Orbit Post Sitemap
The better the earnings, the harder the fall—Wall Street's moves are thoroughly dissected by Lao Mo
Brothers, last night the US stock earnings season staged an absurd drama.
SanDisk's revenue surged 372%, EPS multiplied 135 times, yet it fell 8% after hours. SpaceX's revenue exceeded expectations by $1 billion, AI revenue tripled, but the stock price crashed 13%. AMD beat expectations with data center revenue doubling, yet it still dropped 7%.
The better the performance, the harder the fall—this isn't a fundamental problem, it's a big issue with expectation management.
Let's start with SanDisk. The data is flawless; the killer is the guidance.
Revenue was 8.97 billion, a 372% year-over-year surge, beating market expectations of 8.394 billion. GAAP net profit was 6.903 billion, non-GAAP adjusted EPS $39.25—compared to only $0.29 in the same period last year. Gross margin was 84.6%, up 3.2 times from 26.4% a year ago.
Data center revenue was 2.98 billion, doubling quarter-over-quarter and soaring 1298% year-over-year. Edge computing was 5.43 billion, up 392% year-over-year. Full-year revenue was 20.25 billion, up 175%.
They also approved a $14 billion buyback, with $15.5 billion remaining in authorization. At current market value, this could repurchase about 6.6% of the float.
Then it fell 8% after hours.
Why? The guidance didn't beat the market's imagination.
For fiscal 2027 Q1, revenue guidance is 10.3-10.8 billion, midpoint 10.55 billion. Wall Street expected 11.16 billion. EPS guidance is $44-$46, market expected $45.58. Gross margin guidance is 83%-85%, nearly flat with Q4's 84.6%.
The market wants "shock," you gave "very good"—not enough.
But Lao Mo says fairly: it's not that sales can't be made, it's that production can't meet demand. SanDisk said on the call that over half of fiscal 2027 capacity is locked in by long-term contracts, and about two-thirds of fiscal 2028 is also secured. Signed long-term contract minimum revenue scale is $93.9 billion. Capacity is maxed out, not demand gone. But Wall Street ignores this; a $300 million shortfall in guidance is enough to punish.
Next, SpaceX. The earnings are impressive, but the unlock is the real blade.
Revenue was 7.814 billion, up 92% year-over-year, beating expectations by about $1 billion. AI revenue was 2.56 billion, more than tripling quarter-over-quarter. Net loss narrowed from 1 billion to 540 million.
Then the stock dropped 13%, wiping out 225 billion in market cap. Less than two months after listing, the stock price has fallen nearly 50% from its high.
Two reasons: capital expenditure exploded, and the unlock arrived.
Q2 capital expenditure was 18.4 billion, a 5-fold year-over-year increase, including 15.8 billion related to AI, a 20-fold surge. The burn rate is more than twice the earning speed. CFO said Q3 and Q4 will maintain this pace.
Today (August 6), 911.5 million restricted shares unlock. Equivalent to 1.4 times the existing float, valued over 100 billion at $110 per share. Short positions account for 34% of the float, nominal value 24.6 billion. These players started positioning in June, with huge unrealized gains and are still adding.
Good or bad earnings have to take a backseat to this supply shock.
Back to the market.
The Dow rose 0.49% to 54,349, marking five consecutive gains and a new high. But the Nasdaq fell 0.83%, and the S&P 500 dropped 0.17%.
BTC latest price is about 64,600-65,000, trading in the 63,800-65,000 range over 24 hours. ETH is at 1905. Bitcoin did not fall with the US tech stocks—indicating that the macro logic of easing geopolitical tensions, falling oil prices, and declining US bond yields temporarily outweighs the tech earnings negative impact.
Lao Mo offers some trading advice.
For brothers stuck in SanDisk: 1226-1250 is short-term support; hold it and wait for a rebound to 1310-1320 to reduce positions. If it breaks below 1226 with volume, cut losses. The $14 billion buyback is not just for show, but don't bet your life on it.
For those empty-handed wanting to enter SanDisk: wait for 1226-1250 to stabilize on low volume before acting, stop loss below 1200, target 1310-1320. Or wait for the price to climb back above 1310 before chasing on the right side.
For Bitcoin: 64,000-64,200 has become support. Buy on a pullback to this area if it stabilizes, stop loss below 63,800, target 65,000-65,500. 65,000 is the touchstone; if it passes, the rally may restart; if not, it will continue to consolidate.
Lao Mo's final word: This earnings season teaches us one thing—good earnings don't equal stock price gains; expectation management is more important than earnings themselves. The problem with SanDisk and SpaceX isn't the companies, but that the market's imagination of "infinite growth" has become unreasonable. You gave very good, but it wants unimaginable.
But the mid-to-long-term logic remains unchanged—AI storage shortage until 2027, eight long-term contracts locking in $93.9 billion revenue, HBF new standards just implemented. Short-term volatility is emotion; long-term trend depends on supply and demand.
Are you bottom-fishing or waiting on SanDisk and SpaceX this round? Discuss in the comments.
If you think Lao Mo's analysis is clear, give a like and follow; I'll alert you immediately when key levels are reached. $BTC $ETH $SNDK #交易之声：你的经验值得被听到 我是中线情报哥，刚看到$BICO 的 4 小时 K 线。
最新价 0.02810，24 小时狂涨 21.5%，新币种 Layer1&2 赛道排第三，热度炸了！
EMA5/10/20 呈多头排列，量能 VOL(BICO)827M、USDT24.7M，MACD 红柱放大，KDJ 虽近超买但趋势没拐。
不过 24h 最高 0.03185 后回落，S 信号出现，短期或有回调，但看均线支撑（EMA5 0.02723）仍在，
只要不破位，这波新币行情我继续看多做波段，盯紧 0.02400-0.03185 区间的突破方向！
#波动雷达：币种异动观察$SNDK Over 140 short whales have closed $90 million in short positions, making the intentions of the main funds clear.
The backend data is very realistic: the number of whales and capital shorts completely outpace those who went long. The Air Force had already built heavy positions at the 1324 high, and now they hold nearly $6 million in huge profits, watching the drama.
In contrast, the more than 400 long traders have all their average prices stuck on it, and their accounts are bleeding. This broken market has no big money willing to enter to support and rally. Once these long-holding bulls are completely desperate and cut losses, a wave of crushing can happen at any moment, directly shorting in the direction of the bear whales!$SNDK
SanDisk: The inevitable collapse of the AI bubble
SanDisk's current stock price frenzy is a complete "passing the buck" to a brutal value destruction.
Core Bearish Logic 1. Commodities disguise AI: NAND flash has no moat, with a gross margin as high as 84.6%, which is a death knell at the peak of the cycle, not the start of a long bull market.
2. Samsung's Dimensionality Reduction Squeezing: Samsung is using more advanced and cheaper chips to penetrate the high-end SSD niche, and SanDisk's high gross margin moat could be filled at any moment.
3. Insiders panic and flee: Big stock data from both the east and west sectors is selling off at a 25% discount, and those who know the best cards are fleeing for their lives. #闪迪今日行情
4. Valuation Collapse Imminent: Consumer demand plunged 32%, and the market was wiped out simply because guidance slightly missed expectations, setting an extreme "Davis double kill" on the verge of erupting. #闪迪财报双超预期, an additional $14 billion repurchase authorization was added
When the AI frenzy fades, it's the day SanDisk's crash begins—don't look for support amid an avalanche.#闪迪财报双超预期, an additional $14 billion repurchase authorization was added
I am in a short-term bearish position on Korean AI chip stocks this time, so I'm not in a hurry to buy them.
Today, KOSPI fell more than 4%,
Samsung fell about 6%,
SK Hynix once fell nearly 10%.
But strangely, it was:
AI demand has not collapsed,
Profits haven't collapsed either.
What really frustrated the market was—
Money is made,
Shareholders, however, feel they haven't received much of their share.
Samsung and SK Hynix expect their combined net cash to reach about $263 billion by year-end.
As a result, shareholders now directly demand:
Much of the red dividend,
and many buybacks,
Less is just about burning capital expenses.
So I believe this decline is not just about the "AI market stalling."
It seems the market is starting to ask again:
Money earned by AI,
In the end, who exactly did it go into?
I remain cautious in the short term.
But if Samsung and SK Hynix really come up with more aggressive buyback and dividend plans later,
On the contrary, it may become a catalyst for the next round of rebound.
This AI market is entering its second phase:
Previously, we only looked at growth.
Let's start looking now—
After making money, how much can shareholders actually get? $SAMSUNG $SKHYNIX $Bless It feels more like someone is pulling contracts through small circulation contracts.
Spot trading in one day was only 630,000 U, but OI was 17.72 million U, meaning the money opening positions in the session far exceeded the money spent buying coins. The price was pushed to 0.009, and the fee rate was raised to around 0.06%, indicating that after the rally, long positions have been actively squeezed in. But today, there was no new news to absorb this funding, and the project team did not present any new business progress. Going forward, they could only continue to maintain prices through contract additions.
