the market.
Not comfortable at all, but I think we'll see a 20%+ subsistence soon anyway. (Don't try to time the bottom with a lever unless you're a speculative protégé.)
As for the long-term allocation, my positive outlook on memory has not changed, but the PA indicator now will be very different from the April-June only rally.
After this type of selling, the market needs time to digest and will not just reflect in a V-shape.
Make sure you have a plan in place and respect it... GLSpaceX$SPCX secured new orders for the U.S. Space Force, directly boosting market sentiment in the short term.
It must be admitted that Musk is very skilled at building capital narrative chains, continuously creating catalysts by combining project progress, government orders, and social media.
Based on Tesla's past market performance, the market is naturally willing to give the aerospace sector higher valuation potential, but whether the market can continue to strengthen cannot rely solely on stories.
$SPCX The real key test lies in the financial report released on August 4.
#SpaceX获 $1.6B US military contract, stock price plunge sparks controversy between two camps #SpaceX获 $1.6B US military contract, stock price plunge sparks controversy between two camps
The $1.6 billion Space Force contract edged up 0.31% in after-hours trading. In the past, orders of this level could have triggered a major bullish candle for SPCX. The market has now become numb.
1. What exactly did this 1.6 billion buy with this 1.6 billion yuan?
Eighteen Falcon 9 launches, to be completed by the end of 2027, deploying space-based detection and tracking satellites. In less than two months, the military awarded SpaceX nearly $6 billion in orders. Normally, this reaction shouldn't be the case.
2. Why can't it be pulled?
It had only risen by $1 before the market opened. Several reasons have been added together:
(1) The launch failed on July 21, and Starlink launches were suspended. 1.6 billion yuan can indeed hedge operational risks, but what the market wants is a signal of "safe recovery," not the orders themselves. Money can solve cash flow, but it can't solve trust.
(2) The execution period is set to the end of 2027. Previously, the market expected Starship to fully undertake military launches by 2027, but now the contract explicitly uses Falcon 9, effectively stamping the official seal—the timeline for Starship to replace Falcon has been postponed at least until 2028. While these positive factors have materialized, it has also revealed that Starship's commercialization progress has fallen short of expectations.
(3) First round unlock approaching. At the beginning of September, 44% of the internal shares were available for sale, and the outstanding shares increased by 900%. 1.6 billion orders simply can't withstand unlocking billions in the tens of billions.
Third, the two factions started arguing
Bulls say this is a long-term cash flow ballast stone, with analysts' average target price of $239. Bears say the 1.6 billion order isn't even enough to fill the fraction of the first round unlocked, and the SPCX has dropped from 225 to around 110. The 0.31% post-market gain is the market betting with its feet. Both sides have their own reasons, but their directions are completely opposite.
4. My personal opinion
1.6 billion yuan is indeed real money, but SpaceX's biggest problem right now isn't a lack of orders. At the time of the IPO, the market value was 2.6 trillion, but now it has dropped to around 1.6 trillion. The market is repricing forward narratives in a high interest rate environment, and 1.6 billion orders cannot change this trend. Before the first round of unlocks, any positive news could become a reason to sell. Only after the chip swap in early September is completed will SpaceX's true pricing come to light.OpenAI's "AI stock god" has also been taught a lesson 📉 by the market
According to the Financial Times, Situational Awareness, managed by former OpenAI researcher Leopold Aschenbrenner, recently sought additional funding after the AI stock crash.
Last year, it was hailed by the market as the "AI Stock God of Wall Street," and now it is beginning to feel the power of high concentration and leverage.
However, there are quite a few issues with the claims circulating online about holding positions.
The fund is indeed heavily invested in AI infrastructure, power, and storage. According to Q13F, public long positions include Bloom Energy, SanDisk, CoreWeave, IREN, and Core Scientific.
Additionally, the fund disclosed holding 12.41 million shares of Nebius, accounting for about 5.6% of NBIS's outstanding shares. This is a large number, but it cannot be directly called "NBIS accounts for 40% of the fund's position"; the two are fundamentally different concepts.
