Orbit Post Sitemap
Positioning is "justice" In 1978, Sequoia Capital invested about $150,000 in Apple, and a few years later exited for about $6 million, making a 40-fold profit. In 2016, Buffett began buying Apple, investing about $36 billion cumulatively, earning over $100 billion, and remains a heavy position today.
Sequoia earned multipliers, while Buffett earned in amounts.
Multiples satisfy intelligence, while base determines destiny.
Buffett's position requires finding an elephant large enough; otherwise, the capacity cannot hold his large sum of money.
For individuals, the challenge of position positioning is: the opportunity is right in front of you, and you only bought 1%.
Choosing the right company doesn't mean making money. You can keep naming the top stocks, but your account may not be the best.
But heavy holdings are not a test of courage. The heavy position in value investing comes from its philosophy of "justice":
Because you truly understand a company, understand its business moat, and have a margin of safety in the price, you can hold a heavy position.
Berkshire's internal evaluation items look at "extremely high win rates," "decent odds," and "extremely high confidence." If these factors are high, the position ratio calculated using the Kelly formula will naturally be high.
Heavy investment is a conclusion of "justice," not a decision made at Qidian, nor is it a belief or wish.
Between "always heavy position" and "always dispersed," there is a third kind of wisdom: small positions testing and gradually increasing holdings. #从降息到加息, the Fed's disagreements are fully public $MSTR Selling 1,638 Bitcoins at an average price of $63,957 to repair preferred stock debt is the core contradiction in the asset side losing losses, which translates into contraction of market risk appetite and position rebalancing, and a game of interest repayment pressure on the liability side.
This single sale, which lost $11,462, cashed out approximately $104.7 million, including $52.4 million in dividends and $52.3 million in repurchase discounted Class A perpetual preferred shares. This move validates that the credit risk premium caused by liability-side discounting is suppressing management's long position risk appetite.
The current driving factors are: progress of credit premium recovery on the liability side, overall market risk appetite bias, and the transmission of selling pressure from spot positions. The cap on authorized sales was raised sharply from $1.25 billion to $5 billion, directly boosting risk aversion in the spot market to potential whale liquidation positions.
If macroinflation expectations remain sticky and interest rates remain high, preferred stock financing costs will not be effectively reduced, forcing companies to realize more assets to fill the liability gap. Although the position remains at 842138 and the net increase ratio has reached 97.9% year-to-date, tactical selling has disrupted the single liquidity transmission path of leveraged buying.
The trigger for the upward scenario is that the price of Class A perpetual preferred shares returns to par value, and Bitcoin's spot price climbs back above the average position price of $75,419. When preferred stock discounts narrow and liquidity pressure on the liability side is lifted, arbitrage positions will once again boost risk appetite, and the pace of selling 3,620 shares this year will significantly slow or even pause.
The downside scenario triggers the condition for the preferred stock discount to continue expanding, while rising market risk aversion triggers a chain reduction of long positions at high levels. If the liability-side discount rate remains high and management continues to use spot positions within the $5 billion authorized quota to cover losses, liquidity support in the spot market will face a major test.
The failure condition for this logic is that the company completely abandons preferred share buybacks or directly adjusts its capital structure financing model. Once the liability side constraints are lifted, market trading focus will quickly shift from liability-side risk transmission back to the macro interest rate environment and the supply and demand relationship for Bitcoin spot prices.
In the next 7 days, it is important to closely monitor the discount recovery of Class A perpetual preferred shares, as well as changes in actual realization frequency and spot depth under the $5 billion sale cap authorization.
#BitMine成全球最大ETH质押方 #CLARITY法案剩72小时, the motion was still not submitted #美伊重回谈判桌, and oil prices pulled backSeptember FOMC: It's not a matter of cutting rates or raising rates, but how much to raise
Stop being brainwashed by headlines like "Waller openly supports a rate cut in September." If you look back at his July 13 speech, and then look at the 9:3 vote from July 29, you'll find that what the market is discussing now is fundamentally a false proposition.
The real internal division within the Fed is not about "cutting rates or raising rates."
Let's clarify the facts first. The July 29 FOMC vote showed a 9:3 vote to keep rates unchanged at 3.50%-3.75%. Logan, Kashkali, and Hamak voted against it, all three advocating for an immediate 25 basis point rate hike. Waller, on the other hand, voted in favor—supporting keeping things unchanged, not favoring rate cuts.
These two concepts are worlds apart.
Waller was very blunt in his July speech: core PCE climbed from 3.0% in December 2025 to 3.4% in May 2026, and inflation is accelerating, not slowing down. What he meant in his exact words was that discussing rate cuts in the face of such data is "crazy." He didn't vote for a rate hike, but that doesn't mean he's a dovish figure; it's just that he felt the current data wasn't enough to support it, so he acted immediately.
So, what are the real internal divisions within the FOMC? It's not about "rate cuts vs. rate hikes," but about the difference between "raise now" versus "add again after another round of data."
Back to the trading side. Last week, when I saw three opposing votes come out, I manually cut my risk exposure by 30%. Not out of panic, but because CME FedWatch data is clear—the probability of a 25 basis point rate hike in September fluctuates between 64.7% and 74.5%. The market is not discussing whether there will be a rate hike, but rather pricing in it.
Now let's look at signals from the bond market. The 10-year Treasury yield is now at 4.70%, up 23 basis points over the past month and 48 basis points higher than the same period last year. Bond traders don't lie; this yield level is pricing in that tightening hasn't ended yet.
Now, putting the data together: core PCE 3.3%, overall CPI 3.5%, three Fed chairs publicly calling for rate hikes, interest rate futures setting a 70% probability for a September hike, and US Treasury yields hitting a nearly one-and-a-half-year high. Tell me, there's still room for rate cuts under this combination?
My judgment: on September 16, there is a high probability of a 25 basis point rate hike. Waller will not vote against it again, because there are still two key data windows in the next six weeks—the August 12 CPI and August 26 PCE. If these two data points don't cool off significantly, even Waller will be pushed toward rate hikes.
But I left myself a 25% margin for error. If the core year-on-year CPI falls below 2.3% on August 12, or the core PCE year-on-year on August 26 falls below 3.0%, the logic for rate hikes will loosen. At that time, I will immediately adjust my position direction.
For the crypto market, those who are long on rate cut expectations need to be very cautious. BTC surged to 64,000 on FOMC day before retreating, indicating that market funds are far from fully absorbing rate hike expectations. If a rate hike really happens in September, this round will see a real liquidity contraction pricing — not a flash crash, but enough to push the 63,000 level down one level.
Next, focus on two key points: the core year-on-year year-on-year CPI on August 12 and the core year-on-year PCE on August 26. These two data points determine which side Waller ultimately takes and determine whether the September FOMC will be 9:3 or 12:0.
#从降息到加息, the Fed's disagreements are fully public Fundamental Research Report $BLAST / Blast (L2/Sidechain) $3.20
To summarize: Blast ($BLAST) has an overall score of 53/100, with a rating that narrative emphasizes implementation. Looking at the three layers, the company team has cash reserves, the protocol network already shows signs of paid usage, and token value transfer still needs to be observed.
Project Overview: Blast (token $BLAST), L2/sidechain track. Focusing on native Pacman L2 yields. Benchmarked against ARB and OP. Traditional collaboration between enterprises relies on cloud servers and contract reconciliation, which causes gas surges, TPS constraints, and frequent cross-chain bridge security incidents during high concurrency. Public chains use a unified state machine for trustless settlement, reducing reconciliation costs. Average order value is $50-500/month, with settlement required in USDC or fiat currency. Narrative-driven tracks, bear market usage cut by 60-80%. Positioning the end-to-end vertical platform. Product implementation: The protocol layer is officially operational, and the on-chain dashboard shows protocol fees accumulating, showing signs of paid usage. Latest version not found, 60 valid submissions in the past 90 days.
At the user level, address MAU not disclosed, DAU not disclosed, 24-hour transaction volume $80.00M, TVL not found. Wallet addresses do not equal monthly active users of natural persons; large large addresses holding concentrated positions tend to overestimate the actual number of users. On the revenue side, user fees are not disclosed. Supply-side revenue is about 80-90% of user fees (to LPs and nodes), protocol treasury revenue is $3.4K, token holders buy and burn at an annualized rate, with no burn mechanism. 24-hour transaction volume is business turnover, not revenue. A company making money does not mean the protocol makes money, and protocol profits do not equal token holders making money. On the code side, 60 valid submissions in 90 days, 25 active contributors, latest version not found. GitHub is a Class A evidence that can be directly verified. Investment background: For company equity financing, look to PitchBook/Crunchbase (A-level); for token private and public funding, use whitepapers, release curves, and on-chain unlocked contracts (A-level); market makers and ecosystem funding are B-level and do not represent long-term holdings of tech VCs; for technical integration, look to API/SDK access evidence (B-level); strategic partnerships and logo walls are D-level. The use of NVIDIA GPUs does not equate to NVIDIA investment, and going public on exchanges does not equal strategic investment.
On the token side, total supply is 1,300,000,000, circulating 950,000,000 (73.1%), FDV $4.20B, next unlock is 2026-Q4 (+3.50% circulating), burn buyback annualized rate, no explicit buyback burn. Do you have to buy coins to use the product? Yes, strong value capture (Gas/Collateral/Service Access). Comparing with peers (unified criteria, no cross-sector random comparisons): In terms of circulating market capitalization, Blast $3.00B, ARB undisclosed, OP undisclosed. For FDV, Blast $4.20B, ARB undisclosed, OP undisclosed. In terms of annualized revenue, Blast $3.4K, ARB undisclosed, OP undisclosed. Regarding monthly active addresses or users, Blast has not disclosed, ARB has not disclosed, OP has not disclosed. Figures are based on public data snapshots; any omissions are supplemented by official self-reports or industry standards. Valuation, market capitalization $3.00B, FDV $4.20B, P/S 887,600.2x, FDV divided by revenue 1,242,640.3x. Pessimistic outlook: $3.00B at 50-70% off, oscillating within a neutral range; optimistic outlook: revenue doubling, burns landing, enterprise clients coming in, FDV corresponding to P/S, aligning with the top companies. Overall: Solid fundamentals (rating 53/100). The token value transmission path is unclear, with only governance incentives. Circulating market cap is relatively high relative to fundamentals, overdrawing expectations, and FDV is moderate. Risks to watch for: short-term large-scale unlocking and sell-off, long-term protocol revenue wiping out, token demand relying solely on incentives (once incentives are cut off, usage collapses). Next, focus on these numbers: protocol fee weekly, burn amount, active address retention, TVL/loan balance, and GitHub version release. Information sources are public, logic is self-developed, and does not constitute buy or sell advice. Data deviations exceeding 30% require reassessment.
Once the fundamentals are dismantled, how the market moves is another matter.
#基本面研报 #加密 #研究 #OKXOrbitNVDA climbed back above $206, HPE rose nearly 5%, and MRVL reached $199 before closing at $193. Every time I see the MRVL about to break 200, I feel like it's about to take off, but it always reminds me at the last minute: don't get too happy too soon......
The US manufacturing PMI hits a four-year high, indicating that demand for AI, electronics, and equipment is indeed still strong; The problem is that the economy is too strong, and the Fed may continue to be hawkish.
The most crucial topic tonight is AMD's earnings report. AI chip revenue, data center growth, gross margin, and guidance—if any one is missing, the entire semiconductor sector could be carried on a roller coaster.