I lean toward ending this rally around 0.009. This has failed to break through many times before. Once the OI remains high but the price starts to stabilize, it's easy to first buy long and then return to 0.008. Even if there is a sharp pull after BLESS, I treat it as a bullish incentive rather than a main rally.我入圈以来最贵的一课，不是来自BTC的暴跌，也不是来自某个山寨币归零，而是来自一个叫BULLA的meme币——我亲手把一笔有止损、有纪律的小仓位空单，一步步加成了一笔没有止损、没有纪律的爆仓大单。
事情是这样的。当时BULLA市值从几百万拉到接近4亿美元，纯meme驱动，没有任何基本面支撑。我判断这种纯情绪币早晚要回来，小仓位开了空单，止损设好了，心态很稳。如果打止损，也就亏一点点，完全在计划内。
问题出在它没打我的止损，而是横盘了一阵之后又开始涨。
我当时脑子里的声音是："已经涨这么多了，不可能再涨了。"于是加了一点仓位。还是小仓，还算有理性。但这一笔加仓没有重新设止损——因为我觉得"都涨成这样了，止损设在哪都不合理"。
然后它又涨了。
第三次加仓的时候，我的总仓位已经不小了。但因为是一笔一笔慢慢加上去的，每一笔看起来都不大，加在一起的风险敞口已经完全超出了我的承受范围。这时候浮亏已经很痛了，但我告诉自己"都亏这么多了，现在割肉更不划算，再扛一扛说不定就回来了"。
可是没有回来。最终爆仓离场。
事后复盘，这笔交易教会我的东西，比任何一本交易书都多。我把它总结成了三句话，现在每次开仓前都会默念一遍：
你的止损纪律，不是在你下第一笔单的时候被打破的。第一笔单你很清醒，仓位小、止损明确、心态平稳。纪律是在你第二次、第三次加仓的时候悄悄死掉的——因为每次加仓都会给自己一个"合理"的理由，而这些理由的共同点是：它们都在帮你回避"我可能错了"这个事实。
"已经涨这么多了不可能再涨"——这句话是逆势交易者的致命毒药。市场不在乎你觉得什么价格"合理"。BULLA从市值几百万涨到4亿又跌回2200万，整个过程没有任何一秒是"合理"的，但它就是发生了。你用"合理"去对抗市场的"不合理"，结果只有一个。
亏损越大越不舍得止损，这是人性，但也是爆仓的直接原因。小仓位的时候止损毫无心理负担，因为亏得起。但当你不断加仓把小亏变成大亏，止损的心理成本就会指数级上升，直到你完全丧失执行力。风险管理的核心不是"会不会设止损"，而是"在仓位膨胀之后还能不能执行止损"。
现在我给自己定了一条铁律：任何一笔交易，不管中间发生了什么，总仓位不能超过初始计划的2倍。想加仓可以，但必须同时收紧止损，让总风险敞口始终锁死在开仓时愿意承受的范围内。一旦发现自己在找理由不止损，立刻平仓，不给自己第二次犹豫的机会。
这条规则让我后来避开了不少坑。说到底，交易能力的上限不是由你赚了多少决定的，而是由你在该止损的时候能不能拉得下脸来决定的。
以上内容仅代表个人观点，不构成任何投资建议。DYOR, NFA.$SPCX 凌晨的拉高果然又是诱多陷阱，现在价格已经回落到 110，前几天积累的涨幅基本全部吞没。我 110 开了仓位，最高看到 130 没走，现在又砸回原点，这一轮过山车坐得属实难受。 今晚 9 点 30 分是首期解锁的关键节点，市场出现恐慌情绪是必然的。9.1 亿股的解锁规模看着吓人，但并不意味着全部都会砸向市场，如果按照现有价格卖出 1 到 2 亿股，那么价格大概率回到 100 到 105 的区间之内。只要有人恐慌抛售，就有人敢进场抄底，但问题是跌破两位数的价格或许只是时间问题。 后续的风险远不止这一次，8 月份还有多轮解锁安排，9 月份同样有压力释放。一次解锁可以用抄底逻辑扛住，但连续几轮的抛压叠加，市场承接力会越来越弱。现在这个位置去接飞刀，等于和整个解锁周期做对手盘，胜算真的不高。 $SPCX 的短期走势已经被解锁预期锁死，反弹都是减仓机会，不是加仓理由。守住本金比博反弹更重要，等 8 月解锁潮全部落地之后再评估方向不迟。 SPCX #Circle财报后押注Arc，USDC能否迎来新增长？ #财报观察员：业绩喜忧参半，解禁将至！SpaceX后续怎么看？ #Polymarket洽谈10亿美元融资, with a valuation exceeding $20 billion
In April, it raised only $15 billion, and now it's asking for $20 billion+ right away. A round of $1 billion is exaggerated to $1.2 billion+ in annualized revenue—Polymarket's round isn't ordinary fundraising; it's packaging itself from a "crypto niche betting on elections" into next-generation macro information infrastructure.
But my honest view is simple: valuations can be trusted, but revenue quality must be compromised. Regulation is the hidden danger that hasn't been priced in.
Here are a few points:
1) The rapid price increase is real, but revenue is somewhat inflation from the "rate switching."
Before the end of 2025, Polymarket had many zero-fee rates; in March 2026, with the launch of the expanded fee + US compliant version (QCEX), the annualized rate jumped from nearly 0 to 1.2 billion. This is more about "starting to collect taxes" rather than "tripling natural demand," annualizing changes in fee rules and easily misinterpreting one-time jumps as permanent profitability.
2) The opponent, Kalshi, is actually the bigger one
Of the industry's 50.6 billion nominal volume in July 2026, Kalshi alone will consume 37.7 billion, Polymarket International 7.9 billion, and U.S. 5 billion. Kalshi's valuation has reached 22 billion, with 40 billion in talk, and Polymarket's 20 billion+ is more like a "benchmark quote" than a liquidity premium.
3) CFTC marketing investigation + state-level gambling law conflicts are like a knife hanging over the head
Paid KOL exposed for simulating profits through fake sites; CFTC investigation initiated; At the same time, multiple states treat sports prediction contracts as gambling. The Compliance Edition ≠ Risks Cleared Away, Instead bringing the platform into the spotlight of Washington and state attorneys.
4) Why do I still favor the sector but not chase this round of valuation?
The prediction market turns "collective judgment" into tradable prices, which have long-term value—Google has already taken the data. But Polymarket is now buying a triple sentiment stack of "political cycle + World Cup + US compliance dividends," not steady-state cash flow prices. A valuation of 20 billion corresponds to an annualized revenue of 1.2 billion, which is 16–17 times PS. This is not outrageous in crypto, but relatively expensive in compliant financial infrastructure.
What crypto veterans should focus on is not "whether it's worth 20 billion," but rather: if the next round of liquidity enters the prediction market, will funds choose PM platform tokens/ecosystems, or will they directly gamble on the upstream APIs/oracles of Kalshi and Polymarket? This is what can be mapped to our positions.#闪迪财报双超预期, $14 billion new buyback authorization $SNDK All the good news has materialized, and every time it does, it's a negative one. SanDisk's sharp drop was already destined for the day. SanDisk's financial report showed impressive results, with revenue and EPS both exceeding market expectations, and it even launched a massive $14 billion buyback plan, piling up a wealth of positive hype and attracting a large number of investors to wait for a rally. But the market never looks at past results; funds focus on next quarter's earnings guidance. The midpoint of the FY2027 Q1 revenue guidance range fell short of market expectations, directly bursting the optimistic bubble. The previous rally had already exhausted all the positive news from financial reports and buybacks, and funds had already been selling off in batches on the positive news. This is a typical "positive news delivered = negative news" market: before the positive news is released, the market speculates in advance to drive the market up; The news officially landed, with no new expected support for the market, and profit-taking positions were concentrated in selling and dumping. Massive buybacks may look tempting, but in the short term, they cannot reverse the trend of capital flight, unable to withstand short selling pressure. The story of AI storage demand has been repeatedly hyped, and the market now questions whether storage price hikes and flash demand can sustain high valuations, causing market divides to rapidly widen. Technically, prices plummeted, falling over 6%, and retail investors chasing long positions at high levels were completely trapped. Don't be blinded by superficial positive news; in the capital market, poor expectations are the key to price fluctuations. No matter how good the performance or how big the buybacks, as long as they don't exceed expectations,The market has raised the bar for SanDisk's expectations too high; no matter how good the financial report is, if it falls short of expectations next quarter, it will trigger a sharp decline. However, SanDisk's stock price has pulled back nearly 50% from its peak, largely digesting the market's panic over the storage cycle peaking.
Although SanDisk's last quarter performance exceeded expectations, its guidance for the next quarter was only $10.3 billion to $10.8 billion, below Wall Street's forecast of $11.16 billion. Plus, after the previous surge and accumulated a large amount of profit-taking, capital is so sensitive that at the slightest disturbance, they will concentrate on selling.
A deeper reason is business differentiation. Although AI-driven data centers are strong, consumer revenues such as smartphones and PCs plunged 32%, raising concerns that the traditional storage market is bottoming out and lagging.
However, SanDisk's fundamentals remain intact, and enterprise-level demand remains. After nearly a halfway retracement, risk has been released relatively well, leaving limited room for sharp declines, but in the short term, sentiment fluctuations will still follow.
#闪迪财报双超预期, an additional $14 billion repurchase authorization was added After the earnings season from July to August 2026, the US storage sector has once again become the focus of market attention. From Seagate and Western Digital to the entire AI industry chain, a common trend can be observed: investment in AI infrastructure continues, demand for enterprise-level storage remains strong, and industry prosperity has not changed due to short-term market fluctuations. However, for investors, the more important question is no longer "whether the industry has opportunities," but "whether the price can continue to rise in the next six months." 1. AI capital spending remains the biggest driving force for the future. Over the past month, US tech giants have successively released their latest earnings reports. Microsoft, Meta, Google, and Amazon all reiterated in their earnings reports that they will continue to expand AI infrastructure investments in the future. The market is not only focused on GPU procurement scale, but more importantly, the pace of data center construction has not slowed down. An AI data center is not just about GPUs. Servers require a CPU, GPU, network switching equipment, memory, enterprise-grade SSDs, and large-capacity enterprise-grade hard drives all together. In other words, as long as major global cloud providers continue building AI data centers, the entire storage industry chain will continue to benefit. At present, there has been no significant reduction in capital expenditure plans by companies, which means that enterprise storage demand is expected to remain high in the coming quarters. 2. Enterprise Storage Will Remain the Fastest-Growing Market In recent years, the consumer electronics market has performed relatively weakly, with limited recovery in PC and smartphone sales微软首次在披露文件中量化了OpenAI对其AI业务的收入贡献，数据让市场重新审视这家科技巨头的AI增长质量。 截至2026年6月的财年，微软从OpenAI获得的销售额达到241亿美元。根据微软CEO纳德拉此前披露的AI业务年化收入370亿美元（截至3月季度），OpenAI的贡献超过微软实际AI收入的一半，预计占比约70%。 根据合作协议，OpenAI向微软支付算力费用、AI模型开发相关成本，以及部分收入分成。这使得双方关系既包含云计算服务采购、研发成本分担，也包含股权回报——并非单纯的客户与供应商关系。 微软的应对与市场反应 面对对单一来源的依赖质疑，微软已尝试通过投资Anthropic、开发自有模型来分散AI收入来源。但此次披露表明，这些努力尚未形成足够的分散效果。 KeyBanc分析师Jackson Ader指出，投资者需明确OpenAI贡献中云计算服务收入与收入分成的比例——服务收入体现的是商业交付能力，而投资收益更接近财务安排。 分析师观点出现分化：有观点认为单一客户集中度使微软AI增长承压；也有分析指出，与微软整体营收相比，OpenAI占比不足10%，年度新增商业预订中约51#黄金重返4200美元, why hasn't BTC risen in line with the rise?