Oracle and AMD cannot be considered ordinary bullish losses.
The main positions disclosed by the fund on these two companies are put options. If the stock price falls in July, this position could theoretically be profitable. As for how much the long-short combination ultimately lost, there is currently no complete data from the outside world.
What confirms this is that AI infrastructure, power, and storage stocks generally experienced significant drawdowns in July, making it difficult for funds focused on these sectors to emerge unscathed. ⚠️
But just looking at the declines of a few stocks and announcing a fund liquidation or a loss on a single position is equally unreliable.
This incident really shows only one thing:
There is no ever-right stock god in the market.
What makes Buffett impressive isn't that every investment makes money, but that after countless bull and bear cycles, he can still stay at the table.
A single market rally can create many "stock gods."
Only those who survive the full cycle are qualified to leave their names. 👀 #韩股波动剧烈引监管介入, the finance minister apologized for leveraged ETFs Last night the Fed's drama ended, let's talk about the market.
Bitcoin is now hovering around $64,000-$64,400, Ethereum just above $1,900. After the decision was announced last night, it first dropped about 1%, hitting a low of 63,890, then slowly recovered some.
First, the biggest event last night — the FOMC.
The result: 9 votes in favor of keeping interest rates unchanged, 3 votes against, advocating a rate hike. The Cleveland, Minneapolis, and Dallas Fed presidents jointly voted for a rate hike. This is the first time since 2016 that three dissenting votes aligned.
Nominally no rate hike, but the signal from the 3 opposing votes is even more hawkish than a hike. The market's expectation for a September rate hike has surged to 82%. Previously, the probability priced in for keeping rates steady was about 65%, so the market volatility after the result wasn't large, but tightening expectations remain strong.
Geopolitical issues are also adding chaos.
After the Iran attack, the US and Saudi Arabia launched a joint retaliatory strike against Iraqi armed forces. Oil prices jumped nearly $4 to $83. Inflation pressure remains high, so the probability of a September rate hike stays elevated.
ETF fund flows continue to diverge.
Bitcoin ETFs have seen net outflows for the fourth consecutive day, with BlackRock's IBIT withdrawing $54.83 million in one day. Over the past 7 days, Bitcoin ETFs have collectively outflowed 3,170 BTC, roughly $200 million.
Ethereum is the complete opposite. On July 29, another 2,000 ETH (about $3.8 million) flowed in. Over the past 7 trading days, Ethereum ETFs have accumulated inflows of 20,277 ETH, worth $38.49 million. BlackRock's ETHA is the main buyer. The trend of money moving from Bitcoin to Ethereum continues.
From a technical perspective, Bitcoin faces short-term resistance at 64,700, with support at the lower Bollinger band around 63,500, and further down at 62,500. Ethereum is slightly weaker than Bitcoin, with clear resistance at 1,935 and strong support between 1,885-1,890.
In short, Bitcoin is stuck hovering around 64,000 — no rate hike from the FOMC means short-term negative factors are mostly out, but the 3 opposing votes imply a real possibility of a September hike, so funds are cautious. On the ETF side, Bitcoin is flowing out while Ethereum is flowing in; institutions are rebalancing, not fleeing.
Those with positions should watch 63,500 closely; if it breaks, look for 62,500. For those looking to buy the dip, wait for Ethereum to pull back near 1,900 or try light positions on Bitcoin around 63,800-64,000. There's no need to chase highs at this point; wait for clearer direction. $BTC $ETHSanDisk (SNDK) saw a big rebound on the evening of July 30 (after-hours/night session once surged over 17%, marking the strongest single-day rebound in weeks), with three core reasons:
1) Overall oversold recovery in the storage sector + major banks bullish on NAND prices
After peaking in June, SanDisk fell from 2354 to just over 1000, a retracement of more than half, with the sector also deeply adjusted during the same period. Morgan Stanley's Joseph Moore reiterated that "Q2 to Q3 NAND contract prices will rise at least 25%, with no signs of easing in channels," while JPMorgan and UBS adjusted their views, saying "the pullback is a buying opportunity, not a sign of deteriorating fundamentals." Tonight, Microsoft and Meta expressed optimism about capital expenditure on computing power, and Samsung said that memory shortages in 2027 will be tighter than this year, lifting the entire storage chain (Micron +10%, Seagate +16%, Western Digital +14%, Kioxia ADR +11%). SanDisk is the most elastic among them.