On the crypto side, things are still the same: BTC is holding $63,000, ETH is still lying flat. Now, capital clearly trusts AI companies that can deliver earnings reports.Marvell announced it will release its fiscal 2027 second quarter results after the U.S. market closes on August 27, 2026, and will hold a conference call at 4:45 a.m. Beijing time on August 28; The company will also hold an investor day in New York on October 6, where CEO Matt Murphy and the executive team will present the company's strategy.
August 27 earnings report, October 6 investor day. To be honest, I'm not as excited about a regular financial report right now; I'd rather hear what management has to say about AI ASICs, cloud customers, and revenue targets for the coming years.
What this company lacks now isn't a story, but turning it into clearer orders and profits. If Investor Day can keep making empty promises, I might really be losing control.
$MRVL Whether the US and Iran have actually made a deal remains a Rashomon to this day.
Trump claimed that negotiations had begun. On August 3, he stated that, at the request of Saudi Arabia, the UAE, Qatar, and other countries, the large-scale strike against Iran was canceled, and that the U.S. and Iran would begin negotiations on the afternoon of the 3rd. He then insisted that negotiations are ongoing and will proceed in two phases: the first phase is opening the Strait of Hormuz, with full opening expected as soon as the 4th. Trump even threatened Iran with "either an agreement or a complete surrender," hinting at the possibility of a "decapitation" action.
Iran has completely denied this. On the 3rd, Iranian Foreign Ministry spokesperson Baghae made it clear that Iran is currently not negotiating with the United States, and there are no plans to receive or send delegations "these days." He clarified that all current consultations are focused on bilateral dialogues with Oman on safe navigation in the Strait of Hormuz. A senior Iranian official told Reuters that Foreign Minister Alagazi is making a religious pilgrimage in Iraq and will not be available until at least the weekend.
The statements from U.S. officials are also contradictory. According to CBS, some U.S. officials stated that there are currently no "new" or unusual negotiation plans, only continuations of existing talks conducted by the U.S.-Iran negotiation team through mediators.
The impact on BTC will depend on whether the strait can reopen.
After Trump announced on August 2 that he would pause the strike and negotiate, BTC briefly rebounded from below $63,000 to around $64,000. But after Iran's denial, market optimism about the "negotiations" quickly faded. As of August 4, BTC was fluctuating narrowly between $63,300 and $64,000.
In the short term, if the Strait of Hormuz truly reopens, the decline in oil prices will ease inflation concerns, ease the pressure on Fed rate hikes, and be positive for BTC. However, the strait remains closed to this day. Iran has clearly stated that even if an agreement is reached with Oman, as long as the U.S. does not lift the blockade or resume fulfilling the 14-point memorandum of understanding, the strait will not reopen. There are also fundamental differences between the two sides regarding fees.
This "negotiation" is more like a battle of public opinion than a genuine diplomatic breakthrough. With both sides unable to even agree on whether to negotiate, it is difficult for the market to consistently price a "peaceful premium." BTC will continue to swing between geopolitical uncertainty and macro policy in the short term, with the direction depending on whether the strait truly reopens, rather than any side's verbal statements.Palantir's second-quarter revenue reached $1.94 billion, up 93% year-over-year, with adjusted earnings per share of $0.41, both above market expectations. More importantly, the company sharply raised its full-year 2026 revenue guidance from about $7.65 billion to around $8.15 billion, with the third-quarter revenue guidance also significantly exceeding analysts' expectations.
To be honest, the market has been asking whether AI can truly make money, and Palantir has given a solid answer: customers aren't just trying it out—they're actually signing contracts, paying, and contributing cash flow.
This is good news for AI software, cloud computing, and even the upcoming demand for computing power and networking.
But with such high valuations, we can't just look at the story going forward; we must maintain this exaggerated growth rate.
#Palantir $PLTR #AI #美股 #NVDAStrategy 再次卖出比特币，长期“只买不卖”的叙事正在明显改变。
Strategy 最新向美国 SEC 披露，7月27日至8月2日期间卖出1,638枚比特币，套现约1.047亿美元，平均卖价约63,957美元。卖出后，公司仍持有842,138枚BTC，但平均持仓成本约75,419美元，高于当前比特币价格。公司同期还回购了约8,120万美元的STRC优先股，并继续增加美元储备。
数量相对84万枚总持仓不算多，但意义挺大：以前大家默认Saylor只会买，现在他已经开始卖币补现金、付股息、回购优先股了。
说实话，我觉得最危险的不是这次卖了多少，而是市场以后每次BTC下跌，都会开始猜他还要卖多少。
比特币最大的企业买家，正在从“永久买盘”变成可能出现的阶段性卖盘。
$MSTR $BTC With 72 hours left, the CLARITY Act still hasn't been implemented—most likely, it's over
Monday, August 3rd.
The Senate announced this week's agenda.
The CLARITY Act is not on top.
In its place is the H.R.6500—a persistent resolution with absolutely no connection to encryption.
On August 7, the Senate adjourned.
Today, August 4th. There are less than 72 hours left until the deadline to submit a motion to terminate the argument.
But the problem is—
They haven't even reached the starting line.
You think 72 hours is long?
Let me break down the Senate's inhumane procedures:
The first hurdle: submitting a cloture (termination of debate) application, which requires signatures from 16 senators.
16 people. Sounds like nothing?
But this is a request to "end the debate"—meaning you have to convince 16 senators to agree "enough arguing, just vote directly."
Second hurdle: Cloture voting, requires 60 votes.
100 senators, three-fifths. Even if the Republicans unanimously support it—53 seats—they would still need to bring in at least seven Democrats.
Seven people.
These seven people have been stuck on the bill for two months.
The third hurdle: even if the cloture passes, you can still debate for another 30 hours.
30 hours. Only after the debate is done that a vote on "whether to enter the review" is "not to vote on the bill itself."
Even if the green light is clear, the cloture vote will take place as early as Friday (August 7).
And then? The meeting is adjourned.
The bill itself? The interrogation hasn't even begun.
Do you now understand where the problem lies?
Wednesday (August 5) is the last chance to submit your Cloture application. Thursday? It was too late. Friday? The meeting is adjourned.
So—if 16 people don't sign by Wednesday, the bill doesn't even have a chance to be "vetoed."
Directly stillborn.
What's even more heartbreaking is the vote count.
With 53 Republican seats, even with full support, it still needs 7 Democrats.
But what is the Democratic Party arguing about?
Ethical clauses.
Specifically: the Trump family earned $1.2 billion from crypto, and the Democrats said, "You made too much, we don't believe in this bill."
The Republican Party itself is also unclean — Josh Hawley and Rand Paul have publicly opposed it.
The Republicans can't even gather enough votes themselves, so how can they expect to win the Democrats?
Both sides are acting.
Who was injured? For those holding positions. How does the market view it?
On Polymarket, the probability of the bill passing in 2026 is 28%.
At the beginning of the year, this figure was 74%.
Galaxy Research — cut directly from 50% to 30%.
Bernstein warns: If the bill is delayed, it will trigger a "stress-based" sell-off in the crypto market.
BTC fluctuated around $63,500 today.
But don't be fooled by the numbers—trading volume plummeted by 71%.
Duokong dared not move.
Everyone is waiting for the same signal: is there still any chance for the bill?
To put it bluntly—
Last week, Grayscale was also urging the Senate to vote quickly.
Finance Minister Besent made a public statement.
Industry representatives have contacted Congress over one million times. Is it useful?
As soon as Monday's agenda came out—there was nothing.
A 616-page bill, stuck in just a few words.
Even if the procedural vote starts this week, it will only leave a "measure" for the resumption in September.
Reunion in September? There are still two months left until the midterm elections.
By then, who will still be able to control your crypto bill?
If you still have a large amount of altcoins, bet on a wave of price surges after the bill passes—
You're betting on something that is 72% unlikely to happen.
Before August 5th, if no 16 people have signed—
This bill ends this year.
Jiang Zhuoer said that if the CLARITY Act cannot be passed before the recess, Bitcoin may complete its final bearish dip. The chip concentration in the $62,000–$63,000 range is as high as 8%, similar to the eve of the FTX collapse in 2022.
High concentration + negative news reallocating = violent redistribution.
You're not holding on.
You are waiting to be assigned.US Semiconductor Stocks Rebound Strongly: Micron and SanDisk Surge—Has a New AI Chip Market Already Started?
Yesterday, an important signal appeared in the US stock market:
After a significant adjustment in the semiconductor sector in July, memory chips saw a strong rebound, with Micron (MU), SanDisk (SNDK), and other related stocks performing outstandingly.
The market began to focus on one issue:
Has the July adjustment ended? Is the AI chip market restarting?
1. Why did semiconductors undergo such a sharp adjustment in July?
In recent times, the AI industry chain has become the strongest main theme in the US stock market.
Nvidia, Broadcom, semiconductor equipment, and storage-related companies all attracted massive capital pursuits, causing their stock prices to rise rapidly.
But after the rise, concerns began to emerge in the market:
* Has AI investment already priced in for future growth?
* Will data center demand slow down?
* Are semiconductor company valuations too high?
At the same time, the high interest rate environment has also reduced investors' short-term risk appetite for overvalued tech stocks.
Therefore, the July decline was more like a valuation and sentiment adjustment than a change in the logic of the AI industry.
2. Why are memory chips like Micron and SanDisk once again attracting capital attention?
The core reasons are:
The AI era requires not only computing chips but also massive high-speed storage.
In the past, the storage industry was considered a typical cyclical sector by the market.
But with the rapid development of artificial intelligence, the storage industry is undergoing changes.
AI servers require:
* HBM high-bandwidth memory;
* DDR5 high-speed memory;
* Large-scale data storage.
These demands are increasing the long-term growth potential of memory chip companies.
The market is beginning to re-examine it:
Future AI infrastructure construction will be not just a competition of GPUs, but also of storage capacity.
3. Why does the macro environment support a rebound in tech stocks?
From a global perspective, there are several key factors:
1. Investment in AI infrastructure continues
Major global economies are increasing their investment in artificial intelligence.
AI competition has evolved from simple software competition to:
Comprehensive competition in computing power, chips, power, and data centers.
This means the semiconductor industry may be entering a longer development cycle.
2. The global semiconductor supply chain is being restructured
Due to international competition and supply chain security needs, countries are strengthening their chip industry layouts.
This will drive continued investment in semiconductor manufacturing, equipment, materials, and storage industries.
3. The market has begun to re-trade rate cut expectations
If the interest rate environment gradually improves in the future, funds will typically flow back into growth technology assets.
As a core sector of the technology industry, semiconductors often become a focus for capital.
4. Does this surge mean the July correction is over?
At present, this may be a positive signal, but further confirmation is needed.
To determine whether the adjustment has truly ended, you can focus on several aspects:
First, whether leading semiconductor companies continue to strengthen.
Second, can core companies like Micron, Nvidia, and AMD form sector resonance?
Third, whether the semiconductor index has broken through key resistance levels.
If capital continues to flow back and industry fundamentals keep improving, then the July adjustment could be a healthy shakeout.
However, if the rise is only a short-term capital push, it could trigger another wave of volatility. #从降息到加息, Fed Divergences Fully Revealed: #财报观察员: AMD and SpaceX Close to Close, Circle Closes #Palantir营收增93%, Gains 13% in After-Hours Guys, WAXP dropped 7.17% today, now at $0.003943. Compared to yesterday's rapid rally, the current pullback leans more toward a receding sentiment, but for now, we cannot directly rule out a subsequent evolution into a trend reversal; it depends on the defensive outcome of the support.