This is a bit of a joke, right? Are these two things in the same track? When Bitcoin was $120,000, it didn't enter the gold rally either. Bitcoin's smaller brother Ethereum, gold's little brother silver, let's go head-to-head in 2v2
Let me get straight to the conclusion: gold rose because of rate cut expectations, while Bitcoin didn't follow because it's now classified as a risk asset by the market, and big money in the US keeps selling.
Let's first look at how this bullish candlestick for gold $XAU came to be. On August 6, ADP employment data added only 44,000, compared to an expected 75,000, and the previous value was revised down from 95,000. The market immediately turned its course: the dollar plunged, US Treasury yields fell accordingly, and after gold broke through the $4,200 resistance, short positions were closed and programmed orders were added, pulling up a single line. Combined with the People's Bank of China increasing its gold holdings for 20 consecutive months, this hidden line has been supporting the bottom.
Bitcoin $BTC is in an awkward position. Macro expectations for rate cuts are heating up and the US dollar weakening, which should have diverted funds into the market, but the Coinbase premium index has been negative for nearly 80 consecutive days, indicating continued selling by US institutions, with Asian buying just in place, and the two forces are topping around 64,000. On August 6, nearly $200 million was seen in BlackRock ETF inflows, unable to push the price down, showing the selling pressure.
A deeper issue is that Bitcoin's pricing logic has changed. "Digital gold" is just a metaphor; after institutions enter, it is more regarded as a high-volatility risk asset, with a much higher correlation with the Nasdaq than gold. Although the Fed's data is weak, officials remain hawkish and the timetable for rate cuts is unclear. Bitcoin, a product that requires clear liquidity signals to surge, can only trade sideways and wait for news.
Currently, $65,000 is a hard hurdle. If volume surges, only macro positive news can turn into buying; If it can't rise, just keep holding out. Both tracks are going their separate ways, with no signs of closure in the short term. Gold Surges While $BTC Waits for Its Next Catalyst
Global markets are sending a clear message: gold continues testing the historic $4,200/oz level, while $BTC remains range-bound despite strengthening long-term fundamentals. The divergence suggests investors are still favoring defensive assets before rotating back into higher-risk markets.
Gold's strength is being supported by ongoing geopolitical uncertainty, sustained central bank buying, and growing expectations that the Federal Reserve will gradually adopt a more accommodative monetary stance. At the same time, elevated U.S. Treasury yields and a resilient U.S. dollar continue to cap further upside, leaving gold near a key resistance zone.
Meanwhile, Bitcoin continues to consolidate. Institutional adoption, expanding participation in digital assets, and steady demand for spot ETFs remain constructive long-term drivers. However, the market still lacks a clear catalyst capable of attracting fresh capital and triggering a sustained breakout.
Liquidity across the crypto market also remains below previous expansion cycles. Trading volumes have yet to fully recover as investors wait for key U.S. macroeconomic data—including CPI, PPI, and upcoming Federal Reserve commentary—before increasing risk exposure.
Historically, periods of gold outperformance have often coincided with risk-off market sentiment. As macroeconomic conditions improve, Treasury yields ease, and the U.S. dollar weakens, capital has frequently rotated from traditional safe-haven assets into cryptocurrencies, providing support for both $BTC and $ETH .
Looking ahead, investors should closely watch three key developments:
Whether gold can establish support above $4,200/oz.
Whether $BTC can break out of its consolidation range on rising trading volume.
Whether ETF inflows, moderating inflation, and a more dovish Federal Reserve improve overall market risk appetite.
#MSTRSells1638BTC #Gold4200BTCStalls #TrumpTokenProbe #SandiskBeatAndBuyback #CircleArcLaunch #EarningsRealityCheck On August 6, about 912 million shares held by SpaceX employees and early investors were officially eligible for sale, more than double the previously publicly traded shares, and valued at about $100 billion at current prices. Musk himself is still subject to the individual lock-up period, but early shareholders such as Founders Fund, Craft Ventures, Alphabet, and some employees may choose to cash out.
After the earnings report, it has already fallen for a round, and now with early-stage investors and employees cashing out, the short-term stock price is indeed unlikely to relax.
This doesn't mean the stories of Starlink and Starship suddenly collapsed; it's simply that valuations are too high, spending is too high, and the chips have suddenly increased.
I have long had faith in SpaceX, but I really don't dare to insist on this kind of unlock. The dream is to go to Mars, but the stock price may first experience a free fall. Look for opportunities to enter 👀
$SPCX#Circle财报后押注Arc, can USDC experience new growth?
Circle's Q2 earnings report belongs to the category of "clear direction."
Total revenue was 701 million, up 7% year-on-year, but Wall Street expected 713 million, just a bit short. Net profit was 48 million, compared to a loss of 480 million in the same period last year, marking a significant turnaround. Adjusted EBITDA was 143 million, up 8% year-on-year. USDC circulating supply was 73.3 billion, up 19% year-on-year but down 4.8% quarter-on-quarter, with a market share of 27%.
In plain terms—the ability to make money is improving, but the growth rate makes the market a bit uncertain.
It once rose over 8% before the market opened, then turned downward at the open, ultimately dropping about 2%-3%. This scene is especially familiar this year—the fundamentals are solid, but the market's demands are even higher.
What is truly worth watching is not the financial report itself, but the new story Circle is telling.
Arc mainnet will launch on September 16. The list of founding validator nodes is basically all Wall Street stars—BlackRock, DTCC, Visa, Mastercard, Standard Chartered, ICE. BlackRock plans to deploy the BUIDL fund on Arc, and DTCC will manage asset tokenization on Arc.
At the same time, Circle obtained the OCC federal trust bank license, becoming one of the first stablecoin issuers to hold a federal banking license. Meanwhile, CPN's annualized transaction volume reached 14.7 billion, up 76% quarter-on-quarter. Agent Stack has already run over 900 paid services, with 99.3% of x402 agent payments settled in USDC.
Simply put, Circle is shifting from being an "interest-bearing stablecoin issuer" to an "on-chain financial infrastructure platform."
The full-year guidance for other income doubled directly from 150-170 million to 310 million to 310 million to 330 million, mainly relying on 242 million from the Arc token presale. This 242 million is a one-time book income, but for the market, the fact that someone is paying for Arc's future is itself a signal.
There are still significant disagreements on Wall Street. Morgan Stanley cut the target price from 106 to 38, saying USDC's contraction exposed the vulnerability of reserve income. TD Cowen gave 82, optimistic about platform transformation.
One is 38, the other is 82, a difference of 44 yuan. This 44 yuan represents the market's pricing difference regarding whether Circle is a reserve bank or a financial infrastructure platform.
My judgment is: whether Arc can succeed will determine whether CRCL's valuation moves toward 38 or 82. If Arc can really get on-chain settlement from institutions like BlackRock, DTCC, and Visa, Circle won't just be a company that survives on US Treasury interest. But right now, Arc is still in the 'list, no data' stage. The mainnet will launch on September 16. Whether there will be real transactions and real institutions will be the real answer.
Whether USDC can achieve new growth doesn't depend on USDC itself, but on whether Arc can succeed. If Arc succeeds, USDC will go from being a "stablecoin in reserves" to a "stablecoin in the settlement network." The valuation logic of the two stories is completely different. Circle is betting big, and it's betting the company's entire future on Arc. Whether it can succeed will be answered after September 16.Derivatives market ⚡
The long-short game revealed by futures contracts
📉 Open interest (OI): CoinGlass data shows that $XRP open interest has dropped to about $2.25 billion, with leverage at its lowest level in six months. When price drops coincide with lower OI, it usually indicates that positions are being closed rather than aggressive short positions.
⚖️ Long-Short Ratio: $XRP The long-short ratio was 0.75 on Thursday, near the lowest level in over a month. A ratio below 1 indicates bearish market sentiment, with bears slightly prevailing.
💰 Funding rate: $XRP The funding rate has dropped to 0.0013%, near negative territory. A neutrally biased rate indicates a lack of strong one-sided conviction in the market.
📊 CVD indicator: The 90-day Taker cumulative volume Delta remains neutral, indicating passive absorption rather than active buying or selling.
💡 Derivatives conclusion: The current market is in a "leverage reset" phase, not panic selling. Traders are reducing exposure, and large-scale forced liquidations are not obvious. This pattern usually means a lack of catalysts to stimulate sharp rises and falls.
#闪迪财报双超预期, $14 billion new buyback authorization #财报观察员: Mixed results, lock-up lifting approaching! What do you think about SpaceX's future? #ADP就业降温, the Fed's policy divergence has intensified 黄金重返4200美元，BTC为何没跟涨？数字黄金叙事短期失灵？黄金突破4200，白银破62，BTC却卡在6.5万——避险资金不买账？
8月5日现货黄金盘中突破 4200美元，COMEX期货收于 4245.8美元，白银同步升破 62美元。ADP数据疲软推动美元和美债收益率回落，为贵金属提供支撑。
比特币仍徘徊在 6.4万~6.5万美元 区间，未出现同步突破。本轮资金更多在交易 传统避险逻辑和利率预期，而非“数字黄金”叙事。在真正的宏观风险面前，市场仍将BTC视为 “高风险科技股”，而非避险资产。黄金涨它的，BTC等它的——非农之前，资金更愿拥抱“老牌避风港”。
#黄金重返4200美元，BTC为何没跟涨？ #内存卖方市场延续, can the Korean stock market see a turnaround?
Recently, the Korean semiconductor sector has undergone significant adjustments, with Samsung and SK Hynix $SKHYNIX share prices under continued pressure, and two voices have emerged in the market:
One believes this signals that the AI storage cycle has peaked;
Another view is that this is an opportunity after an over-release of emotions.
I prefer the latter, but the premise is: don't ignore cycle risk.
From industry data, the global DRAM market remains relatively tight. Demand for AI servers continues to grow, especially HBM (High Bandwidth Storage), which has become an important part of AI infrastructure like Nvidia, with Samsung, SK Hynix, and Micron still holding the main market share.
Goldman Sachs recently also believes that Korean chip stocks have seen significant corrections before, and the market may have overpriced in competition and cyclical risks. At the same time, this memory cycle may be longer than before.
Why?
Because previous storage cycles were mainly driven by PC and phone demand, but the biggest variable this round is AI.
The demand for high-performance memory in AI data centers is completely different from that of traditional consumer electronics. Previously, the storage industry often experienced cycles of "demand rising→ expansion→ oversupply→ and plummeting prices."
But now the problem is:
The new demand brought by AI may be changing this cycle.