2) The AI data center enterprise SSD demand narrative remains intact, validated by Seagate's earnings
Seagate's strong earnings the previous trading day confirmed that "AI inference clusters consume enterprise storage" is not empty talk; SanDisk itself will release Q4 earnings on August 5 (next Wednesday), and the market is already pricing in a "233% quarter-over-quarter increase in data center revenue and sustained 78% gross margin" scenario. The earlier sell-off was about valuation and sentiment, not orders — institutional investors are aware of this, so value funds dared to buy near 1000.
3) Technical extreme oversold + macro risk appetite rebound
SanDisk's stock price is nearly 40% below its 50-day moving average, Wall Street consensus target prices imply doubling potential, short covering + dead cat bounce + Nasdaq's 2% rise tonight bringing back risk appetite, with three forces combining to pull the single candlestick into a V-shape. The options market is pricing high volatility around the August 5 earnings report; tonight is a "pre-earnings expectation repair" rather than a new fundamental catalyst.
#美股全线走高，加密股领涨 #美股 #闪迪 $SNDK $MU Same drop, different speed
Put the data together and get a sense of it:
Bitcoin fell 54% — in 268 days
Silver fell 54% — after 169 days
SanDisk (SNDK) fell 55% — after 36 days
SK Hynix fell 53% — after 34 days
Similarly, semiconductors have seen a halved-level pullback, moving seven to eight times faster than crypto and precious metals.
From 268 days to 34 days, the intensity of this round of storage adjustments is truly impressive.🚨 The Biggest Weakness in Some Stock Perpetual Markets: Oracle Risk
The recent SK Hynix perpetual incident highlighted a key issue in synthetic stock markets:
There isn't always a single, unified reference price.
Here's what happened:
📉 During early trading, an oracle pricing anomaly caused the SK Hynix perpetual price on one platform to briefly collapse by nearly 20%.
The interesting part?
Another platform also experienced a drop—but it was much smaller.
Why?
Arbitrage algorithms reacted immediately.
Their logic is simple:
✅ Buy where the price is abnormally low.
✅ Sell or hedge on another venue.
✅ Capture the spread.
This activity can transmit volatility across platforms, even if only one platform experiences the original pricing issue.
Why didn't every venue fall to the same level?
1️⃣ Different oracle designs and pricing sources.
2️⃣ Transmission delays.
3️⃣ The abnormal price wasn't a genuine market price and quickly reverted.
The broader lesson:
Oracle-dependent products introduce an additional layer of risk beyond normal market volatility.
For traders, it's important to understand:
• How prices are calculated.
• Which data sources feed the oracle.
• Whether the platform has protections against erroneous prints.
Technology reduces friction—but it doesn't eliminate market infrastructure risk.