Yesterday's WAXP surge did not come from substantial ecosystem positive developments, but rather a reflexive rally driven by three overlapping forces: millions of dollars in spot buying in a low-liquidity environment drove the gains; Extreme negative funding rates have led to short covering and short squeezes; After the price surged, the Square platform narrative spread, old stories were reheated, and community discussions exploded.
The shortcoming of this type of market is obvious: once there is no sustained incremental capital to take over, the rise is fast, and the tide retreats just as fast. Today's correction is a manifestation of this structure.
Technically, prices have fallen below the short-term MA7 and MA25 moving averages, which are under pressure; Trading volume failed to sustain and expand, and the strong bullish momentum from the previous wave is fading. The previously driven short squeeze conditions quickly disappeared as prices retreated.
Whether another rally can emerge depends on two factors: whether trading volume can increase again after a pullback, and whether the price can hold at key support levels.
If the market enters a continuous shrinking volume decline, then confirm that the past two days were just a one-time capital pulse rally.
Personal market view analysis and market information compilation, not investment advice.
$BTC $ETH $WAXP
#从降息到加息, the Fed's disagreements are fully public
#财报观察员: AMD and SpaceX are about to hand over, with Circle as the grand finale
#Palantir营收增93%, up 13% in after-hours trading. Crypto investment firm The Spartan Group just liquidated three investment tokens worth $3.92 million:
2.045 million PENDLE ($2.74 million) + 5.377 million SYRUP ($860,000) + 813,000 ETHFI ($320,000) were transferred into Coinbase Prime.
These three are all their investment projects, and tokens are acquired through investment unlocking ownership.
Address: 0x0DC874Fb5260Bd8749e6e98fd95d161b7605774D$BTC long-bear analysis on August 4
Bias: Mostly oscillating
Bullish probability: 60%
Bear Probability: 40%
Key location
First support: 63,500
Second support: 63,000
Strong support: 62,500
First pressure: 64,300
Second pressure: 64,800
Strong pressure: 65,500
Trading strategy
Long Approach (Priority)
If stabilizing near 63,450~63,600, consider going long in batches.
Stop loss: below 63,180.
Take profit:
First target: 64,300
Second target: 64,800
Strong potential target at 65,500
Empty thinking
If the rally near 64,700~64,900 is blocked by increased volume, you can try a light position to short.
Stop loss: 65,150.
Take profit:
First target: 64,000
Second target: 63,500
Monitor all day
Holding above 64,300: Bulls have a chance to further challenge 65,000~65,500.
Below 63,500: Bears may test 63,000 or even 62,500.
Most of the time, the expected operating range is: 63,500~64,800.
⚡️ Daily market analysis 📈: bloggers publish short-term resistance and support levels daily. Since everyone's trading system and mindset are different, group members refer to resistance and support levels, combined with real-time geopolitical and economic policies, and choose the method that suits them best. From a cyclical perspective, yes, a few platforms that truly have users, cash flow, and network effects will continue to radiate vitality.
The current challenges facing cryptocurrencies—regulation, compliance, and AI-driven vulnerability detection—have made attacks more convenient, and security and trust have become common issues this year. Currently, it is happening on-chain, and in the future, we should see thefts in exchange-custodial assets.
Regulatory and compliance issues will drive funds to flow into projects with greater long-term value, rather than some tail-end fund schemes.
Especially after the passage of the U.S. Clear Act, tail assets should be quickly cleared out, with leading assets taking on institutional liquidity, which may come from Wall Street, Silicon Valley, AI companies, and others
The excessive profits on exchanges will also end within 2~3 years, returning to the level of a normal asset custody platform.
I remember when Da Bing was 30,000, I wrote that the survival space for crypto-native main players would shrink more and more.Nvidia's next-generation AI chip may reduce HBM configuration—can memory stocks still buy the dip?
Many people have recently bottom-fished storage stocks, betting on an industry reversal after an oversold period.
But it's important to note that what you buy may just be a price correction, not necessarily a new round of accelerated profits.
Over the past year, the biggest growth logic in the storage sector was not the recovery in demand for phones and PCs, but AI.
The original market script was very simple:
AI server demand surges → NVIDIA GPUs continue shipments → each GPU is equipped with more HBM → HBM prices rise→ SK Hynix, Micron, and Samsung all benefit.
This logic drove a major rally across the entire storage sector.
US Stock Investment Network Now, this script is undergoing new changes.
According to SemiAnalysis's supply chain information, some SKUs of Nvidia's next-generation Rubin Ultra platform are being previewed to key customers.
One change the market is focusing on is:
The previously highly anticipated 12-layer HBM solution may have room for adjustment, with some versions possibly adopting an 8-layer HBM configuration.
According to current rumors, the HBM capacity per GPU may be reduced from 288GB to 192GB, a reduction of about 33%.
But here you need to note:
This does not mean that Nvidia's AI demand has declined, nor that Rubin Ultra's overall performance has been significantly weakened.
What truly changed was the market's expectations for a certain question:
How much high-end storage will each AI chip actually consume in the future?
In the past, investors defaulted:
The more GPUs sold, the more HBM each GPU can hold, naturally leading to rapid revenue growth for storage manufacturers.
But now, the AI chip competition is entering a new phase.
NVIDIA is not just concerned with how much HBM a single GPU stacks, but rather on the overall efficiency of the AI system.
For customers, what they ultimately purchase is not memory capacity, but lower training costs, lower inference costs, and the computational efficiency corresponding to each token.
If system architecture optimization reduces the HBM configuration per GPU while meeting performance requirements, it may actually be more attractive to NVIDIA and cloud providers.
The revenue logic of storage vendors can actually be simply broken down as:
GPU shipments × HBM capacity per GPU × HBM price.
In the past, the market believed that all three variables would rise simultaneously.
But now, the first variable may continue to grow, the third still has demand support, but the second variable has the potential to decline.
Assuming HBM prices remain unchanged, if the HBM capacity per GPU drops from 288GB to 192GB, GPU shipments would need to grow by about 50% to fully offset the impact of the drop in unit capacity.
This is also what storage stocks really need to pay attention to.
The risk is not the disappearance of AI demand.
Instead:
AI demand continues to grow, but the growth rate and the value per chip may not be as high as the market previously imagined.
U.S. Stock Investment Network believes this news is not simply a bearish outlook on SK Hynix, Micron, and Samsung.
HBM remains a crucial component of AI infrastructure, and demand will continue to grow in the coming years.
But the investment logic is changing.
In the past, when buying storage, I looked at:
The more AI gets popular, the more profitable all storage manufacturers are.
When buying storage in the future, what you look at is:
Whoever can secure more high-value orders and maintain higher profit margins.
SK Hynix's greatest current advantage is its HBM customer relationships and mass production experience.
Micron needs to prove its scale and profitability in the HBM4 era.
Samsung needs to prove whether its high-end HBM products can regain more customer recognition and expand its market share.
The future market will not reward all storage companies equally.
What truly determines valuation are several key issues:
What are the final production specifications of the Rubin Ultra?
Is the high-capacity HBM version still mainstream?
Can HBM4 contract prices be maintained?
Can the overall scale growth of AI servers offset the decline in HBM capacity per chip?
If the number of AI servers grows quickly enough in the future and system-level HBM demand may continue to rise, current concerns will be alleviated.
However, if single-card HBM capacity declines and the upside for HBM prices weakens, then the recent rise in storage stocks may come more from valuation recovery rather than a sharp increase in earnings expectations.
The AI story is not over.
But storage stocks can no longer be simply bought according to the logic of the past two years.
The real competition in the next stage is not who stands at the AI front, but who can maintain their storage value as AI systems continue to be optimized.Fundamental Research Report $METIS / Metis (L2/Sidechain) $3.20
Conclusion First: Metis ($METIS) overall score 59/100, rated with narrative over implementation. Looking at the three layers, the company team has cash reserves, the protocol network already shows signs of paid usage, and token value transfer still needs to be observed.
Fundamental breakdown: Metis (token $METIS), L2/sidechain track. Featuring Optimistic L2+DAC. Benchmarked against ARB and OP. Traditional collaboration between enterprises relies on cloud servers and contract reconciliation, which causes gas surges, TPS constraints, and frequent cross-chain bridge security incidents during high concurrency. Public chains use a unified state machine for trustless settlement, reducing reconciliation costs. Average order value is $50-500/month, with settlement required in USDC or fiat currency. Narrative-driven tracks, bear market usage cut by 60-80%. Positioning the end-to-end vertical platform. Product implementation: The protocol layer is officially operational, and the on-chain dashboard shows protocol fees accumulating, showing signs of paid usage. Latest version not found, 60 valid submissions in the past 90 days.
At the user level, address MAU not disclosed, DAU not disclosed, 24-hour transaction volume $80.00M, TVL not found. Wallet addresses do not equal monthly active users of natural persons; large large addresses holding concentrated positions tend to overestimate the actual number of users. On the revenue side, user fees are not disclosed. Supply-side revenue is about 80-90% of user fees (to LPs and nodes), protocol treasury revenue is $10.6K, token holders buy back and burn at an annualized rate, with no burn mechanism. 24-hour transaction volume is business turnover, not revenue. A company making money does not mean the protocol makes money, and protocol profits do not equal token holders making money. On the code side, 60 valid submissions in 90 days, 25 active contributors, latest version not found. GitHub is a Class A evidence that can be directly verified. Investment background: For company equity financing, look to PitchBook/Crunchbase (A-level); for token private and public funding, use whitepapers, release curves, and on-chain unlocked contracts (A-level); market makers and ecosystem funding are B-level and do not represent long-term holdings of tech VCs; for technical integration, look to API/SDK access evidence (B-level); strategic partnerships and logo walls are D-level. The use of NVIDIA GPUs does not equate to NVIDIA investment, and going public on exchanges does not equal strategic investment.
On the token side, total supply is 1,300,000,000, circulating 950,000,000 (73.1%), FDV $4.20B, next unlock is 2026-Q4 (+3.50% circulating), burn buyback annualized rate, no explicit buyback burn. Do you have to buy coins to use the product? Yes, strong value capture (Gas/Collateral/Service Access). Let's look together with peers (unified criteria, no cross-sector random comparisons): In terms of circulating market cap, Metis $3.00B, ARB undisclosed, OP undisclosed. For FDV, Metis $4.20B, ARB undisclosed, OP undisclosed. In terms of annualized revenue, Metis is $10.6K, ARB undisclosed, OP undisclosed. Regarding monthly active addresses or users, Metis has not disclosed this, ARB has not disclosed, and OP has not disclosed. Figures are based on public data snapshots; any omissions are supplemented by official self-reports or industry standards. Valuation, circulating market cap $3.00B, FDV $4.20B, P/S 283,419.9x, FDV divided by revenue 396,787.9x. Pessimistic outlook: $3.00B at 50-70% off, oscillating within a neutral range; optimistic outlook: revenue doubling, burns landing, enterprise clients coming in, FDV corresponding to P/S, aligning with the top companies. To wrap up: solid fundamentals (rating 59/100). The token value transmission path is unclear, with only governance incentives. Circulating market cap is relatively high relative to fundamentals, overdrawing expectations, and FDV is moderate. Potential pitfalls: short-term large-scale unlock and sell-off, long-term protocol revenue reverting to zero, token demand relying solely on incentives (once incentives are cut off, usage collapses). Ongoing monitoring: protocol fee cycles, burn amounts, active address retention, TVL/loan balances, GitHub version releases. The above judgment is based on publicly available data and does not constitute any investment advice. Conclusions need to be revised when key indicators deviate significantly.