However, I wouldn't simply assume that a drop in Korean chip stocks is an opportunity. Because what the market really worries about is not whether there is demand now, but whether the competitive landscape will change in the coming years.
After Changxin Technology's IPO, the global DRAM market gradually shifted from a "three-way competition" to new variables, and the market began to reassess the future profit margins of Samsung, SK Hynix, and Micron.
Therefore, when investing in the storage sector now, you shouldn't just look at short-term fluctuations; instead, pay attention to three core indicators:
First, whether HBM orders will continue to grow;
Second, whether AI capital expenditure remains high;
Third, can the rise in storage prices truly translate into corporate profits?
For me, I am currently more focused on the segments of the industry chain that truly have technical barriers.
Storage manufacturers benefit from the cycle, but equipment, materials, and core infrastructure also deserve attention, because regardless of future competitive changes, AI development requires sustained investment.
This adjustment in Korean chip stocks, in my view, is more like the market moving from "blind optimism" to "repricing"SpaceX's earnings day was a big roller coaster. 🫥
Before the first earnings report after listing on August 4, the stock price first rose 9.43% to close at $125.33. After the report was released, it immediately plunged over 8%, then fell another 13.61% on August 5, closing at $108.27, a new closing low since listing, with a trading volume of 208 million shares throughout the day.
Starlink posted a quarterly operating profit of $1.66 billion, which is actually a decent figure, but it couldn't withstand the pressure of unlocking 900 million shares ahead of time. Coupled with the official announcement of a joint venture with NVIDIA to develop Starmind AI satellite computing payloads, data center-level computing power is about to soar, and expectations are being fully smashed first.
But the positive news turned negative, with nearly $200 billion evaporating in one day. Its market value now stands at $1.43 trillion, more than the total market value of many listed companies.
#财报观察员: Mixed results, lifting restrictions imminent! What do you think about SpaceX's future? &P 500 Sets Another Record High: Selling Now to Wait for a Pullback Could Be August’s Biggest Mistake
Wednesday, August 5, 2026 | Q3 · Issue 93
Aspirin – A Data Scientist’s Perspective on Cyclical Analysis
The S&P 500 closed at a new all-time high of 7,736.52, gaining 1.8% on the day and approximately 13% year-to-date. Despite the strong rally, I neither chased the upside nor reduced my core index exposure. While the possibility of a second-half correction remains, even a meaningful pullback could still leave the index trading above current levels.
Bitcoin continues to consolidate around $64.4K. For now, I’m keeping fresh capital in reserve for a potential S&P correction and have only two BTC contract alerts in place at $63.5K and $69.2K.
A 10% Correction Doesn't Guarantee a Better Buying Opportunity#SandiskBeatAndBuyback #CircleArcLaunch #EarningsRealityCheck 美光 $MU 正处在存储周期向AI基础设施重新定价的枢轴点，核心矛盾在于美股流动性收紧预期与高端HBM结构性供需紧缺之间的博弈。
在跨市场资金流动中，美元指数与美债收益率的高位震荡压制美股科技板块整体估值弹性，而黄金与加密资产在避险与流动性分配中持续分流宏观风险偏好资金。存储龙头 $MU 的定价逻辑正在脱离传统消费电子周期，转由AI算力资本开支与HBM3e/DDR5出货量所驱动。
驱动因素排序上，AI数据中心对HBM与DDR5的确定性需求挤占高阶产能占据主导地位，NAND闪存库存出清带来的资产负债表修复次之，宏观利率环境对成长股折现率的挤压影响相对滞后。当美股科技巨头算力资本支出维持高位，存储环节的溢价能力便优先于大盘系统性估值提升。
上行剧本需满足美股科技股资本开支确认增量，且美债收益率下行释放科技股估值压力。若美光在高端HBM市场份额与出货结构超预期，资金将加快从BTC等风险资产向美股AI基础设施核心标的集中，带动 $MU 完成从周期股向AI算力基础设施的估值重构。触发条件为企业级存储订单持续追加，失效信号为宏观流动性紧缩导致科技巨头大幅削减AI资本开支。
下行剧本源于美联储维持高利率导致美元强势，美股高估值AI板块遭遇系统性获利了结，资金转向黄金避险或加密资产阶段性轮动。若主要存储厂商扩张产能过快破坏供需均衡，或下游AI推理需求转化不及预期，存储行业盈利将再度承压。触发信号为存储均价增速放缓或下游客户库存积压，失效信号为供给端再次达成协同减产。
未来7天需重点观察美债收益率变化、美元指数波动对美股科技板块资金流向的影响，以及 $MU 在高端存储领域的出货结构与盈利兑现节奏。
#CLARITY法案推进受阻，参议院分歧扩大 #财报观察员：业绩喜忧参半，解禁将至！SpaceX后续怎么看？ADP employment hits a six-month low, and the probability of a rate hike in September still exceeds 50%? Nonfarm payrolls + CPI are key to setting the tone
US July ADP private employment increased by 44,000, well below the expected 75,000 and the lowest in six months. The job market has cooled significantly, but it has not reversed the divergence in the Fed's policy path.
Where are the market disagreements? One side believes that the slowdown in employment weakens the reasons for further rate hikes, easing the pressure of policy tightening. Another side warns: inflation risks remain high, and CME data shows the probability of a 25bp rate hike in September is still above 50%.
The key variable in the crypto market is no longer just employment data, but "whether cooling employment can overcome inflationary pressures." This Friday's nonfarm payrolls and next week's CPI will jointly determine the September interest rate pricing.
ADP is just the appetizer; non-farm payrolls are the main course. If employment remains weak but inflation remains stubborn, the Fed will still be unwilling to relax.
#ADP就业降温, the Fed's policy divergence has intensified #Circle财报后押注Arc，USDC能否迎来新增长？
Circle这份财报，数字本身不是重点。
总营收7.01亿，同比增7%，略低于预期。USDC平均流通量同比增了25%，但季末只剩733亿，比一季度末少了4.8%。这说明利息收入还在撑着利润，但真正在生态里流转的稳定币反而在缩。营收增长靠的是加息环境，不是规模扩张，这是两份不同的答卷。
真正值得看的是财报之外的那个东西——Arc。
这个机构级区块链平台已经进了private mainnet阶段，9月16号上线公共主网。贝莱德、DTCC、Visa、万事达这些全球金融基础设施的核心玩家，全在创始验证者名单里。Arc这套东西，本质上是把稳定币结算、代币化资产和机构金融基础设施打包成一套服务，让华尔街直接在链上做清算、结算、资产交割。
如果Arc跑通了，USDC就不只是散户在交易所里买买卖卖的稳定币，而是机构间结算和代币化资产流转的基础货币。这才是真正的增长引擎——量级不在一个维度上。
我的看法很简单，短期流通量下降是现实，但Arc打开的赛道，比零售市场的天花板高得多。Circle现在押注的不是USDC的季度流通量，是华尔街的底层清算系统。这才是值得长期看的东西。
$BTC $ETH $BICO In the previous trading day, the S&P corresponding to $SPY only dipped slightly, the Nasdaq weakened, and the Dow continued to close at a new high. Even more interestingly, Brent crude oil hovered around $79, US Treasury yields edged down, and the VIX fell back to 15.81, indicating that market panic hadn't risen. The real change was that tech stocks themselves started picking stocks.
Today, I will treat $SPY as a thermometer of overall risk appetite, and $QQQ as a thermometer of AI and big tech crowding. As long as the $SPY is stable and the $QQQ weak, it means funds are not leaving US stocks but rebalancing from high-valuation technology sectors.