Understanding how a market works can be just as important as understanding where price is going
#Fed3Dissents #MSFTCutsCapex #AIStoryDiverges
$BTC $ETH $SNDK $MU, $SNDK, and $SKHYNIX all saw a strong rebound today, with the entire storage sector collectively rallying. A critical time window is about to arrive. Tonight, after the US stock market closes, Apple and Amazon will successively release their earnings reports, followed by Seagate Technology tomorrow. The entire storage industry chain's expectations will soon undergo a re-pricing. Many analysts habitually focus solely on K-line charts to find support and resistance, easily overlooking the most important core variable in this storage sector cycle: the real end-user demand. Whether NAND and DRAM prices can hold steady, and whether Micron, Samsung, and SK Hynix's price hike expectations can continue, fundamentally depends on consumer electronics demand and the purchasing strength of AI servers for storage. The upcoming earnings reports from these giants provide the most direct window to observe demand changes. Amazon AWS's capital expenditure plans directly affect upstream HBM and enterprise flash memory order expectations. Apple's hardware shipment and inventory data influence long-term consumer flash memory demand. Seagate, mainly focused on mechanical hard drives and enterprise storage, provides downstream customer guidance that directly impacts the overall sentiment in the NAND sector. Sandisk's business foundation is tied to flash memory demand, so its performance naturally fluctuates with the sentiment across the entire industry chain. The sector's price volatility ultimately reflects the market's continuous adjustment of expectations for this sector. Looking back at the recent performance of the storage sector, many stocks have steadily declined from their highs, with many experiencing a halving in value. After this deep decline, the current strong rebound has prompted a large amount of capital in the community to start bottom-fishing and positioning. However, I have recently been discussing with traders around me #财报观察员: Microsoft's cloud revenue surpasses 100 billion, but Meta's guidance is disappointing—Is the AI story diverging?
On earnings night evening, Apple and Amazon took turns handing over their papers. But the real highlight of the night may not be the numbers themselves.
🍎 Apple: Driven into 5 trillion yuan by "AI hedging," tonight it will be tested
Wall Street expects Apple's Q3 revenue to be about $108.9 billion, up 16% year-over-year, with earnings per share of $1.89. iPhone revenue is expected to grow by 20.5%, making it the fastest-growing segment; Services business revenue is expected to be about $31.4 billion, up 14% year-on-year.
But what truly made Apple the center of attention was its AI narrative.
Apple's stock price has risen 24% this year, making it the best performer among the "Seven Tech Giants." It doesn't win by spending money on data centers, but by "not spending money"—capital expenditure for fiscal year 2025 is only about $12.7 billion, which is on a completely different scale compared to Microsoft and Google's hundreds of billions in investment. While the market priced the uncertainty of returns on massive AI investments into its stock price, Apple unexpectedly became the beneficiary of the "AI risk-off trade."
The question now is: can this financial report hold up to this logic?
Apple's current 12-month expected price-to-earnings ratio is about 35 times, the highest since 2008. The options market's volatility in tonight's earnings report is close to 4%, far above the average actual volatility of about 1% over the past year. If the report shows gross margin pressured by rising memory chip costs, the 35-times PE valuation anchor could be shaken.
📦 Amazon: AWS's growth rate determines the direction of AI narratives
Wall Street expects Amazon's Q2 revenue to be about $196.5 billion, up 17% year-over-year, with earnings per share of $1.82. AWS growth is the biggest highlight—Bank of America forecasts it could reach 33%, a significant increase from 28% last quarter.
Amazon has missed earnings expectations for the past two quarters in a row, and tonight it needs to prove that AI investment is being converted into real revenue for AWS. If not only does growth meet targets, but a clear path to capital expenditure returns can be provided, the entire AI narrative will be reignited.
📅 There are still two major dramas ahead
On August 4th, the Future Memory and Storage Conference — Samsung, SK Hynix, Micron, and NVIDIA all attended, highlighting its high importance for the memory industry chain. HBM production capacity and 2027 contract pricing may be the most critical signals.
August 23-25: Hot Chips 2026 — the most anticipated technology conference in chip architecture and AI hardware. Next-generation technologies such as NVIDIA Vera CPUs, HBM architecture, and advanced packaging will be announced. Vera is based on the Arm architecture, featuring 88 custom cores, and is expected to be delivered to the Chinese market starting in August.
The tech stock narrative is shifting from "who has more GPUs" to "who can turn computing power into profit." Tonight's earnings reports from Apple and Amazon are the next touchstone for this switch.
$AAPL $MU $AMZN The reason behind SK Hynix's rally has been found!!
The main reason for SK Hynix's recent rally! and sharing methods for handling multiple orders!!
I'm Ci Ge. I've gotten long positions at 908.37 so far, and the floating profit is already very substantial. Today's big bullish candlestick directly pulled the market back. First, let's talk about why it rose, then how to handle it.