That's all for the content—judge for yourself.
#基本面研报 #加密 #研究 #OKXOrbitETHUSDT (Market Linked Short Position)
- Admission period: 1955~1980
- Stop-loss level: 2005
- Take-profit tier 1: 1900
- Take profit in tier 2: 1850
BTCUSDT (Core Short Position on the Main Market)
- Entry section: 66,500~67,000
- Stop-loss level: 67,700
- Take-profit tier 1: 65,000
- Take-profit second tier: 63,800
SOLUSDT (Public Chain Leader Rebounds Short)
- Entry Section: 77.5~78.8
- Stop-loss level: 80.5
- Take-profit tier 1: 73.0
- Take-profit tier 2: 70.2
ZECUSDT (Privacy Coin Under Pressure)
- Entry section: 505~512
- Stop-loss level: 525
- Take-profit tier 1: 480
- Take-profit second tier: 462
XAUUSDT (Gold Commodity Under Pressure)
- Entry section: 4220~4260
- Stop-loss level: 4310
- Take-profit tier 1: 4050
- Take profit second tier: 3960
LABUSDT (Oversold Rebound Short Position, High Priority)
- Entry range: 0.150~0.156
- Stop-loss level: 0.164
- Take-profit tier 1: 0.136
- Take profit tier 2: 0.125ETHUSDT (Market Linked Short Position)
- Admission period: 1955~1980
- Stop-loss level: 2005
- Take-profit tier 1: 1900
- Take profit in tier 2: 1850
BTCUSDT (Core Short Position on the Main Market)
- Entry section: 66,500~67,000
- Stop-loss level: 67,700
- Take-profit tier 1: 65,000
- Take-profit second tier: 63,800
SOLUSDT (Public Chain Leader Rebounds Short)
- Entry Section: 77.5~78.8
- Stop-loss level: 80.5
- Take-profit tier 1: 73.0
- Take-profit tier 2: 70.2
ZECUSDT (Privacy Coin Under Pressure)
- Entry section: 505~512
- Stop-loss level: 525
- Take-profit tier 1: 480
- Take-profit second tier: 462
SOXSUSDT (Inverse Index Under Bear Pressure)
- Entry Range: 53.2~54.2
- Stop-loss level: 55.8
- Take-profit tier 1: 48.5
- Take-profit tier 2: 46.2
SKHYUSDT (SKHYNIX ADR Going High)
- Entry range: 152.5~155.0
- Stop-loss level: 158.5
- Take-profit tier 1: 142.0
- Take-profit second tier: 137.5
6. Technology Blue Chips and Commodity Stocks
UBUSDT (Small and Mid-Cap Bear Pressure)
- Entry Range: 0.198~0.203
- Stop-loss level: 0.209
- Take profit level 1: 0.182
- Take profit second tier: 0.173🚨 Tomorrow might be the biggest crypto day this month.
Charts matter, but Washington might move them more.
The U.S. Senate is set to vote again on the CLARITY Act — the bill that would finally draw clear rules for digital assets. Talks are still ongoing, so the outcome is anyone’s guess.
Why it matters:
🟢 If it passes: More clarity = more institutional trust. More capital. Better sentiment for $BTC, $ETH, and quality alts.
🔴 If it stalls: Uncertainty sticks. Volatility picks up. Risk assets stay under pressure.
With $BTC sitting on a key level, this isn’t just politics. It’s about flows and confidence.
Crypto isn’t driven by TA alone anymore. Regulation, institutions, and macro are running the show just as much.
👀 Breakthrough or more limbo? Drop your take ⬇️
$BTC $ETH $SOL #DailyOrbit
#BigTechEarningsWatch
#FedSplitGoesPublic The 30-year yield fell from 5.27% on Friday to 5.23% on Monday, just 4 basis points away. Some in the comments section are already shouting "peak."
I'll hold back for now.
For $TLT, the decline in yields is good news, but two days isn't enough to issue a graduation certificate for long-term bonds. If the 30-year term continues to fall below 5.15%, then trend relaxation will follow; If it returns above 5.27%, that breath will be for nothing.
Long-term bonds are best at using a slight rebound to trick away patience. Slow down. This does not constitute investment advice.
#30年期美债, the top or a new beginning? #美债 #TLT #利率观察At dinner at dinner, the most common question you hear is: Why is AI trending again? This time, the excitement wasn't about a new app or someone releasing another model, but about a sum of money. Castle Securities expects that by 2028, tech companies will also raise over $500 billion in debt financing in both public and private markets to fund chips needed for AI park construction. On the surface, this incident affects tech companies, the credit market, and the chip supply chain, but on the deeper level, it also impacts everyone currently using AI products. Faster replies on your phone, smoother tools in the office, or whether your company is willing to continue buying AI services don't come out of thin air; ultimately, it all comes down to computing power, chips, campuses, and money. What readers should most verify now is not how a company's stock price will move, nor whose story is louder, but three questions: Is this money continuing to flow into AI infrastructure? Can chip demand really be sustained? As debt financing grows larger, who can absorb the costs, and who just rides along with the excitement? [The AI bill is already on the table] In the past, when people talked about AI, it was easy to keep the conversation light. A chat window, a video, a self-generated PPT—it all seems like things happening on the screen. Click once, and the answer will appear. In the office, some people use it to revise copy, some to write code, some to make spreadsheets—it's like having a colleague who's always on call. But behind the screen is no magic. For AI to run, train, deploy, and enable more people to use it simultaneously, chips and data centers are needed#MSTR再卖1638枚比特币，规模腰斩
From July 27 to August 2, Strategy sold 1,638 bitcoins at an average price of $63,957 each. Compared to its overall holding cost of $75,419 per bitcoin, this represents a loss of about $11,462 per coin. The proceeds of approximately $52.4 million were used to pay preferred stock dividends, and the remaining roughly $52.3 million was used to repurchase discounted Class A perpetual preferred shares. This is the company's second public reduction after a tentative sale of 32 bitcoins in May, and so far the largest "loss-making sell-off".
My view is that this is not a bearish signal. After the reduction, the company still holds about 842,138 bitcoins, accounting for approximately 4% of the global mined supply. Since the beginning of the year, it has cumulatively bought 174,895 bitcoins and sold only 3,620, resulting in a net increase of 97.9%. The "Bitcoin whale" status remains unchanged.
What is truly worth noting is the change in operational logic. In the past, Strategy was known for "only buying, never selling," and its founder Saylor even publicly declared he would personally never sell bitcoin. But this year, the company has clearly revised this stance: Saylor clarified that "never selling" is a personal belief, and the company can trade for capital management needs.
The pressure behind this comes from the liability side. Strategy previously financed its bitcoin accumulation by issuing preferred shares, but the Class A perpetual preferred shares have been trading at a discount, indicating the market is pricing in a risk premium on its credit. If the discount persists, it will not only increase future financing costs but may also affect the stability of the entire capital structure.
Therefore, the company chose to sell bitcoin to gain liquidity, prioritizing dividend payments and repurchasing discounted preferred shares. This is a typical capital structure optimization move—using asset liquidation to proactively repair liability pricing. The company has raised the authorized bitcoin sale limit from $1.25 billion to $5 billion, showing management's willingness to sacrifice some bitcoin exposure to stabilize preferred share prices.
Overall, Strategy's long-term strategy has not shifted. Bitcoin remains the absolute core asset on its balance sheet. This "loss-making sale" looks more like a tactical adjustment by a giant institution holding 4% of the global supply, responding to capital market regulation and financing pressures.
The key points to watch going forward are no longer the amount sold in a single transaction but two indicators: the progress of the STRC preferred shares' discount recovery and the correlation between bitcoin volatility and the company's financing costs. If the discount continues to widen, the company may continue selling bitcoin to intervene; if it recovers near par value, the pace of reduction will slow significantly. This balancing act of "hoarding bitcoin while managing debt" has only just begun. $SOL image shows data from the voting support phase of Solana's governance proposal "Double Disinflation" (SGP-0002). Below is an analysis of the current state and potential impact:
📜 Core of the proposal: Accelerating "deflation"
The proposal, proposed by Solana infrastructure company Helius, centers on modifying inflation parameters:
· Accelerating inflation decline: raising the "inflation decline rate" from -15% per year to -30%.
· Achieving the target ahead of schedule: The timeline for lowering the SOL inflation rate from the current approximately 3.8% to the final target of 1.5%, from 2032 to around 2029.
· Reducing SOL supply: It is expected that approximately 18.9 million new SOL issuances will be reduced over the next six years.
📊 Voting status: Support rate meets target, progress is smooth
· Current support rate: 39.97%, well above the 10% threshold.
· Required votes: 17.29M SOL supported, with 25.97M SOL remaining before the official vote can begin.
· Time and participation: About 14 days and 16 hours remain in the voting support phase, with only 5 validators (0.7%) currently participating.
⚖️ Potential impact on SOL
Supply and Inflation (Positive)
Reducing net new supply of SOL is a long-term positive factor. If combined with proposals to increase daily SOL burning (SIMD-0553), deflationary pressure may arise.
Staking Yield (Short-term Bearish)
Falling inflation directly lowers staking yields. The annualized yield on nominal staking is expected to drop from the current approximately 6.41% to 5.04% in the first year and to 2.42% in the third year.
validator (potential resistance)
Reduced yield may lead to opposition from some validators (especially smaller validators). The proposal model predicts that about 18 validators may turn from profitable to loss-making in the first year.
Price and sentiment (complex)
The market usually digests expectations in advance. Prices reacted when the proposal was proposed in June and when voting began in August. As of today (August 4), SOL's price is fluctuating around $73.5.
💎 Summary
The "Double Disinflation" proposal is currently progressing smoothly, reflecting the community's inclination toward optimizing tokenomics. If ultimately approved, it would provide structural support for SOL's long-term value, but in the short term, it may face resistance from some validators due to declining staking yields. After testing the 64,239 level at midday, Bitcoin quickly pulled back and is now near 635.
On the four-hour chart, the candlestick has already crossed our target level, but unfortunately, it hasn't stabilized. However, it is worth noting that the candlestick is still near the upper band of the Bollinger Band.
At the one-hour level, a three-day consecutive bearish pattern has formed, and the Bollinger Bands have formed a tightening pattern. The candlestick is about to touch the middle band of the Bollinger Bands, and a correction is currently underway. If it stabilizes, it will trigger our second rebound, giving friends who missed out a second chance to catch up. $BTC $ETH $KORU ，15.02 开多，+131.82%。
开仓逻辑不仅基于技术面，更叠加了宏观基本面的隐性驱动。近期该项目方释放了重大合作利好，且行业整体处于回暖期，基本面支撑价格上行。
15.02的开多点正是基于“有基本面支撑的价格更易持续”的判断。
没踩准的，学会将宏观基本面与技术面结合分析。等我下一个基本面与技术面共振的布局。$BTC $SNDK SOL discussions have sped up slightly, with a clear advantage in the upper hand position: the two most easily confused things in the popular rankings
SOL's community snapshots provide both heat and tone, but not necessarily on the same side.
OKX Onchain OS recorded 22 mentions of SOL in one hour at 13:00 (China time) on August 4, including 22 x mentions and 0 news articles; The total 24-hour volume was 466 times.