Personal observation: This is not a bear market signal; it seems more like the market at a high point is starting to reallocate chips. Non-investment advice.@币圈超短王马大帅 2026.08.06 午间币圈快讯完整版#闪迪财报双超预期，新增140亿美元回购授权
一、盘面整体行情（截至8月6日12:30）
1. 加密总市值2.29万亿美元，24小时涨幅+0.9%，现货成交量570亿美元，市场震荡修复，大饼强势、山寨分化明显
2. BTC比特币：现价64516美元，24H+0.3%
短线支撑63800，第一压力64800、强压力67000–68000；地缘避险降温后依托支撑走出空头回补反弹，比特币市场占有率56.5%持续走高
3. ETH以太坊：现价1897美元，24H+1.25%，走势强于BTC
支撑1850、压力1920，链上借贷、DeFi板块异动带动独立买盘，结束长期联动弱势行情
4. 主流山寨分化
走强：SOL、TRX、BNB；走弱：XRP、ADA、DOGE、LINK整体偏弱；DeFi赛道板块日内跌幅最高38.7%，稳定币资金流转缩减5%，资金集中流向BTC、ETH两大核心币种
5. 市场情绪：恐惧贪婪指数小幅修复，脱离极致恐慌区间，偏向谨慎观望
二、全网合约爆仓数据（24小时）
全网合约清算总额2.35亿美元，以空头踩踏止损为主
• 空单爆仓1.42亿美金，多单爆仓0.93亿美金，本轮反弹由短线空头集中平仓驱动
• BTC合约爆仓约4100万美金，ETH合约爆仓3800万美金，无超大额单笔爆仓，属于中小合约集中止损行情
• 头部交易所爆仓量占比过半，短线合约资金博弈加剧
三、宏观重磅消息
1. 美英联合加密监管协同落地（核心政策）
美英财政部发布联合声明，双方统一稳定币监管框架、代币化金融监管标准，落地对接美国GENIUS稳定币法案，对合规机构资金入场形成中长期利好；短期无强监管利空落地，缓解市场政策焦虑。
2. 地缘利好支撑盘面
美方暂缓对伊朗军事打击，霍尔木兹海峡局势缓和，避险资金从黄金出逃，部分资金流入加密市场做短线风险资产博弈，是BTC企稳反弹核心诱因。
3. 美联储降息预期维持不变
CME利率期货定价9月降息概率87.5%，官员偏鸽发言延续，美债收益率小幅下行，纳指走高，风险资产整体环境偏暖。
4. 美国CLARITY加密法案再度推迟投票
议员以银行经营风险为由提出反对，法案落地概率降至31%，合规立法落地延后，短期削弱机构长线布局预期，但规避了短期强监管打压利空。
四、行业热点动态
1. Uniswap上线链上借贷产品「Earn」
DEX龙头正式跨界借贷业务，以太坊链上活跃度提升，直接助推ETH短线走强，Layer2生态活跃度同步回暖。
2. 印度更新加密税务规则
接入OECD全球加密资产报税框架，境内交易所强制申报用户加密交易盈利，东南亚散户合规成本抬升，土狗币种抛压增加。
3. 香港稳定币牌照传闻澄清
市场传言国庆发放第二批稳定币牌照，香港金管局官方辟谣，仅专注现有两家持牌机构运营测试，暂无新增牌照计划。
4. Robinhood英国FCA加密资质稳定运营，欧美零售交易渠道持续扩容。
五、盘面核心总结&观察要点
1. 行情逻辑：地缘缓和+空头回补+降息预期三重支撑反弹，但整体成交量没有放量，属于存量资金博弈，不适合高位追涨；
2. ETH迎来阶段性结构性走强，短线性价比优于比特币，山寨轮动极快，DeFi板块回调规避；
3. 关键时间窗口：本周美国CPI通胀数据，直接改变美联储降息预期，决定本轮反弹能否延续； Although the expansion of AI infrastructure is reshaping the memory semiconductor industry, market attention has already shifted beyond simple performance improvement to structural pricing power. On the surface, the logic behind the rise in $SNDK, $MU, and $SKHYNIX converges on 'expanding AI demand,' but the core price reflection lies in each company's dominance in the HBM and enterprise SSD markets. Does the current premium really match the pace of fundamental improvement? - The HBF Open Spec unveiled at the Flash Memory Summit is not just a simple technical collaboration, but an attempt to standardize industry standards to resolve storage bottlenecks in AI data centers. This creates the conditions for the establishment of the customized memory market, differentiating it from the competition of general-purpose products in the existing NAND market. - SK hynix is strengthening its price negotiation power by emphasizing its advantage in the HBM market, while Micron is expanding its share of DRAM for AI servers. SanDisk presents the rising demand for enterprise SSDs as a new growth engine. - AI Datasen for HyperscalersCircle二季度最亮眼的数据，不是营收，而是USDC的使用强度。
截至季末，USDC流通量为733亿美元，同比增长19%；季度链上交易额达到14.8万亿美元，同比增长151%。持有超过10美元USDC的有效钱包增至700万个，同比增长24%。
交易额增长速度远高于供应量，说明同一批USDC在链上的流转频率明显提高。但供给端还没有进入持续扩张：二季度共铸造830亿美元USDC，赎回870亿美元，净流出约40亿美元；季度平均流通量为765亿美元，季末降至733亿美元。到8月3日，流通量又回落至约720亿美元，较季末减少1.8%。
收入结构依然高度依赖利率。
Circle二季度总收入及储备收入为7.01亿美元，同比增长7%。其中储备收入6.68亿美元，占总收入约95.2%；其他收入只有3360万美元，占比不足5%。
USDC季度平均规模增长25%，储备收入却只增长5%，主要原因是储备收益率由去年同期下降66个基点至3.5%。利率下行后，发行量必须增长得更快，才能抵消单枚USDC创造的利息收入下降。
渠道成本同样不低。二季度分销、交易及其他成本达到4.12亿美元，相当于总收入的58.8%。扣除这些成本后，Circle留下2.89亿美元，利润率约41%，较去年提高3.02个百分点。
这也解释了Circle为何在财报后把Arc推到更重要的位置。
Arc计划于9月16日开放主网，目前已有超过100家机构和生态项目参与建设。首批外部验证者包括BlackRock、DTCC、Mastercard、Visa、ICE、渣打银行和MoneyGram等11家机构，USDC将直接作为网络Gas资产。
BlackRock计划把BUIDL部署到Arc，DTCC则准备在2027年下半年接入托管资产代币化。Circle Payments Network的年化交易量也已达到147亿美元，环比增长76%，加入网络的金融机构增至175家。
公司把2026年其他收入指引从1.5亿—1.7亿美元上调至3.1亿—3.3亿美元，接近翻倍。不过官方明确注明，这里面包含已经确认的ARC代币预售收入，不能全部当成稳定、可重复的业务增长。
Arc合作名单已经足够大，接下来要看的是实际资金：Arc上的USDC流通量、日均交易额、付费应用数量，以及非储备收入能否持续提高占比。
USDC要迎来下一轮增长，不能只靠美债利息。Arc需要证明，它能把稳定币从Circle的负债规模，变成真正产生手续费和服务收入的网络资产。
#Circle财报后押注Arc，USDC能否迎来新增长？ #闪迪财报双超预期, an additional $14 billion repurchase authorization was added
Many people saw SanDisk's earnings report exceeding expectations and their first reaction was: "Why is it still falling?" "
On the contrary, I feel this isn't positive news being realized, but rather that market expectations for AI have been raised too high.
No matter how much you earn this season, it only proves the past; What truly affects the stock price is whether it can continue to grow rapidly in the future. As long as next quarter's revenue guidance falls short of expectations, the market will reprice.
So, what makes me focus more on this financial report is not the daily rise and fall of the stock price, but whether the demand for AI storage is truly a long-term trend.
My answer is: I am still optimistic.
The reason is simple: AI models are getting larger, inference is increasing, and data center expansion hasn't stopped. As long as AI computing power continues to expand, the demand for high-performance storage will not disappear.
However, what truly deserves attention in the future is not whether memory chip prices will rise, but who can truly turn demand into profit.
Growing industry demand does not mean all companies can make money. Only companies with strong product competitiveness, high bargaining power, and continuously improving profitability can truly enjoy the benefits of AI.
Therefore, this financial report does not change my long-term judgment on the AI storage sector.
In the short term, the market may fluctuate repeatedly due to overly high expectations, but if the AI industry continues to develop, I prefer to view this adjustment as an emotional release rather than a trend end.
For me, AI storage remains a long-term track worth tracking, but the focus of investment has shifted from "whether there is demand" to "who can keep making money." @币圈超短王马大帅 2026.08.06 Full version of Midday Crypto News #SanDisk's earnings report both beat expectations, with $14 billion added for buyback authorization
1. Overall Market Trends (as of 12:30 on August 6)
1. Total crypto market capitalization is $2.29 trillion, with a 24-hour increase of +0.9%. Spot trading volume is $57 billion. The market is recovering from volatility, with strong Bitcoin and obvious differentiation among altcoins
2. BTC Bitcoin: Current price $64,516, 24H +0.3%
Short-term support is at 63,800, with first resistance at 64,800 and strong resistance at 67,000–68,000. After geopolitical risk aversion cooled, the market rebounded with bearish covering rebounding relying on support, and Bitcoin's market share of 56.5% continued to rise
3. ETH (Ethereum): Current price $1897, 24H +1.25%, showing stronger performance than BTC
Support at 1850, resistance at 1920, on-chain lending and DeFi sector movements have driven independent buying, ending the long-term weak linkage market
4. Differentiation among mainstream knockoffs
Strengthening: SOL, TRX, BNB; Weakening: XRP, ADA, DOGE, LINK overall weak; The DeFi sector saw a day-high drop of 38.7%, stablecoin fund flows shrank by 5%, with funds concentrated in BTC and ETH
5. Market sentiment: The Fear and Greed Index has slightly rebounded, moving out of the extreme panic range, leaning toward cautious observation
2. Full-Network Contract Liquidation Data (24 hours)
Total contract liquidation across the network was $235 million, mainly with short stamping stop-losses
• Short positions were liquidated at $142 million, and long positions were liquidated at $93 million. This rebound was driven by concentrated short-term short liquidations
• BTC contract liquidations amounted to about 41 million USD, ETH contracts 38 million USD, with no large single liquidations, indicating concentrated stop-loss conditions for small and medium contracts
• Liquidations on leading exchanges account for more than half, intensifying competition among short-term contract funds
3. Major macro news
1. US-UK Joint Crypto Regulation Coordinated Implementation (Core Policy)
The US and UK Treasury departments issued a joint statement, unifying the regulatory framework for stablecoins and regulatory standards for tokenized finance, implementing alignment with the US GENIUS Stablecoin Act, which will provide medium- to long-term benefits for compliant institutional funding; In the short term, the implementation of weak regulatory heads will ease market policy anxiety.
2. Favorable geopolitical conditions provide support
The U.S. has temporarily suspended military strikes on Iran, easing tensions in the Strait of Hormuz, and safe-haven funds fleeing from gold. Some funds have flowed into the crypto market to engage in short-term risk asset trading, which is the core driver for BTC's stabilization and rebound.
3. Fed rate cut expectations remain unchanged
CME rate futures price pricing at an 87.5% probability of a rate cut in September. Dovish officials' remarks continued, US Treasury yields edged down, the Nasdaq rose, and the overall risk asset environment was warm.
4. The vote on the U.S. CLARITY crypto bill has been postponed again
Lawmakers opposed the bill citing operational risks in banks, lowering the probability of its implementation to 31%. The delay in compliance legislation weakened institutions' long-term positioning expectations in the short term but avoided the negative side of strong regulatory pressure in the short term.
4. Industry Hot Topics
1. Uniswap launches on-chain lending product "Earn"
Leading DEXs have officially crossed over into lending business, with increased on-chain Ethereum activity, directly boosting ETH's short-term strength and boosting Layer 2 ecosystem activity.
2. India updates crypto tax rules
By integrating with the OECD global crypto asset tax reporting framework, domestic exchanges are required to declare users' crypto trading profits, raising compliance costs for retail investors in Southeast Asia and increasing selling pressure on Tautou coins.
3. Rumors of a Hong Kong stablecoin license clarified
Market rumors have claimed that a second batch of stablecoin licenses will be issued for National Day, but the Hong Kong Monetary Authority officially denied these rumors, stating that it is only focusing on operational testing with the two existing licensed institutions, and currently has no plans to add new licenses.
4. Robinhood UK FCA crypto qualification operates stably, with retail trading channels in Europe and the US continuously expanding.
5. Key Summary & Key Points for Market Observation
1. Market logic: Geopolitical easing + short covering + interest rate cut expectations triple support for a rebound, but overall trading volume has not increased, indicating a game among existing funds and not suitable for chasing rallies at high levels;
2. ETH is experiencing a phase of structural strength, with short-term cost-effectiveness better than Bitcoin. Counterfeit rotation is extremely fast, and the DeFi sector is avoiding pullbacks;
3. Key Time Window: This week's U.S. CPI inflation data directly changes expectations for Fed rate cuts and determines whether this rebound can continue; #ADP就业降温，联储政策分歧加剧
Bitunix分析师观点：美联储鹰派信号升温，全球资本成本迎来重估。
ADP就业走弱，但美联储官员依旧保留加息选项，经济降温不等于马上宽松。
美债供给持续释放，长端利率高位运行，推升全球资产折现成本。
霍尔木兹海峡谈判接近达成临时方案，但能源地缘风险并未完全出清。
加密市场ETF短期资金回流，但季度依旧大幅流出，机构整体态度保守。
BTC行情取决于美元流动性，ETF能否形成趋势流入是重要风向标。高估值高杠杆资产仍承压。
$BTC $ETH $SNDK How long does it take to go from loss to profit?