Today's aggressive rally is due to four overlapping drivers
First, Choi Tae-won's first personal purchase directly ignited the emotions. On July 30, SK Group Chairman Chey Tae-won bought 3,620 shares of SK Hynix, totaling about 4.8 billion KRW. This is the first time Choi Tae-won has directly held SK Hynix shares in his personal name. Before this, he had never directly held shares and only controlled them indirectly through SK Square. After the stock price dropped from 2.18 million won to 1.32 million won, the chairman personally stepped in to buy shares, signaling a strong signal. After the regulatory documents were disclosed, SK Hynix turned from a decline to a gain in pre-market trading.
Second, the storage sector collectively shorted and rebounded. The Philadelphia Semiconductor Index rose more than 7%, SanDisk surged about 20%, Seagate Technology rose about 18%, Western Digital rose about 16%, Micron Technology rose about 11%, and SK Hynix rose about 11%. The US chip and semiconductor sector surged collectively before the market opened, with a strong sector synergy.
Third, Microsoft's earnings exceeded expectations, and AI narratives were restored. Microsoft rose about 8.5% in after-hours trading, Azure cloud revenue grew 43%, surpassing $100 billion for the first time this fiscal year. As the first major tech company to lower its capital expenditure guidance for fiscal year 2027, Microsoft is spending less and earning more. Market concerns about AI hardware demand have been dispelled, and storage chips as AI infrastructure have benefited simultaneously.
Fourth, institutions collectively bullish, with Daiwa's target price at 3 million won and UBS's target price at 204 USD. On July 28, Goldman Sachs held a webinar for Korean storage experts, where experts believe that traditional DRAM prices will maintain double-digit percentage growth this year, and HBM has room for significant price increases or even doubling next year. Multiple forces have resonated together.
908.37 Follow-up Handling Method for Long Orders
The profit was taken in four installments. The first batch sold 20% near the current price (around the 1100-1150 range), locking in some profits. In the second batch, the 1200-1250 range is further flattened by 25%. In the third batch, the 1350-1400 area is further flattened by 25%. In the fourth batch, the remaining 30% above 1500 yuan was completely liquidated.
Moving stop is executed. The current stop loss has been moved up below 950, and for every 100-point increase, the price moves up by 50 points. At 1100, the stop loss was moved up from 950 to 1000; To 1200, moving up from 1000 to 1050. This ensures that profits are not all given back. If the price pulls back to the 1000-1050 area without breaking below the moving stop, you can add positions in that area, with overall stop-losses consistently below 950.
Ci Ge finished speaking. This deal was a chip brought out from the deepest panic, and Choi Tae-won personally stepped in—don't let short-term fluctuations scare you off. Set your stop-loss and take profits in batches—hold on. Think carefully. #韩股波动剧烈引监管介入, Finance Minister apologizes for leveraged ETFs. #美联储三票主张加息, PCE becomes a new highlight tonight. #比特币与纳指相关性大幅下降: Independence or Illusion $SKHYNIX $BTC $ETH SK Hynix finally stood firm once! Took profit on the long position, although not fully, but at least didn’t hold in vain 😭
---
【Take Profit Report】
📌 Asset: SKHYNIXUSDT (SK Hynix)
🎯 Direction: Long 17x
📈 Opening Average Price: 932.48
📊 Closing Average Price: 958.65
💰 Realized Profit: +3.03 USDT
📈 Realized Return: +39.24%
📊 Current Price: 1,033.65, 24h up 8.58%, peaked at 1,049.67
Although it kept rising after closing, missing out on some gains, after consecutive liquidations earlier, surviving is a victory. No greed, really no greed.
【Microsoft and Meta Earnings: Worlds Apart】
Last night after US market close, two tech giants released earnings with vastly different results:
Microsoft: Big beat, surged nearly 9% after hours
· Q4 revenue $90B, beating expected $87.6B
· Azure cloud revenue up 43% YoY, fastest growth since 2022
· Full 2026 fiscal year, Azure annual revenue surpasses $100B for the first time
· No increase in full-year capital expenditure, no "continued burning money" feared by market
In short: AI spending wasn’t wasted, now turning profitable.