The latest hour is 1.13 times the long-window hourly average, which is about 13% higher than the 24-hour average, which can be considered a "slight acceleration." This speed describes new discussions and is not necessarily related to market fluctuations.
The tone of the text is bullish at 64%, bearish at 9%, and neutral at about 27%, currently indicating a clear bullish bias. 49% bullish in the 24-hour period, 9% bearish; If there is a gap between the two windows, it should first be understood as a change in the discussion structure, rather than directly deriving a price target.
I will draw these two lines separately. If the tone is too heavy but the speed of mention is slower, it means the current discussion is more positive, but the new attention hasn't accelerated; If mentions are rising and bearish are dominant, it may be risk or fault news attracting people. Even if the hype and tone are in the same direction, it still cannot be directly equated with genuine buying.
Source is another limitation. Currently, SOL is "almost entirely driven by X." Social channels respond fastest, and the same topic can be reposted repeatedly; The more concentrated the source, the more the next window needs confirmation. News mentions that an increase does not automatically mean the event is true; the original announcement remains the final verifying standard.
Within twenty-four hours, SOL's X and news mentions were 433 and 33 times respectively; One hour means 22 and 0 times. If short windows are more focused on X than long windows, sensitivity should be increased to forwarding and single narratives; If the proportion of news increases, also check whether the same material is actually being restated.
What really matters are SOL's on-chain transaction success rate, fees, active addresses, and main application usage, combined with spot trading, perpetual contract funding rates, and open interest. These data answer usage needs and leverage participation, and popular rankings cannot replace them.
Time differences also need to be watched for. The 466 24-hour samples span different market periods; dividing by 24 is just for comparison convenience and does not mean the same volume of discussion every hour. A single deviation from the mean should be observed first, not as a trend completion.
How can you tell that it was just noise? The next round of mentions increased, but the tone quickly returned to neutral. This time, the sense of direction was mostly due to a small sample size. If the speed of mentions continues to rise and the sources expand from a single community, attention will gradually stabilize. In the end, what can change judgments is still continuous data, not a louder slogan.
Let's note three things for now: SOL discussions have slightly accelerated, the Short Window tone is clearly dominant, and it's almost entirely driven by X. If the speed continues and the sources become more diverse, and transactions and on-chain data also echo, this observation can be pushed forward further; Before that, put them on the watch list and don't rush to run.$SNXX ，20倍杠杆，+395.65%。
持仓期间我严格优化了盈亏比。该笔交易预设止损在8.8（约4.5%风险），目标看向12.0（约30%收益），盈亏比接近7:1。
高倍持仓必须配合极高的盈亏比，才能在承受波动的同时获得超额回报。我选择在突破后回踩确认时入场，提高了胜率与盈亏比的平衡点。
没跟上的，重新计算你的盈亏比与仓位匹配度。等我下一个盈亏比极致的实战案例。$BTC $ETH Let's talk about the Federal Reserve. To put it bluntly, half a year ago there was debate over whether to cut rates, but now it's a direct debate over whether to raise rates, and the differences are laid out for everyone to see. The July 29 rate meeting voted 9 to 3, keeping the range unchanged from 3.5% to 3.75%. However, three regional Fed presidents—Hamack of Cleveland, Kashkari of Minneapolis, and Logan of Dallas—all voted against it, arguing that a 25 basis point rate hike should be introduced.
Among these three, Kashkari, who had long been labeled by the market as a dovish, now sided with the rate hike camp, indicating that this is not a handful of hawkish faces opposing the issue, but rather that the internal political landscape of the entire committee is truly loosening.
The new chairman Walsh's statement this time was also interesting. When asked how he responded to the three votes against it, he said he had called for a fierce internal family debate, and he indeed got it. He has said this thirteen times in his last five public appearances, basically treating the disagreement publicly as a narrative strategy rather than covering it up. Before the meeting, the market's implied probability of a rate hike once surged to 34%. This figure is unusual before a meeting that remains unchanged, indicating that the market itself has sensed a shift in the wind direction.
Looking back, the June dot plot already included a 25 basis point rate hike within the year. Board member Waller has publicly stated that if inflation does not improve further, rate hikes may be necessary. The September meeting is now widely seen as a key window for whether to continue raising rates or continuing to wait and see, depending on whether the Middle East situation can cool down and the next two CPI data points.
The logic behind this is actually not complicated. Inflation has failed to return to the 2% target for more than five consecutive years, and combined with tariffs and energy costs driven by the Iran conflict, these supply-side shocks should be temporary by textbook logic, but now they have persisted enough that some officials feel inflation expectations will derail if they wait any longer.
My own judgment is that this public disagreement is not a random communication lapse but rather a deliberate effort to help the market absorb the possibility of a rate hike early. Whether or not it will be raised in September, the Fed's policy focus has clearly shifted from how to continue easing to whether to tighten again. #从降息到加息, the Fed's full disclosure of #从降息到加息 of divides, the #从降息到加息 of the Fed's full disclosure of the $BTC $ETH $SKHYNIX #亚马逊市值破3万亿, a 50 billion bet to win the first round
Amazon's market capitalization closed above $3.05 trillion, closing at $284.02, officially joining the "$3 trillion club" and becoming the fifth company globally—ahead of Apple, Microsoft, Alphabet, and Nvidia.
But this wave wasn't driven by sentiment speculation; it was the financial reports that pried open the valuation door.
Q2 revenue was $200.6 billion (+19.6% year-on-year), AWS revenue +37% year-on-year, marking the highest growth rate in 18 quarters. Most AI computing power contracts are locked for more than five years, and the company believes the supply gap can be delayed until 2028. After the earnings report, it rose about 4.5%–5.7% in a single day, reaching an intraday high of 287.16, marking the strongest bullish candlestick since 2012.
On the capital side, the 'first move' claim is also confirmed: before the earnings report, institutions have already pushed their positions up, and Wall Street's consensus target price still leaves a double-digit margin above the current price; The dynamic P/E ratio is just over 22 times, below its own ten-year average, so "3 trillion" looks alarming. Discounted cloud + retail cash flow is actually not yet in the bubble zone.
The only minor issue is Bezos's plan to reduce his holdings by 15 million shares, about $4.1 billion. But this is a 10B5-1 plan filed in November 2025, well before this round of surges. It's a personal asset allocation and doesn't touch the company's fundamentals, so it can't suppress the bulls.
With 30-year US Treasuries at 5.23%, Amazon still being priced as "undervalued repair" shows the market values not cheap money, but AWS's ability to rent out AI infrastructure—$3.05 trillion is more like a revaluation starting point, not the end of an awards ceremony. #从降息到加息, the Fed's disagreements are fully public
Damn! The Fed has now completely torn off all facades. At the end of July meeting, the rate was firmly fixed at 3.50%-3.75%, but the vote was split 9 to 3.
The three regional chairs—Hamak, Kashkari, and Logan—have directly stepped forward to raise rates by 25 basis points, with only one reason: inflation hasn't returned to 2% in five years, the energy shock is still fueling the flames, and current rates are simply not harsh enough.
Chairman Kevin Walsh said, "What I want is a proper internal argument within the family," while throwing all forward-looking guidance into the trash, essentially telling the market: don't expect me to provide a roadmap; data speaks for itself. When necessary, I will act without hesitation.
As a result, the bond market tightened for them first, with the 10-year yield surging and the 30-year yield breaking through key psychological levels. Institutions were in complete chaos—some were still hoping for rate cuts, while others had already included one or even multiple rate hikes within the year in their reports.
This isn't just an ordinary disagreement—it's a directional-level internal conflict. The fantasy of rate cuts was crushed on the spot, and raising rates went from 'impossible' to 'could happen at any time.' The crypto market is still fantasizing about a 'liquidity spring,' but it's just self-deception.
The KOL on X also analyzed very thoroughly: this is a typical hawkish faction, holding its ground. The market achieved the expected flatness but did not receive the desired dovish signal. The bulls deserved to have hundreds of millions of dollars liquidated.
Others say BTC is hovering around 63,000, ETFs continue to flow out, and leveraged positions remain high. This wave of uncertainty is the biggest negative factor. The scenario of interest rate hikes and continued wait-and-see is hanging on the wall simultaneously, leaving funds to hedge for now.
Others argue that those still pricing in a rate-cut bull market are just catching the knife on the left side. If it really hits again in September, the negative news might be the next trend entry point, but only if you live until then.
For BTC, don't expect a one-sided rally in the short term. Volatility will wash out high-leverage players to the point of losing everything, and the ups and downs will make you question your life. The medium term is even less simple.
If it ultimately leads to rate hikes and liquidity tightens again, the altcoin season will be postponed until when the monkey is born; If data unexpectedly weakens and forces a turnaround, recession fears must be traded first, then liquidity easing will be discussed.
Don't tell me that 'whether you add or down, it's good for BTC'—that's just hindsight nonsense.
The Fed's own people are sabotaging each other, so what gives you the right to think you can bet on the other side precisely? Spend less time watching the market these past two weeks, sleep more. Waiting for the boot to really fall before picking up the benefits is a hundred times better than being cannon fodder in their internal strife.Everyone sees that old Zhihu post “Bought $BTC for 6000 yuan” and thinks: what if I’d bought then? Would I be free by now? 🤔
But read the comments. You’re seeing the story with 2026 hindsight.
If you actually went back to 2013-2016, it wasn’t a straight line up. It was chaos.
2013: policy panic, cliff dive.
2014: forced capitulation.
2015: bear market so long even the maxis broke.
2016: Bitfinex got hacked. Exchanges felt unsafe. 😰
The real question isn’t “why didn’t I buy?”
It’s “could I have held through all that?”
Hindsight makes it look easy: buy, lose the password, get rich. Reality? Fear, FUD, friends telling you to quit, 3am crashes making you question everything. 🌪️
That’s the lesson of “crypto archaeology.” The rare skill wasn’t the entry. It was conviction, risk management, and emotional stamina when nothing made sense. 🛡️
Did I miss $BTC? Yeah. But if I wasn’t built for that volatility then, buying pizza and living normal wasn’t wrong either. 🍕
The hardest part of investing isn’t reading history.
It’s holding when history hasn’t been written yet. ⏳
#DailyOrbit #FedSplitGoesPublic
#BigTechEarningsWatch $BTC From rate cuts to rate hikes, divisions within the Federal Reserve are becoming public.
In the past, the market has been waiting for an answer:
When will the Federal Reserve start cutting interest rates?
But now, the market is seeing a different signal—
Within the Federal Reserve, increasingly obvious divisions are emerging regarding the future policy path.
I believe the biggest change right now isn't whether the Fed will raise rates soon, but rather the market's beginning to realize:
The future interest rate path may not be as simple as previously expected.
In the past period, the market mainly traded the following:
Inflation falls→ The Fed shifts to → rate cuts.
But the reality is:
Although inflation has eased somewhat, it is still some distance from the 2% target;
Although the job market has cooled, economic resilience remains;
Consumption has not clearly collapsed, indicating that the high interest rate environment can still be sustained economically.
This has led to differing voices within the Federal Reserve.
Some officials believe:
If inflation continues to improve, policies need to shift gradually to avoid prolonged high interest rates putting excessive pressure on the economy.
Another group of officials is concerned:
Recurring inflation, energy price fluctuations, and fiscal pressures could cause premature rate cuts and lead to inflation heating up again.
So the market is now focused on more than just "whether interest rates will be cut."
Instead:
Within the Federal Reserve, who will prevail in judgment about the future direction of the economy?
For asset markets, such divergences can lead to greater volatility.