$SPCX opened at 110, pulled up to 130, then dropped back to 109 within 3 days.
And the battle for the bears has only just begun.
Tonight's unlocking will bring selling pressure, but it won't be full circulation.
910 million shares—what does full circulation mean? Currently, only 410 million are circulating.
No matter who buys, they can't keep up; the stock price will directly drop to 30-40.
Then the Mars project is basically scrapped, yet some still won't sell and hold long-term.
All IPO projects face selling pressure during their initial circulation.
Don't be misled by too much information; it takes time for the price to fall.
Tonight, once it reaches a certain price, just close the position and wait for the next opportunity.
#SandiskBeatAndBuyback #CircleArcLaunch #EarningsRealityCheck Circle’s next growth bet is no longer just more USDC. It is building the financial infrastructure around it.
Q2 revenue and reserve income reached $701M, up 7% YoY, while adjusted EBITDA rose 8% to $143M.
Average USDC circulation grew 25% YoY, but quarter-end circulation fell 4.8% QoQ to $73.3B. Onchain volume reached $14.8T, up 151% YoY but below Q1’s $21.5T, showing sequentially softer supply and activity.
Arc has entered private mainnet ahead of a Sep 16 public launch. More than 100 ecosystem and institutional builders are participating, alongside 11 third-party founding validators including BlackRock, DTCC, Visa, Mastercard and Standard Chartered.
Arc is designed around:
· USDC-denominated gas fees
· Sub-second settlement and stablecoin FX
· Payments, tokenized assets and institutional infrastructure
· Integration with Circle’s platform, including StableFX
Circle still depends heavily on income from USDC reserves, leaving performance sensitive to circulation and interest rates. By building infrastructure around USDC, Arc could help diversify Circle beyond reserve income if institutions use it for payments, FX and tokenized-asset settlement.
The institutional groundwork is expanding. Circle National Trust has final OCC approval, BNY has added USDC to its Digital Asset Custody platform, and Standard Chartered offers institutional access to USDC minting and redemption.
But recognizable validators do not guarantee adoption. After Sep 16, the key signals will be assets issued, settlement volume, active institutions and fee-generating activity on Arc.
What would signal real Arc adoption: renewed USDC supply growth or institutional settlement volume?
#CircleArcLaunch #EarningsRealityCheck #闪迪财报双超预期, $14 billion in new buyback authorizations. Micron's earnings exceeded expectations, but the market began to hesitate
Micron's financial report actually sends a very strong signal.
FY2026 Q4 revenue reached $8.97 billion, exceeding market expectations; adjusted EPS also significantly exceeded forecasts, and the company announced an additional $14 billion in stock repurchase authorization, indicating that demand for AI storage remains strong.
Interestingly, after the earnings report was released, the market did not simply respond with a "positive rise" but instead experienced a pullback.
The reason is that the market's focus is no longer on whether AI demand is present, but on whether AI demand can continue to exceed expectations.
Over the past year, HBM, high-end memory, and AI servers have driven rapid growth in the storage industry chain, with Micron, as a core supplier, directly benefiting. But once everyone agrees with the AI trend, the focus of capital transactions shifts from "future space" to "future fulfillment."
In short:
Previously, the market asked: Is there demand for AI?
Now the market asks: How much longer can AI growth continue?
This is also the stage many AI concept stocks are currently going through.
Good performance does not necessarily mean the stock price will rise, because the price reflects future expectations. If growth in the coming quarters falls short of market expectations, even with excellent financial reports, valuation adjustments may occur.
However, from an industry perspective, Micron's data still proves one thing: AI infrastructure construction continues, and storage remains a key link in this wave.
Next, the market's real focus is whether the AI investment cycle can continue to deepen. @OKX planet Behind the excitement lies a thorn: the dream of flash memory price hikes was shattered by a single sentence. 🫧 Have you noticed that every time a financial report comes out, the biggest fear isn't the ugly data, but the management suddenly putting away their smiles? On the surface, SanDisk's report card looks pretty good, with both revenue and profit passing the mark. But what really changed the market's mood were those few remarks about the future price rhythm of NAND—too cautious, so cautious that it was completely on the same wavelength as the previous market expectation of "AI storage going wild." After the market closed, it plunged directly, hitting a low of 1244, with a drop of over 12% at one point, and now it is barely holding on to 1304. Behind this lies an easily overlooked pricing logic: the market was not buying current performance but the slope of future price increases. Everyone is betting that flash memory prices will soar beyond expectations, igniting AI storage demand like a rocket. But the management stepped forward and said, 'Don't rush, take it slow.' This statement shattered expectations, and funds ran faster than anyone else. - The event transmission path is clear: the earnings report itself -> management guidance -> spot price expectations correction -> stock price repricing -> which affects risk appetite across the entire storage sector. - What's even more noteworthy is that this kind of "performance meets targets but provides conservative guidance" scenarios are often not isolated incidents. If other storage chain companies show similar sentiments next, the valuation anchor for the entire sector will have to move downward. On the positive side, the medium- and long-term demand logic for AI storage has not been overturned; only the pace has changedThe fundamentals were fully realized, and the coin price broke the bear market's lowest point
This is not a hypothesis; I bought it with real money.
In 2023, I spent a lot of time researching DeFi leaders. Protocol revenue is high, growth is rapid, and market value is falling cheaply.
Scoring each item within the framework of value investing: good business, good prices, wide moat—almost perfect marks.
After buying, I have high hopes for them. When prices rise, they think they have foresight; when they fall, they think the market is wrong. No matter how others laugh, I don't care. That's value investing—hold for the long term.
Years later, the fundamentals delivered on all their promises. Protocol revenue hit a new high, with the moat far ahead of the second place.
But the coins we hold don't make money. The price even fell below the lowest point of the 2022 bear market.
The fundamentals are all correct, but the prices are all wrong.
Some say: don't touch altcoins, only buy Bitcoin. This helps you avoid losses and misses out on every round of excess returns.
Some say: Tokens are just governance rights and have nothing to do with protocol revenue. This explains why it was wrong, but doesn't tell you what to do.
The continuous blows made me deeply reflect and determined to find the root cause
Later I realized that the main bull wave in crypto bull markets is determined by narrative, not by fundamental realization.
But the altcoins I used only looked at fundamentals and didn't consider narrative.
Every bull market has a main story, and none of these stories ultimately deliver on promise, but that doesn't stop them from driving the entire market cycle.
LTC is the cleanest specimen, having surged dozens of times in 2017 thanks to Bitcoin's "silver" narrative. An asset with almost zero value capture and zero real use still skyrocketed.
At that stage, no one cared about accounting issues like value capture, and it really didn't affect the price increase.
Because value capture determines the floor, not the ceiling.
It doesn't care how much is left after you crash, no matter how high you can rise. Otherwise, these DeFi leaders wouldn't have missed new highs.
Narrative is the router of capital. It doesn't create value; it points money off-market to a direction of inflow. When money flows in, prices rise. As for whether the story will be fulfilled ten years later, the bull market won't wait for that day, nor will it need to.
Therefore, altcoins should be secured during bull markets, not held long-term.
Fundamentals are not the engine of a bull market; narrative isFundamental Research Report $EGLD / MultiversX (Public Chain/L1) $3.20
Essentially: MultiversX ($EGLD) has a comprehensive score of 57/100, with a narrative rating focused on implementation. Looking at the three layers, the company team has cash reserves, the protocol network already shows signs of paid usage, and token capture has been implemented.
Fundamental breakdown: MultiversX (token $EGLD), public chain/L1 track. Focuses on AdaptiveState sharding. Benchmarks ETH and SOL. Traditional enterprise collaboration relies on cloud servers and contract reconciliation; during high concurrency, gas surges, TPS is limited, and cross-chain bridge security incidents occur frequently. Public chains use a unified state machine for trustless settlement, reducing reconciliation costs. Average order value is $50-500/month, requiring USDC or fiat currency settlement. Narrative-driven track, bear market usage cut by 60-80%. Positioned as an end-to-end vertical platform. Product launch: protocol layer officially operational, on-chain dashboard shows protocol fees accumulating, showing signs of paid usage. Latest version not found, 60 valid submissions in the past 90 days.
On the user side, address MAU is not disclosed, DAU is not disclosed, 24h transaction volume $80.00M, TVL not found. Wallet addresses do not equal monthly active individuals; concentrated holdings of large addresses overestimate actual user numbers. On the revenue side, user fees are undisclosed; supply-side revenue is about 80-90% of user fees (belonging to LPs and nodes), protocol treasury revenue is $2.00M, token holders buy and burn annualized without a burn mechanism. 24h transaction volume is business turnover, not revenue. Company profits do not equal protocol profits, protocol profits do not equal token holders profit. Code side: 60 valid submissions in 90 days, 25 active contributors, latest version not found. GitHub is A-level evidence that can be directly verified. Investment background: For company equity financing, look at PitchBook/Crunchbase (A-level); for token private and public funding, refer to whitepapers, release curves, and on-chain unlock contracts (A-level); market makers and ecosystem funding are B-level but do not represent long-term holdings by tech VCs; for technical integration, see API/SDK access evidence (B-level); strategic partnerships and logo walls are D-level. Using NVIDIA GPUs does not mean NVIDIA investment, and going public on exchanges does not mean 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 (accounting for +3.50% circulating volume), burn buyback annualized rate, no explicit buyback burn. Must you buy coins when using the product? Yes, strong value capture (Gas/collateral/service access). Looking at it together with peers (unified caliber, no cross-sector random comparison): In terms of circulating market cap, MultiversX $3.00B, ETH undisclosed, SOL not disclosed. FDV: MultiversX $4.20B, ETH undisclosed, SOL not disclosed. In terms of annualized revenue, MultiversX is $2.00M, ETH is not disclosed, SOL is not disclosed. For monthly active addresses or users, MultiversX is not disclosed, ETH is not disclosed, SOL is not disclosed. Figures are based on public data snapshots; some omissions are supplemented by official self-reports or industry standards. Valuation: circulating market cap $3.00B, FDV $4.20B, P/S 1500.0x, FDV divided by revenue 2100.0x. Pessimistic: $3.00B is 50-70% off, fluctuating in a neutral range; optimistic outlook: revenue doubles, burns are implemented, enterprise clients are inflowing, FDV corresponds to P/S, aligned with the top companies. To sum up: Solid fundamentals (score 57/100). Token value capture has been implemented (buyback/burn/gas). Circulating market cap is relatively high relative to fundamentals, overdrawing expectations, and FDV is moderate. Three major risks: 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). Key points to look at next: protocol fee cycles, burn amount, active address retention, TVL/loan balances, GitHub version releases. The above is the logic and judgment based on public information and does not constitute buy or sell advice. Core financial indicators deviate by more than 30%, conclusions need to be re-evaluated.