Meta: Record revenue but profit collapsed, dropped over 10% after hours
· Q2 revenue $60.8B, all-time high, but net profit down 14% YoY to $15.8B
· Q3 revenue guidance below expectations
· Capital expenditure floor raised from $125B to $130B, still heavily investing in AI
· Free cash flow only $784M, four-year low, compared to $8.5B last year
In short: AI money burned, but no returns yet.
【What do these earnings mean for semiconductors?】
Microsoft’s report tells the market: AI computing demand is real, cloud providers are still aggressively buying chips. Azure growth accelerating from 40% to 43% means Hynix’s HBM memory is selling well.
Meta’s report warns the market: AI spending is intense, not every company can sustain it.
These two forces tug and pull, causing storage sector volatility. But Microsoft’s beat at least proves AI infrastructure investment logic remains intact — a solid positive for Hynix.
【Tonight’s biggest catalyst: PCE data】
The biggest positive tonight is actually the PCE inflation data:
· June overall PCE YoY 3.7%, in line with expectations, previous 4.1%
· Core PCE YoY 3.3%, in line with expectations, down from 3.4%
· MoM down 0.1%, first monthly decline in six years
Inflation cooling weakens the case for rate hikes. Plus, SK Group Chairman Choi Tae-won bought 3,620 shares of Hynix for the first time, a double positive that pushed Hynix from 900 to 1033.
Once the PCE data came out, my long position went straight from floating loss to take profit.
【Summary】
Tonight’s PCE data proves inflation is cooling, rate hikes are less urgent, and with the chairman personally bottom-fishing, this rebound in Hynix has fundamental support.
Though I closed early and earned less, surviving is what matters. After being liquidated 4 times this week, finally got some blood back 😭
$SKHYNIX
#美联储三票主张加息，今晚PCE成新看点
#财报观察员：微软云收入破千亿，Meta却指引拉胯——AI故事分化了？ Storage Backlash Variety: Hynix, Micron, SanDisk—Who Is the Real King, and Who Is Riding the Wave?
The US stock storage market collectively recovered its funds, and the three representative companies have completely sealed their respective personas, even their rebound postures mirror their own positioning.
$SKHYNIX SK Hynix: HBM king in competition, with the most restrained gains
Riding the halo of being the global HBM market leader and tied to Nvidia to reap the biggest AI dividends, it should be at the forefront of the rebound. But the earlier speculation was too aggressive, with institutions holding positions clustered, and every rally triggered take-profit orders, forcing the leading stocks to lose their sluggish slow bull vibe.
A quick comment: the technique is truly impressive, the stock character is truly tough, suitable for long-term trading, but for short-term traders who want quick profits, entering is torture.
$MU Micron Technology: North America's top student, steady as the priority
HBM4 has successfully entered mass production, with quarterly revenue exceeding $1 billion, data center business doubling year-on-year, and financial reports consistently beating expectations. A 13% increase is neither more nor less, neither standing out nor holding back, perfectly fitting its positioning as a "steady leader."
After all, it's the beloved son of North America, with a strong geopolitical buff and the strongest earnings certainty. It drops less when it drops, and never goes crazy when it rises—perfect for stable capital.
$SNDK SanDisk: Invisible dark horse, silently rising the most
Many people still cling to the impression that "SanDisk just sells USB flash storage cards," but they have long made a fortune from enterprise-grade SSDs and long-term contract orders with cloud vendors. Backed by Western Digital, the consumer sector is basically stable, the enterprise sector is growing rapidly, and when it falls, no one cares; when it rises, it leads by 20%.
The most ironic thing is: the stocks you usually don't mention in research reports actually rebound the strongest. Sure enough, A-shares and US stocks are the same—stocks that no one hypes up when they rise, so there's no selling pressure.