If the market resumes trading with "higher interest rates and longer duration," the dollar and Treasury yields could strengthen, putting pressure on risk assets.
If data continues to support a decline in inflation and expectations for rate cuts heat up again, capital may flow back into stocks and crypto assets.
The same applies to the BTC market.
Short-term price fluctuations often stem from changes in expectations.
What truly determines medium- to long-term trends remains:
Has liquidity improved?
Whether institutional funds continue to flow in.
Does the macro environment support the expansion of risk assets?
What the market fears most is not interest rate hikes or cuts, but uncertainty. When significant divisions begin to emerge within the Federal Reserve, asset prices enter a repricing phase. Next, what really matters is not who speaks, but whether future data supports their judgment. $BTC #从降息到加息, the Fed's disagreements are fully revealed #财报观察员：AMD and SpaceX Reports Are Imminent, Circle Takes the Spotlight
The real test of earnings season has never been about how much money is made, but whether the market is still willing to believe your next story.
Palantir has already given the answer: 93% revenue growth, combined with an upward revision of the full-year guidance, led to a 12% after-hours surge, indicating that what the market values most now is not just impressive data, but whether a company can continue to raise expectations.
So tonight’s AMD and SpaceX earnings reports, I believe, are not about the numbers themselves.
AMD is expected by the market to report $11.3 billion in revenue, a 47% year-over-year increase. This result has actually been priced in by the market already. What will truly determine the stock price direction is whether AI chip demand can continue to accelerate and whether the gross margin can prove that the AI business is not just a short-term bonus.
My judgment is that AMD’s biggest risk is not that the performance is insufficient, but "how good it is." At this stage of the AI rally, investors are no longer satisfied with growth alone; they need to see stronger orders, higher profit margins, and certainty of continued delivery over the next few quarters.
SpaceX follows a completely different logic.
Securing a $1.6 billion military contract proves its commercial capability is solid, but the large-scale unlocking starting August 6 is the real stress test. The market will focus not only on revenue but also on whether early investors are willing to continue holding.
This is also a common issue faced by many high-valuation assets this year:
The company hasn’t deteriorated, but valuations need to be repriced.
As for Circle, I actually think it is the most variable to watch among the three companies.
USDC’s reserve size has dropped to around $72 billion, indicating a change in the stablecoin market growth rate. On the other hand, if interest rates remain high, reserve yields can still support profitability.
So Circle’s real issue is not how much it makes now, but whether the stablecoin scale can continue to expand in the future.
My preference leans more toward watching AMD.
The reason is simple: AI remains the most important sentiment anchor for risk assets this year. If AMD can provide AI demand guidance that exceeds expectations, it will not only impact chip stocks but may also reignite the market’s risk appetite for AI-related crypto assets.
But if the earnings are good and the guidance is average, the market may continue to trade on concerns about the "AI investment return cycle."
The biggest opportunity this earnings season is not chasing the rally after numbers are released, but understanding what capital truly cares about before the market reprices.
Tonight’s AMD guidance may be more important than the revenue number.
$XAMD $XSPCX $XCRCL Palantir has taught us one thing: the financial report numbers don't matter; the guidance for next quarter is what matters
Palantir rose nearly 15% in after-hours trading.
Revenue grew by 93%, EPS beat expectations, and the full-year guidance was raised by nearly $500 million. The CEO bluntly stated—"Strong growth can last at least another 18 months."
Is it beautiful? Beautiful.
But if you really apply Palantir's formula of "earnings beating expectations = surge," you'll lose so much that you won't even lose your underwear.
Why?
Because tech giants have already proven that numbers beating expectations don't matter; the guidance for next quarter is what matters.
Microsoft, Amazon, Meta, Apple—all of which exceeded expectations in revenue. And what happened? Microsoft rose 15.5%, Meta fell over 9%, and Apple fell over 4%.
What's the difference? Guidance for the next season.
Palantir rose 15%, mainly because it raised its full-year revenue from 7.65 billion to 8.15 billion. The rise is said to "last another 18 months."
The quarterly figures are just tickets; the guidelines set prices.
Now there are three papers left: AMD, SpaceX, and Circle.
Three companies, three tracks, three completely different scripts.
Don't use the same formula to apply three different tickets.
Let's start with AMD (after market close on August 4).
Wall Street expects revenue of $11.3 billion, with adjusted EPS of $1.61. The company itself is guided at 11.2 billion (plus or minus 300 million), with a Non-GAAP gross margin of about 56%.
Data center revenue is expected to double to 6.5 billion, with server processors growing by over 70%.
What is AMD's biggest variable? AI chips.
In Q1, data centers have already generated 5.8 billion yuan (+57% year-on-year). The MI400 series accelerators and Helios rack platform were just released in July.
If AI chip demand exceeds expectations—go long on AI computing power-related tokens.
If it falls short of expectations—hedge.
It's that simple. Don't bet on direction—bet on "expectations are off."
Now, let's talk about SpaceX (after market close on August 4).
This is SpaceX's first financial report since going public.
Analysts expect revenue of $6.88 billion and a loss of $0.23 per share.
Starlink is the core—the internet business is expected to generate 3.83 billion revenue, with operating profit of 1.42 billion. But there's a painful detail: ARPU (Average Revenue per User) has dropped nearly 25% year-over-year—growth comes from price cuts.
But the real game isn't in the financial report itself.
On August 6th.
About 20% of the restricted shares will be unlocked, corresponding to a market value of approximately $116 billion.
Heavy positions before the financial report.
Observe the support force after lifting the ban before taking action. SpaceX's stock price has fallen from $135 at its IPO to below $110, more than halving its post-IPO high of $225.
First earnings report + unlocking hundreds of billions = maximum volatility.
Want to bet before the earnings report? Good luck.
Finally, let's talk about Circle (pre-market on August 5).
This is the most directly impactful part of the crypto market.
The earnings were released before the US stock market opened—right during the Asian trading session.
Wall Street expects revenue of $717 million, with EPS of $0.16-0.19.
Circle's revenue is almost entirely tied to USDC circulation and short-term interest rates.
The problem is—USDC circulating supply dropped from 77 billion at the end of Q1 to about 73.7 billion at the end of Q2. It fell by 4.6%.
Volume is shrinking, but interest rates are rising.
The energy shock triggered by the Iran war pushed up short-term market interest rates, partially offsetting the impact of declining USDC balances.
The core suspense is the RLDC margin—how much Circle can keep from USDC's reserve returns. Q1 was 41.4%, already exceeding management's full-year guidance of 38-40%.
If Circle's USDC circulating supply data exceeds expectations, it could trigger an overall valuation recovery in the stablecoin sector.
If it falls short of expectations—the stablecoin sector will face short-term pressure.
The three companies are located in three completely different tracks: AI chips, space economy, and stablecoins.
Split bets are safer than betting on just one piece.
Palantir has validated the formula of "beating expectations + upward guidance = surge."
But the remaining three companies have different scripts.
AMD is looking at the quality of its AI chips.
SpaceX is watching the acceptance after the ban is lifted.
Circle looking at USDC data.
Don't put Palantir's script on every ticket.
$AMD $SPCX $CRCL #财报观察员: AMD and SpaceX are about to hand over, with Circle as the grand finale When many people dig up that old Zhihu question about "buying BTC for 6000 yuan," their first reaction is: If you had bought it back then, would you have been financially free by now? 🤔
But one comment below was really heartbreaking. What we see today is the answer after 2017, the reverse after BTC became legendary. If you really went back to 2013, 2014, 2015, or 2016, what ordinary people experienced was not an upward curve, but a series of crashes, thefts, policy crackdowns, plummets, and public skepticism. 📉
In 2013, policy risks arose, and the market plunged; In 2014, heavy investors couldn't hold on and cut losses to exit; In 2015, during a long bear market, even the bulls began to question their lives; In 2016, Bitfinex was hacked, and the exchange risk hit everyone's face. 😰
So the question has never been "Why didn't you buy it back then?" but "Even if you had bought it then, could you really have lasted until today?" ”
💡 Many stories of getting rich, when read in hindsight, seem as simple as nonsense: buy, hold, forget the password, and get rich a few years later. But the players in the market are not just cold, backtesting data. You feel fear, doubt, can't help but check the news, get discouraged by those around you, and ask yourself over and over in the middle of the night when prices crash: Did I make a mistake? 🌪️
That's why I believe the greatest value of this kind of "crypto archaeology" is not to create regrets, but to remind everyone: what is truly rare is never the one "right buy," but that in the midst of great uncertainty, one still has awareness, position management, emotional resilience, and ultimately survives. 🛡️
It's truly a pity to miss out on BTC. But if you weren't prepared at all, just having fun, eating, eating pizza, living an ordinary life, isn't a bad thing. 🍕
The hardest part of investing is never understanding history when looking back. It's about whether you can get through it before history becomes history. ⏳$CORE CORE (Core DAO) Mining Degradation Mechanism (distinct from Bitcoin-style halving)
1. Core Conclusion: CORE does not have a Bitcoin-style halving event where 50% is cut every four years; it follows a smooth, decreasing annual model
5
The total CORE supply is capped at 2.1 billion tokens, with a total mining reward of 839.9 million tokens, released linearly and smoothly over 81 years, not a cliff-like halving
6
。
1. Deceleration trigger rules
Every 1,0512,000 blocks produced triggers a reward reduction, with a cycle of about 365 days (1 year), and the block reward is reduced by a fixed 3.61%.
Formula: Block reward for the next year = Current reward × (1 − 3.61%).
Features: Output is slowly declining, with no sharp halving nodes, inflation rate steadily decreasing year by year, block rewards after 1981 are close to zero, and miners' earnings will rely entirely on network transaction fees
9
。
2. Comparison with BTC halving mechanisms
Project Bitcoin BTC CORE Core DAO
Deduction method: Every 210,000 coins ≈ 4 years, the reward is directly reduced by 50% (cliff halved), decreasing by 3.61% annually, with smooth decay
Issuance cycle: approximately 140 years, 81 years
Total supply cap: 21 million tokens, 2.1 billion tokens (100 times BTC)
Mining reward pool: All block rewards are gradually halved until they disappear. Mining rewards account for 39.995% of the total amount, released evenly over 1981
6
3. Mining Reward Distribution Structure (Node Staking Mining)
A total of 839.9 million CORE exclusive node mining rewards, all distributed through block rewards, fully released in 1981;
Previously, block rewards were high, but as the annual decrease of 3.61% decreased, the output per block continued to decrease; With the total network hash rate unchanged, single-user mining earnings will gradually decline year by year, avoiding a halving where earnings are cut in half;
Early protocols burned part of the block rewards and on-chain fees, but now DAO governance has gradually removed the burn rule, instead using rewards for ecosystem support and node incentives, and no longer burning tokens permanently.
4. Easily confused point: Early BTCs versions of CORE halving rules
The initial BTCs testnet project design was modeled after Bitcoin: for every mined amount, the block reward is halved by 50% over about four years;
After the mainnet launched the official Core DAO version, the code was restructured, abandoning the 4-year halving mechanism and switching to the current smooth release model with an annual decreasing rate of 3.61%. The two versions have completely different rules and cannot be used interchangeably.
5. Impact on miners
Returns don't suddenly plummet; they decline slowly over the long term, with miners' income fluctuating less;
Over time, the proportion of block rewards continues to decline, and future node earnings will mainly rely on on-chain transaction gas fees and ecosystem staking dividends;
By the end of the 1981 cycle, block minting rewards were basically zero, and the network relied entirely on fees to maintain node incentives.BREAKING: US Treasury Secretary Bessent is asking the Fed to help Japan raise dollars without selling US Treasuries.