Once the fundamentals are dismantled, how the market moves is another matter.
#基本面研报 #加密 #研究 #OKXOrbitSanDisk delivered a truly "explosive" financial report—revenue surged 372% year-on-year and 51% quarter-on-quarter, driven by strong demand for data centers and rising pricing. However, this report card failed to hold the post-market stock price. Why did the stock price fall despite explosive earnings? There are two core reasons: Options Competition and Earnings Competition Battle: Data for options expiring in the week of August 7 shows that the $1,370 strike price holds the highest open interest among both call and put options. After the earnings release, intense options competition became one of the factors putting short-term pressure on the stock price. Next quarter guidance falls short of expectations: SanDisk expects revenue of $10.3 billion to $10.8 billion next quarter, below market expectations. For AI storage leaders that often grow by over 100%, the market wants not just "good," but "better than expected." The logic behind AI storage still holds true. From a fundamental perspective, SanDisk's strategic direction is sound—shifting to a high-value customer mix (data center business grew 437%) and price increases continue to support performance. The market is still watching how long the storage price upward cycle will last, and whether SanDisk can maintain its pricing power amid the surge in AI data center demand and gradual supply release. $BTC $ETH $AI #闪迪财报前夕, HBF and storage shortages spark heated discussion. #临时通航协议待落地, oil price risks have yet to reverse. #财报观察员: Mixed results, restrictions imminent! What do you think about SpaceX's future? 别光看涨势，$APT这些数据更关键最后说几句实话 $APT涨了2.7%，评论区比我还惨，心理平衡了点。我多了$APT，0.6032进的，5倍，浮亏38.75U。 后市我怎么看布林带上轨0.6005，下轨0.5851，价格就在这里面折腾。 DIF在0.00这个位置，动能的事急不来。均线报0.5961，价格贴着它走，短线的命就系在这。 技术面给出的信号我个人一直觉得，每次说"这是最后一次"的人，下一次还会再来。 多数人都搞错了不追第三根阳线，这条纪律救过我好多次。 仓位管理这事，说起来简单，做起来全是人性。看不准的时候就小仓位试错，错了也伤不到筋骨。 周末的拉升水分大，量能跟不上就是耍流氓。睡前必检查一遍挂单，不带止损的单子我睡不着。 突破重要关口需要回踩确认，不回踩的突破大概率是假突破。复利是有的，但亏损 50% 需要 100% 才能回本。 羊群效应让人跟着大 V 喊单，但大 V 平仓时不会通知你。止损线我固定在成本下方8%，到了就走，不商量。 大饼横盘的时候，山寨就开始各演各的戏。挂单必带止损，裸奔的单子我一张都不留。 交易所维护前后，盘面容易出幺蛾子。计划你的交易，交易你的计划，老话#闪迪财报双超预期，新增140亿美元回购授权
After SanDisk's earnings report was released last night, the stock price fell steadily after hours.
Many people's first reaction was that the earnings missed expectations, but the truth is quite the opposite. FY2026 Q4 revenue was $8.97 billion, adjusted EPS was $39.25, both exceeding market expectations, and the company also added a $14 billion stock repurchase authorization. This report card is not bad.
The real issue lies in the next quarter's revenue guidance.
Although it is only slightly below market expectations, it is enough to make investors choose to take profits.
The market is becoming increasingly difficult to satisfy.
In the past, such an earnings report exceeding expectations would likely have driven the stock price higher. But now, the market is no longer trading on "how much was earned this quarter," but rather "whether more can be earned in the future."
The pricing logic of the AI sector is changing.
When growth for the next few years is already priced into the stock, beating earnings expectations is just the baseline. Only by continuously raising future expectations can higher valuations be supported. Once guidance fails to bring new surprises, even if fundamentals remain strong, investors will not hesitate to take profits.
Actually, this kind of trend is not unfamiliar.
Before good news is announced, investors rush to buy in advance, pricing all expectations into the stock; when the good news is finally announced, it becomes the moment to take profits.
Unfortunately, I am the one who went long at a high level and am still holding a position at a loss.
However, I have not changed my view because of this single down day.
The AI industry is still developing rapidly, computing power continues to expand, data volume keeps growing, and high-performance storage chips remain an important infrastructure in the entire AI industry chain. Short-term stock prices will be affected by sentiment and expectations, but long-term demand has not fundamentally changed.
Therefore, I prefer to see this decline as an adjustment after expectations were realized, rather than the end of the AI logic.
Over the next decade, I remain firmly optimistic about the AI industry and continue to be bullish on the storage chip sector for the long term. Oil prices plunged 12% in two days, and the Dow broke through 54,000 points—Bitcoin's "dual identity" has finally been understood by the market
Yesterday, did you see the news of the oil price crash?
Brent crude oil plunged nearly 12% over two days, breaking below the $80 mark, and on Wednesday in early Asian trading, it briefly touched $78.43 per barrel.
At the same time, the Dow surged over 1,000 points intraday, breaking through 54,000 for the first time, while the S&P 500 and Nasdaq hit record highs simultaneously.
What about Bitcoin? It rebounded from the low of $62,200, reaching a high above $65,000, firmly holding around $64,000.
Oil prices fall, stocks rise, Bitcoin follows suit—how long has it been since you saw this logic?
What happened?
The Strait of Hormuz, a vital artery carrying one-fifth of the world's oil transport, has been blockaded for over five months since the US-Israel surprise attack on Iran in February 2026.
Now, the turning point has arrived.
On August 4, U.S. Treasury Secretary Besent publicly stated: "We are negotiating with Iran, and it is very likely that an agreement will be reached today or tomorrow to open the strait." ”
The specific outline of the agreement is already clear: 60 days of temporary arrangements, 30 days of mine clearance, and no tolls during this period. Iranian Deputy Foreign Minister Ghalibabadi confirmed that the Iran-Iraq agreement is "close to finalization," and the strait will establish a "new mode of passage different from the past 60 years."
Trump himself said, "The result will be known within 48 hours." ”
The biggest landmine in global geopolitics is being defused.
The market's reaction says it all.
The transmission chain is as clear as could be:
Strait navigation expectations → oil prices plunge → easing inflation concerns → easing rate hike expectations → risk assets collectively celebrate
The Dow Jones rose nearly 1,000 points in two days, and the S&P 500 closed above 7,700 for the first time.
Gold has risen. Bitcoin also rose.
Here's a point many people haven't figured out—
For the past six months, Bitcoin has been traded as a "safe-haven asset." With war in the Middle East, Bitcoin is rising; High inflation is driving Bitcoin up.
But this time, the logic is completely reversed:
Geopolitical easing + falling oil prices + cooling inflation = risk appetite returning = Bitcoin rising alongside US stocks
Bitcoin rebounded from 62,200 to 65,000, an increase of nearly 2,800 points. It's not a safe haven; it's just following the rise.
This is Bitcoin's most alluring trait—
It is both a "digital gold" and a "risk asset."
When geopolitical conflicts escalate, it is gold—funds flowing into safe-haven markets.
When geopolitical tensions ease and risk appetite is rising, it is a tech stock—capital pouring in to chase gains.
Others have only one identity, but it has two. When others fall, it may rise; when others rise, it may rise too.
This is the power of dual storytelling.
But don't get too happy too soon.
Iranian Deputy Foreign Minister Ghalibabadi denied direct negotiations with the U.S., saying the agreement "has nothing to do with immediate opening of the strait," and that opening the strait depends on "whether the U.S. can correct its violations."
There were still internal divisions within the Revolutionary Guard, and the risk of attacks on cargo ships remained.
The formal agreement has not yet been signed.
If negotiations break down, there is room for oil prices to rebound, and risk assets may give back all their gains.
Finally, three honest words—
First: The biggest geopolitical shadow weighing on global risk assets over the past six months is now dissipating. Bitcoin shifted from a "safe-haven narrative" back to a "risk asset narrative," which actually opened up greater upside potential.
Second: $64,000 is an important support level. If it can hold and break through 65,000, the upper range could see 65,800-66,200. But if geopolitical tensions change, 62,000 will be the next critical line of defense.
Third: This round of rebound is essentially a technical recovery driven by geopolitical easing, not a full reboot of the bull market. Don't shout "bull return" just because it rose in two days, and don't cut losses because of one piece of bad news.