Looking at long-term logic, SK Hynix and Micron are the core of AI storage; But for short-term rebound elasticity, you still need to look at edge-bound stocks that have been completely bearish.
When it comes to investing, you always look at fundamentals in the long run, and in the short term, you look at sentiment and chips.
#美光暴跌后: Is it at the bottom or halfway up the mountain? $MSFT
From Frenzied Expansion to Prudent Spending: Microsoft is Changing the AI Investment Logic, Negative Impact on Storage Optical Modules and Other Semiconductors
The most noteworthy aspect of Microsoft's latest earnings report is not just the adjustment of capital expenditure from $190 billion to $175 billion, but the shift in AI infrastructure investment logic.
The previous market logic was:
AI demand growth → Cloud providers frantically build data centers → Massive procurement of GPUs, CPUs, HBM, servers, optical modules, power equipment.
Now Microsoft might be:
AI demand increases → First lease mature data centers → Directly install GPUs.
However, if Microsoft reduces capital expenditure and adjusts its data center strategy, it will put short-term pressure on the AI infrastructure supply chain.
The most affected areas include:
GPU: If cloud providers slow down data center expansion, it may impact the future GPU procurement pace.
Servers: AI server demand depends on large cloud providers' capital expenditure; a CapEx decline will affect order expectations.
HBM and Storage: Slower GPU deployment will affect high-bandwidth memory demand expectations, putting pressure on the DRAM and HBM supply chains.
Optical Modules: Demand for 800G and 1.6T high-speed optical modules depends on AI data center expansion; construction slowdown will affect market expectations.
Data Center Infrastructure: Including power, cooling, liquid cooling, transformers, etc., will also be impacted.
But it is important to note that Microsoft is not abandoning AI investment; it is changing the investment approach.
Previously:
Buying land and building data centers themselves.
Now:
More leasing of mature data centers and rapid GPU deployment.
The reason is that building large AI data centers usually takes 2 to 4 years, while leasing mature data centers allows for faster internal deployment of computing power.
This means AI demand may still exist, but the industry is moving from "unlimited expansion" into an "era of capital efficiency."
Future market observations should focus on:
1. Whether Microsoft continues to reduce CapEx.
2. Whether Amazon, Google, and Meta adjust capital expenditure in sync.
3. Whether NVIDIA GPU orders slow down.
4. Whether HBM inventory and prices change.
5. Whether server and optical module order growth declines.
If these signals appear simultaneously, it indicates the AI infrastructure cycle may be transitioning from a rapid expansion phase to an adjustment phase.
The core change is not the disappearance of AI, but the market shifting from "who invests more" to "who can achieve higher investment returns."Tonight's market was the most abnormal: the tech sector in the US stock market seemed to be on fire, but the crypto world still seemed like a beast crouching at the gate, only defending without attacking. Let's look at the strongest signals: QQQ was at $681.99, up about 3.06% intraday; NVDA at $196.47, up about 3.40%; TSLA at $307.70, up about 3.14%. This isn't just a sector getting lively on its own, but rather risk appetite resurging. Yesterday, people were worried about AI spending, valuations, and interest rates; today, funds suddenly started chasing elastic assets again. But the crypto world didn't surge immediately. BTC is near $64,755, with an intraday high of $65,040 and a low of $63,252, still holding above $64,000; ETH is oscillating around $1,624.95. On the surface, crypto assets are half a beat behind; But the most dangerous and tempting aspect of trading is often this kind of "half a beat behind." If US tech stocks keep rallying, why might the crypto world be re-named? Because global funds don't switch channels first; they first test the waters where liquidity is deepest and the consensus is strongest. Core assets like QQQ and NVDA rise first, indicating big money is willing to endure volatility again; Only when this sentiment is confirmed to be not a one-day trip will funds spread into more resilient BTC, ETH, SOL. What we really need to watch now is not whether BTC has risen by a few hundred dollars, but whether three things are happening simultaneously: First, can QQQ hold this momentum until the close; Second, whether NVDA will continue to be the flagbearer of the AI main theme