Japan holds around $1.1 trillion in US Treasuries.
When it defends the yen, it needs dollars, and most of its reserves are sitting in those bonds rather than in cash.
Once the cash runs out, the only way to keep intervening is to start selling Treasuries.
That selling would push US bond prices down and yields up. The 10-year already went above 4.7% last week, and Bessent has said he watches that number closely.
The FIMA Repo Facility avoids it.
Japan pledges its Treasuries to the Fed, borrows dollars against them, then returns the dollars later and gets the bonds back.
The bonds never reach the open market, so yields are not affected.
The problem is size.
The facility is capped at $60 billion a day, and Japan spent an estimated $60 to $80 billion in a single week.
That is why Bessent wants it expanded.
This is also why he sold euros instead of dollars on Friday.
But it is not his decision. Expanding FIMA needs a vote from the Federal Open Market Committee, and the Fed has declined to comment.
$USDS $ETH 1855, the highest reached 1875 in the early morning, but still couldn't hold and came back.
This wave is not just about weak technical skills; capital is also making choices.
Yesterday, BTC ETFs saw a net inflow of $170 million, while ETH ETFs saw a net outflow of $11.9 million, interrupting several consecutive days of ETH inflows. It's not that the money hasn't entered the market; it's just that it was prioritized for Da Bing, while Er Bing has been temporarily neglected.
Externally, they are still waiting for data on US JOLTS job openings.
Before the data is released, funds generally won't add positions rashly, and highly volatile ETH is easily reduced first.
The market also clearly shows:
After inserting the pin at 1827, ETH rebounded to around 1875, but then pulled back, repeatedly contesting the 1862 level before returning to 1855.
1850 is the short-term bottom; if you hold it, you can still touch 1862 or even 1874;
If 1850 falls, the next move will likely be to watch 1840, and if weaker, it will retest 1827.
So what does ETH look like now?
It's not that no one buys it,
The buying interest is not as firm as BTC.
Not chasing long positions in the short term; first see if 1850 can hold.
Before it can't return to 1862, any rebound can only be treated as a repair.
$ETH #从降息到加息, the Fed's disagreements are fully revealed MU technical position
Resistance: 885–900 (previous support turned to resistance, strong resistance at 4H) → 960 (right reversal confirmation line, only spot viewed) → 1031 (daily super trend line, medium-term bull-bear dividing)
Long-short divide: 830–840 (underlying stock closing at 829.5 + on-chain consensus zone, you above 850 = closed premium long market, below 830 = return to underlying stock logic)
Support: 800–820 (heavy trading volume in 4H) → 770–786 (intraday low on 8/3 + platform support) → 740–750 (extreme pullback / major bearish target)
and SNDKUSDT (both are US stock synthetic contracts)
SNDK 1316: Anchor 1288, premium +2.2%. SanDisk's AI NAND logic is purer, with high elasticity and wild volatility
MU 850: Anchor 829.5, premium +2.5%. Micron DRAM + HBM dual lines, but institutions are fighting between long and short positions (Burry short vs. data center long). Long positions are crowded, funding rates are positive, open interest is declining→ and there is a risk of a stampede amid the rebound
Both are subject to the constraint of 'tonight's US stock market gap-up convergence.' The 850 price level is overly high during the market break, so don't place orders anchored at 850; set 829.5±2% to reserve slippage
Trading Closure Reminder
Currently, the Asia-Europe session is a closed window for US stocks, and MUUSDT liquidity is much thinner than BTC. Inserting a pin near 850 to 865 or 820 is normal
21:30 The underlying stock most often sweeps in the opposite direction 5–10 minutes before market opens. Heavy positions above the watershed at 830 = betting on premiums without taking losses
50x leverage near the anchor point of 829.5 means a gap will be wiped out immediately#ISM创四年新高, U.S. Treasury yields have reversed
The 30-year Treasury yield once surged to 5.23%, and the last time it reached this high was on the eve of the 2007 financial crisis.
Right now, the market isn't arguing about whether it's rising, but whether it's over.
The optimistic side is straightforward: as long as the US economy starts cooling and rate cut expectations return, long-term yields will naturally have reason to fall. But the other camp is not buying it—the US fiscal deficit is widening, Treasury issuance is hitting new levels every year, and the market is repricing for 'sovereign credit risk.' The ultra-long-term 30-year term may have to last a long time at high levels.
The 30-year U.S. Treasury has long been used as the pricing anchor for global assets. Once it firmly hovers above 5%, it's not just bonds themselves that will be revalued: the discount rates of long-duration U.S. growth stocks will be pushed upward, mortgage and commercial real estate financing costs will tighten, the opportunity cost of holding gold will change, and even high-beta crypto assets will have to be recalculated.
This wave isn't just a minor fluctuation in interest rates; it's a loosening of the underlying assumption of 'cheap money' over the past decade—whether the danger is just beginning depends on whether fiscal policy can be contained. If not, 5.23% may not be the top.
$BTC $SNDK Biconomy想让普通人轻松用上Web3，结果自己的代币却成了最难“轻松持有”的资产之一。
$BICO 曾经是“让Web3更好用”的基础设施叙事代表，如今却沦为高集中度、低关注度、几乎被市场遗忘的小市值代币——技术故事还在，但价格、流动性和真实采用都已经讲不下去了。
前100个钱包掌握约96%的供应，Gini系数接近1，典型的“巨鲸控盘”结构。一旦大户出货，价格瞬间崩盘。流动性占市值比例极低，稍有波动就容易出现大幅滑点，普通散户进出都困难。
从2021年底ATH约21-22美元一路跌到目前0.017美元附近，跌幅超过99%。即使偶尔有短线反弹（多由空头挤压或鲸鱼拉动），也很快被抛压打回。长期持有者基本处于深度套牢状态。
历史上多次出现团队相关钱包解质押后往交易所充值的操作，且常发生在价格相对高点附近。这种行为严重打击市场信心，让人怀疑项目方对代币长期价值的态度。
Account Abstraction（账户抽象）、无Gas交易、跨链体验优化这些概念在2021-2023年很热，但如今竞争者众多（Gelato、Pimlico等），Biconomy并没有形成明显护城河。用户和开发者真正留下来的比例不高，生态活跃度与宣传严重脱节。
$BICO 主要用于支付网络费用、质押和治理，但实际需求有限。协议本身的收入并没有强有力地回流到代币持有者身上，导致“用得上但买不动”的尴尬局面。
$BTC $ETH #交易之声：你的经验值得被听到 #ISM创四年新高, U.S. Treasury yields have reversed
ISM data is extremely strong, but the bond market simply isn't buying it.
In July, the ISM Manufacturing PMI soared to 55.6, the highest since May 2022, exceeding expectations. Logically, with such strong economic data, expectations for rate hikes should heat up and U.S. Treasury yields should rise. But the actual trend was completely reversed—the US and Iran returned to the negotiating table, oil prices fell over 7% in a single day, government bond futures surged across the board, and long-term yields actually turned downward.
Data calls for rate hikes, the market is moving toward rate cut expectations, two forces are fighting, and the bond market has chosen to trust oil prices and exchange rates rather than strong ISM data.
What does this contradictory signal mean for the crypto world?
In the short term, falling U.S. Treasury yields provide a breathing room for risk assets. After hovering around 64,000 for so long, if macro pressure can ease a bit, at least it won't be pushed downward. But in the medium term, the ISM data is clear: the real economy remains overheated, and expectations for rate hikes have not truly been eliminated. How long oil prices can fall and what negotiations can reach are what truly determine the direction of U.S. Treasury yields, not the ISM monthly data.
My view is clear—ISM data is no longer important; the logic of market trading has shifted to the line of "oil prices and exchange rates."
There are currently only two variables that truly determine the direction: whether the US-Iran negotiations can be reached, and how long the joint US-Japan intervention can last. These two variables determine the next stop for Treasury yields, not the quality of ISM data.
If negotiations succeed, oil prices continue to fall, U.S. Treasury yields fall, and the Bitcoin market has a chance to move upward. If negotiations break down, oil prices rebound, U.S. Treasury yields return to 5.3% or even higher, and the market will have to keep grinding.
In my opinion, the second scenario is more likely. The reason is that Trump and Iran have fundamental contradictions, the negotiation foundation is weak, and the situation could flip at any time.
Bitcoin's current pricing anchor is not on the chain but on U.S. Treasury yields. The 5.27% level determines the direction for the entire month of August. Before the direction appears, it's better not to move than to move recklessly.
$BTC $SNDK $ETH #从降息到加息, the Fed's disagreements are fully public
Divisions within the Federal Reserve have never been as blatantly exposed as they are today. Logan, Hamak, and Kashkari advocated for rate hikes; Waller publicly supported rate cuts; but Chairman Wash refused to provide any forward-looking guidance, passing the blame to two CPI data points.
My view is clear: betting on a rate hike in September and believing that the current market's one-sided pricing is not only reasonable but may even be the only correct direction. The core reason is not whose voice is louder, but a harsh reality—inflation has exceeded target for more than five consecutive years, and the Fed's "credit gap" is being priced in real money by the market.
Why raise rates instead of cut rates? The reasons for supporting rate cuts sound promising: the economy may weaken, and inflation may fall. But the logic of hawkish officials is more convincing at the moment:
1. Inflation stickiness far exceeds expectations. In June, PCE year-on-year remained high at 3.7%, far from the 2% target. Cleveland Fed President Hamack bluntly stated that inflation has stubbornly stayed above target for more than five years, and she does not believe prices will return to normal on their own.
2. Hawkish rhetoric and restrained actions are backlashing at the Fed. Walsh repeatedly emphasizes his determination to fight inflation but refuses to raise interest rates, which the market interprets as "bluff." The result was a surge in long-term U.S. Treasury yields—the 30-year Treasury yield briefly surpassed 5.2%, reaching a new high since 2007. This is not the market relaxing, but rather punishing the Fed's hesitation with even harsher tightening.
3. The prerequisites for rate hikes are already self-realizing. Walsh said the financial markets have partially tightened the Fed's work. But it is precisely this surge in rates that work for the central bank that makes rate hikes even more necessary—if this market signal is not confirmed through formal rate hikes, the Fed's credibility will be completely ruined.
The market now rates a rate hike in September with a probability of over 67%. Some believe that betting before the CPI data is released is too extreme, but I believe this is extremely reasonable.
The current market is not blindly biased, but a rational correction of the Federal Reserve's inaction. After Wash's removal of forward-looking guidance, the market lost its anchor. The only thing we can anchor is the reality of inflation. If the numbers decline, Wash's approach can go with the flow and hold back; But if the data rebounds, he will face the dire situation of "losing credibility if he doesn't act." The market is now pricing in rate hikes in advance, essentially forcing the Fed: either you prove your credibility through actions, or the market raises rates for you. Talk about SNDK
The daily chart has retraced nearly 45% from the historical high of 2376 in late June, recently rebounding from the 972 low to around 1360. The price is below the Riblin middle band (≈1464) and above the lower band (≈1041), representing a "recovery phase after the daily low golden cross, no reversal."
Technical level (short-term contract, not recommended)
Resistance: 1350–1380 (4H previous high/trapped zone) → 1430–1464 (Middle Band of the Jiblin Ring, only talk of trend recovery after it breaks) → 1887 (Upper Band of the Jiblin Ring, medium-term target)
Bull-Bear Divide: 1300–1320 (your 1316 is close to the upper boundary; holding = rebound continuation, losing hold = pullback to 1280)
Support: 1280–1265 (early morning rebound starting point) → 1200–1160 (yesterday's intraday low in US stocks)→ 1040 (daily Blin lower band/extreme defense)
and the differences in BTC/ETH linkage
BTC 63750 and ETH 1865 are crypto-native assets, driven by on-chain liquidity + macro factors;
SNDKUSDT essentially serves as a thermometer for sentiment in US semiconductor stocks, ultimately anchoring to the underlying stock. At the US market opening (tonight at 21:30 Beijing time), there will be a one-time gap converging toward 1288, and the 'premium' at 1316 during the closed period may be wiped out or widened, depending on pre-market futures.
SanDisk Logic: AI data center NAND demand + explosive performance (gross margin 78%+), but the stock price has risen too much this year. In the short term, it's a 'high-level oscillation searching for a central point,' not a bottom-starting move.Palantir's quarterly report just dropped, and whether AI has truly reached its fulfillment phase can be set aside for now; just looking at the numbers, it definitely stunned Wall Street.
Q2 revenue hit $1.935 billion, a year-over-year jump of +93%, marking the highest growth rate in the company's history. After hours, the stock price surged about 13%.
But what really excites the bulls isn't the total size, but the structure: U.S. commercial business revenue was $764 million, soaring 149% year-over-year; U.S. government business was $809 million, up 90% year-over-year—both sectors booming together shows it's not just propped up by military contracts, but enterprises are genuinely starting to pay for "deployable AI."
Supporting metrics are solid too: total new contract value for the quarter was $3.37 billion, adjusted free cash flow surpassed $1.2 billion for the first time, and the full-year revenue guidance was raised sharply from a range of $7.65 billion to $8.15–8.16 billion.
Previously, the market often grumbled that "Palantir is just a government project contractor," but this time U.S. commercial clients (including large enterprises) rewrote that narrative with 149% growth—AI demand has moved from PPT slides to purchase orders, and Palantir is among those who have secured their entry pass.
#Palantir营收增93%，盘后涨13% $PLTR $SNDK $BTC This signal is also good news.
Based on my summary and analysis of historical trends.
In past cycles, whenever a weekly MACD golden cross appeared, the period typically saw a minimum 10% increase from the opening price to the highest point.
Some one-sided bull markets even saw prices doubling.
Here's something interesting: now the weekly MACD has shown a golden cross again.
However, the current price has not risen by 10% from the opening price; instead, it has dropped by over a thousand dollars compared to the opening price.
According to historical patterns, before the weekly golden cross cycle price increases did not exceed 10%, the win rate for going long was quite high, especially since the price has fallen now.
So I think if the price drops a bit further, we can build positions in batches and aim for a minimum 10% gain in this cycle.1.73亿美金期权押注：比特币9月前难破7万！空军正在集结？
大饼在63500美元晃悠，上面有1.73亿美金期权压着不让涨，下面有20亿美金多单等着爆——这盘棋，你看懂了吗？
消息面：三座大山压顶
第一座山：1.73亿期权压顶。 某大户卖出了2709张比特币看涨期权，名义价值1.73亿美元。未来52天涨幅超过9.5%他就亏钱——大户的钱都在看空，你细品。
第二座山：美债市场告急。 30年期美债收益率飙到2007年以来最高。债市这个“沉睡巨人”一醒，全球借钱成本都要涨，股市和大饼都得跟着遭殃。
第三座山：ETF资金分化。 大饼ETF昨天净流入了1.7亿美金，看着是利好。但以太坊ETF净流出1190万——机构钱往大饼跑，不代表马上拉盘，更多是避险配置。
技术面：4小时图已露疲态
4小时级别从64000附近承压回落。布林带中轨63178，上轨63987价格贴着上轨下方，说明上面抛压不小。MACD多头缩量，小时级别也在缩，短期上攻动能不足。
清算地图：20亿美金的筹码博弈
涨破65000，将清算20亿美金空单；跌破62000，同样清算20亿美金多单。
主力最爱干嘛？先往上拉一波爆空头，再反手砸盘爆多头，两头通吃。现在价格63800附近，往上空间只有1200点到65000，往下空间却有1800点到62000。往上的“路”更短，往下的“坑”更深。
个人观点
激进玩家：63500-64000附近轻仓空，目标看62500-62000。理由：64000抛压密集，加上1.73亿期权压顶和美债利空，短期很难一口气冲上去。
做多：等61500附近再考虑接多。理由：62000以下有20亿多单要被清算，主力很可能先砸穿62000把多头扫干净，到61500附近才是真正值得捞底的区域。
#从降息到加息，联储分歧全公开 $BTC #美日确认联合购汇
The US and Japan joining forces to buy yen appears to be a move to save the exchange rate on the surface, but in reality, it is a forced brake on the "borrowing yen to speculate globally" game — the yen gets a short-term bargain, but the dollar, US stocks, Japanese exporters, and global liquidity all suffer as a result.
Previously, Japan defended the market alone, and any pull was quickly reversed. This time, with the US directly involved, the significance is different: the USD/JPY rate approaching 164 is no longer just a Tokyo issue; it’s something that can shake the global table.
Why? Because the yen has depreciated so much over the years due to interest rate differentials. Everyone borrows yen at near-zero cost, then turns around to buy US Treasuries, US stocks, gold, Bitcoin — this chain is called "carry trade," with a scale exceeding one trillion dollars. When the yen suddenly rises sharply, the chain reverses: cutting global risk positions → covering yen → repaying loans → yen rises further → more forced liquidations. The scariest part of the joint intervention isn’t how much money was spent, but that it made this chain start to creak.
The US was smart this time, not dumping dollars but having the New York Fed sell euros to buy yen. The benefit is: the yen rises, the dollar index won’t collapse, and US import inflation won’t be fueled by a weak dollar. So the situation is roughly — USD/JPY falls, EUR/JPY is pressured, the dollar index may not crash, but the 164 level is marked by the market as a "red line," and every time it approaches 160–164, shorts have to consider whether they’ll get slapped again. This is not a new Plaza Accord; the US doesn’t want the dollar to depreciate against all currencies, just to keep the yen from running wild.
On the US stock side, the short-term pain point is leverage. Previously, yen financing heavily flowed into Nasdaq, semiconductors, and AI leaders; a yen rebound forces some money to sell stocks to repay yen loans, amplifying volatility in high-valuation tech stocks. But it’s not purely negative: a relatively softer dollar makes the overseas profits of multinationals like Apple, Microsoft, and Amazon look better when converted back to dollars. In the medium term, the key for US stocks remains long-term US Treasury yields — if Japan sells long bonds to intervene, yields rise and tech valuations come under pressure; this time Japan uses the Fed’s FIMA repo tool to pledge Treasuries for dollars, aiming to minimize bond sales. So the US stock outlook is: short-term liquidity is tight, medium-term watch Treasury yields, long-term depends on AI profitability resilience.
Japan itself is more divided. A stronger yen makes oil and food prices cheaper when converted to yen, benefiting airlines, power, retail, food, and domestic banks; but companies like Toyota, Sony, and Tokyo Electron that earn overseas money see their earnings shrink when converted back, dulling export price advantages. The Nikkei 225 may not rise in sync with the yen and might underperform more domestic-focused indices.
Expanding outward, the truly global risk is concentrated carry trade unwinding. The yen is the world’s most critical financing currency; when exchange rate risk rises, not only US stocks but emerging market equities and bonds, high-yield bonds, private credit, gold, oil, BTC and highly leveraged crypto, and high-valuation Eurasian tech are all hit. Calm unwinding = global deleveraging; if the yen jumps 5%–10% in a few days, a chain reaction like the August 2024 plunge could replay.
So can this move really save the yen? Short-term it can knock out speculative positions, but medium to long term, not necessarily. Intervention can’t change the fundamental US-Japan interest rate differential. As long as US rates stay high and the Bank of Japan is slow, the carry trade will heal and come back. Only three real signals matter: whether the Bank of Japan continues to raise rates, whether US long-term yields fall, and whether Japan’s fiscal policy stops expanding recklessly. Intervention alone without policy support means the yen will weaken again after a sharp rise; intervention + rate hikes + falling US yields could mark a reversal starting point.
In summary: the US-Japan joint yen purchase is not just market defense, it’s a global liquidity reshuffle. The short-term pain is not for the dollar but for those high-leverage trades relying on cheap yen; the real alarm to watch is "yen rising continuously + global risk assets falling together" happening simultaneously. $GOOGL $TER $XAU I was just squatting on the toilet to check the news, saw MSTR selling coins again, my first reaction was to pull up my pants and run away, but when I looked down, it was only 1,638 coins? Compared to last time, it was cut in half, and he sat back down right then and there.
Last time, I really got totally ripped off. At that time, I added 63,000 to my position, bragging to my brothers in the group, saying that with MSTR faith support, reaching 70,000 was only a matter of time. But that very night, it broke out that over 3,000 units had been sold, and the next day it opened low to 59,000. I was stunned, holding out for nearly three weeks before breaking even, unable to sleep well in between, and waking up in the middle of the night to check the market was common.
So this time, I've learned my lesson: don't rush to cut the news when it comes out, settle the score first. This time, 1,638 pieces were sold, with an average price of over 63,900 yuan, nearly $10,000 less than its own cost of over 70,000 yuan—pure loss clearance sale. If the market outlook is truly bearish, is it really necessary to cut losses at low levels? Simply put, the company is pooling money to pay dividends and conduct buybacks, purely financial operations that have nothing to do with BTC itself.
The most interesting aspect is the market. Most of the day after the news broke, BTC was hovering around 62,000, without even a 1,000-point fluctuation. Last time, he would have already dived, but this time he acted as if nothing happened. What does this mean? The chives have been cut smart; last time they were scared once, and this time when the wolf comes, no one will run away. The last time you should have cut off, the rest are either stuck and lying flat, or truly taking the long term. The selling pressure is much less intense by now.
As for ETH, don't even mention it—it's purely a stock that follows the decline but not the rise. They sell Bitcoin, which has fallen even harder than BTC, and when it rises, it lags behind. My 2000 budget is still hovering near the cost line, truly a hopeless idle. I used to think the Ethereum ecosystem was strong, but now it seems that in the face of real money, any ecosystem is just empty; having institutions to support the bottom line is the real deal.
I checked, and its board of directors can sell up to $5 billion worth of coins. These two rounds combined only sold just over 300 million, so there will definitely be more to come. But this time, the quantity was cut in half, showing that it doesn't want to invest too hard. If it has enough turnover, it will stop. It can't just break through its own cost line, right? If you really want to dump, wouldn't it be better to sell a bit more at once?
Now, I actually think it's a good thing. The worst thing about negative news is that it hangs in the air and you never know when it will hit. Now the brand is clear, selling in batches, selling less and less, which actually feels reassuring. One boot had already fallen to the ground, and the other was about to be done—better than living in constant fear.
I still stick to my usual approach and keep my bottom position untouched. Buy when it drops to 60,000, reduce when it rebounds to 64,000, and trade swings back and forth. I really don't want to add ETH anymore. I'll wait until it can hold above 2050 before talking. Don't touch weak coins—every time you touch them, you lose.
Do you think MSTR will be sold again next time? Has Polykun really finished coming out yet? Are there any brothers like me who got buried last time?
#MSTR再卖1638枚比特币, scale halved