Information-driven markets aren't about who can see things accurately, but who runs fast. $BTC $BZ $CL #伊朗阿曼临时通航协议近落地 在最顶级的棋局里，落子前的第二十步就已经规划好将军路线。而今天，边线传来一个微妙的侧翼出子——万事达加密凭证正与 Borderless 试点跨境稳定币共享身份验证机制——这让我重新打开了中局的算度。
这一步棋，不是来争抢“运输货币”的小兵，而是把一枚“王翼贴身侍卫”放上了棋盘。万事达加密凭证提供治理与验证层，却不直接处理或结算资金。这正如一场比赛中，裁判不会替你走棋，但会认定每一步是否合法。跨境稳定币支付由此被装上了公开的行棋谱：对手每走一步，身份与风险信号都要通过标准化审批流程对齐谱库。这不再是“快零”的速攻，而是将整盘棋的合法性前置到开局的王前兵。
在特级大师的残局库里，信息比子力稀缺。多数人只看见支付速度与费用，我却看见这步棋把客户身份识别与反洗钱的“私人笔记”升级成了全盘的“对局协议”。以后每一笔跨境稳定币支付，都像一次正式比赛中的“短易位”申请：王与车必须被裁判确认从未移动，资格核验通过后才能履行“将军”的潜力。参与机构把身份与风险数据纳入审批流程，本质上是把“冒名顶替”的布局陷阱从棋盘中彻底删除。
再看 $XUSAR 的市场联动。普通棋手会盯着这条新闻的日内线形，像初学者只盯着眼皮底下的中心兵。而我看到的是，当身份验证成为跨境支付的强制开局库，真正的战略点在于那个既能接入合规层又能触碰资金流的连接位置。$XUSAR 如果立在“验证层”与“结算层”之间的象位上，它的每一步斜线都会同时控制更多格子。月级影响周期意味着这不是一次闪击，而是棋谱结构的替换。传统金融机构把手伸进稳定币棋盘，并不是为了吃两个兵，而是要改变“行棋资格”的定义方式。
这让我想起古典棋局里的封锁策略：先控制底线所有关键格，再逼迫对手走出一个不义的边兵。现在万事达就是那个控制底线的裁判棋，它不需要亲自持有资金，它只要握着“合规信号”这把王翼钥匙，就足以决定哪些棋手有资格进入中局。那些倚赖信息差和匿名路径赚取价差的棋手，将像过时的早期弃兵一样被现代理论放弃。
真正的大师此刻不会忙着移动王后。因为这一手棋最深的意图藏在它的时间戳里：当验证成为支付前的第一步，所有后续棋谱都会留下可回溯的足迹。残局阶段的“升变”权利，将不再只取决于棋盘上的剩余子力，而取决于你是否通过了那份标准化的身份凭证。
钟声滴答。传统金融的深远易位已经启动，棋盘上所有暗流都在重新排列。那些看不见的边缘格，正在被慢慢染上颜色——这不是中局前的试探，而是残局提前到来的信号。
#影响周期·月级 #传统金融·稳定币支付 #Mastercard·Borderless·身份验证Revenue surged 372%, 14 billion yuan in buybacks—SanDisk gave everything it could, yet the stock price still crashed 8%. Last night, SanDisk released its Q4 financial report—revenue of $8.97 billion, up 372% year-on-year. Adjusted EPS was $39.25, up 135 times from $0.29 a year ago. Gross margin was 84.6%, up from 26.2% a year ago, tripling in size.
The board made a bold move: an additional $14 billion buyback, bringing the total remaining buyback authorization to $15.5 billion.
"This financial report is just too explosive, isn't it?" Buy it! ”
When I woke up, SanDisk had dropped 8% after the session.
He was completely stunned again.
"No, such impressive performance—what makes it possible?"
Why? You still look at this market with outdated perspectives.
After the U.S. market closed on August 5, SanDisk and Western Digital both released earnings reports.
SanDisk's data center business—revenue of $2.98 billion, up 1298% year-over-year. You read that right, it's 1298%, not 12.98%.
Edge computing reached 5.43 billion, a year-on-year increase of 392%. Annual revenue was 20.25 billion yuan, a year-on-year increase of 175%. Eight long-term NBM agreements, with guaranteed minimum revenue of $93.9 billion and customer default coverage of $16.5 billion. More than half of the capacity for fiscal year 2027 has been locked in ahead of schedule, and two-thirds for fiscal year 2028 has been scheduled.
The CEO said during the call: "The growth rate of AI storage demand has already outpaced our supply capacity, and even after the largest customer signed contracts, they are still placing additional orders. ”
Even with all the capacity sold, customers are still chasing after more orders.
In any normal market, wouldn't this stock price skyrocket?
But the market doesn't see it that way.
SanDisk expects revenue next quarter to be between $10.3 billion and $10.8 billion, with a median of $10.55 billion, nearly 5.5% below analysts' forecast of $11.16 billion.
EPS guidance is $44 to $46, with the market expectation of $45.58—just below expectations.
Gross margin guidance is 83% to 85%, roughly flat compared to this quarter's 84.6%, showing signs of peaking. The market follows only one logic: you were good in the past, but can the future be better?
SanDisk's answer was: Yes, but not as good as you think.
After the stock price rose 468%, the earnings report beating expectations was no longer enough.
So far this year, SanDisk's stock price has risen nearly fivefold, while the S&P 500 has only gained 13%.
A 468% increase has already exhausted the story of the "AI storage demand explosion."
Now, what the market wants is not "how much money you have made"—what the market wants is "how much more money you can make."
If the guidance is even a little weak, the stock price will crash for you.
Western Digital released its earnings report on the same day—revenue of 3.75 billion yuan, earnings of 3.56 beating expectations, and guidance also beating forecasts.
And what happened? It fell 11% in after-hours trading.
Both families collapsed together.
What does this indicate? This shows it's not SanDisk's problem alone—it's the entire storage sector, and the market is repricing.
Goldman Sachs, JPMorgan, and a bunch of institutions gave "strong buy" orders, with an average target price of $2,400. But the stock price has already fallen 40% from its June high.
Expectations were set so high that even the earnings that exceeded expectations couldn't be filled.
The market isn't trading what you do.
It's about trading—what else can you do?
SanDisk gave everything it could: explosive performance, 14 billion yuan buybacks, eight long-term contracts locked in for the next four years, and a guaranteed minimum income of 93.9 billion yuan.
But the market is focused on that one sentence: next quarter's guidance is just a little lacking.
So 8% is gone.
This isn't SanDisk's fault; it's a blow that every stock pushed to the skies by AI stories will eventually face. Is it the market that's too harsh, or is the valuation just too high?$META is one of the most "squid game" companies I've seen in Silicon Valley. I'm curious if you really have friends who truly believe in it from the bottom of their hearts.
No matter where you eat, as long as it's a Chinese restaurant, you can always hear a Meta employee cursing at the next table, including but not limited to: Hunan Impression, Xiaoxiang Beef Bone Rice Noodles, Tang Bar, Yang Guofu
A few days ago, I chatted with a friend at Meta, and their entire group was transferred to the data annotation team. The daily task is to chat with AI, constructing challenging enough tasks for the AI to solve, and the company then processes their solutions to train the model. The company uses employees chatting with AI to assess the number of tasks and their specific difficulty each day
Meta now is worlds apart from the Meta I had four years ago, and the turning point should come in 2023.
Will Meta succeed? I don't know, but the company's culture and management's lack of clarity in the North Star indicator clearly aren't on my investment radarPersonally, I am very optimistic about $sive's performance in the second half of the year, especially after other companies' earnings exceed expectations
Personally, I believe the year-end stock price range is between $10-15, and if it's $15 per share, the market cap will reach $5 billion, about 30% higher than before this year (SEK 110/~$3.9B). This is the "top break and breakout of previous highs," and my reasoning is simple and divided into six points:
First: In fact, this company has already proven it can reach ~3.9 billion USD this year, relying on fewer collaborations and less news (mostly our guesses). So the SEK 110's previous high this year mainly came from a GF cooperation in June. From now until the end of the year, the catalysts are stacked (8/27 earnings report + US listing + Q3 full optical chain earnings report + possible new design introduction). Using "more catalysts" to surpass "the high set by fewer catalysts" is the most basic foundation of this logic. At the same time, the subsequent benefits and more customers from $GFS cooperation will give us greater opportunities.
Second: Dual listing on Nasdaq in the US = a shift in valuation structure. This is the real engine that can push $12 to $15+. SIVE is now being overshadowed by the "Stockholm small-cap" valuation framework; Once it goes public in the US, it is thrown into the US AI optical module/connector basket, alongside CRDO, LITE, COHR, AAOI, POET, and other "AI light" targets, and the valuation multiplier for "pure AI narrative players" in this basket is much higher than in Sweden. Listing is not simply about listing elsewhere; it is about connecting it into a "capital pool willing to pay high multiples for AI optics." Of course, this won't happen immediately, but as long as we hear more progress, it will be a catalyst.
Third: Four-layer cross-validation, reducing the "story" to a "low-risk story," so the market dares to offer higher multiples. This is exactly what makes my article most worth reading: AXTI (scheduled until 2027), AEHR (FY27 triple), AAOI/COHR/LITE (explosive), GF (silicon photonics doubling)—four independent financial reports verify SIVE's TAM and timeline. When a pre-revenue story is backed by the entire industry chain, the market's "belief discount" narrows and the willingness to pay multiples increases. Confirmed TAM = acceptable higher multiple = higher market value.
Fourth point: TAM's narrative math makes the $5 billion market cap make sense. GF says the pluggable optical market will be about $25 billion by 2030, and SIVE is an external laser pure player designed by GF SCALE + Jabil 1.6T. Giving it a $5B market cap is only about 0.2 times TAM's 2030 value, or a reasonable share of the laser segment—in a booming AI optical market, this is a very sellable story.
Fourth point: High beta + small-cap + private placement have already been implemented, naturally giving it the flexibility to "surge again."
Its range over the past year was SEK 2.85 → 110, with a ridiculously large beta; After the July private placement, the recent dilution pressure has been released. A small cap that can increase 3–4 times within weeks and has just cleared selling pressure, as long as the catalyst is strong, momentum overshooting fundamentals are the norm, not accidents.
Point six: Sector beta
If the entire AI light line continues to rip before year-end, SIVE will be the most elastic pure laser stock. As long as COHR/LITE/AAOI/CRDO/FN continue to guide upward in Q3/Q4, with 1.6T volume expansion and CPO becoming the consensus, as the "purest laser call option," SIVE will capture the most excess capital flow in this sector.
That's my view—the logic is clear. Let's see how the specific financial report goesKorean stocks have plummeted, but memory is still rising: Did chip stocks really get wrongly killed?
#内存卖方市场延续, can the Korean stock market see a turnaround?
In recent days, South Korean semiconductor stocks have fallen significantly, and Goldman Sachs has judged it as "excessive selling."
When people see the words "wrong kill," many people's first reaction is to buy the dip. But for now, I won't rush in to buy because of that one sentence.
Memory prices and chip stock prices are never in the same rhythm.
Global DRAM supply remains tight, and manufacturers still have bargaining power, indicating that business hasn't significantly worsened at the moment. But stock prices depend on whether they can still make money in half a year. Once capital starts to suspect that the most tight supply-demand phase has passed, even if spot prices are still rising, stocks may still decline.
This round is different from before.
AI servers support the demand for HBM and high-end DRAM, and manufacturers prefer to reserve capacity for high-profit products, so the supply of ordinary memory naturally won't suddenly become loose.
Therefore, I don't quite agree with the idea that "the memory cycle is about to peak." The cycle supported by AI demand is likely to be longer than the past driven solely by phones and computers.
But a longer cycle doesn't mean stocks are cheap now.
What really matters next is how long contract prices can hold firm, whether inventory will pile up again, and whether manufacturers will suddenly accelerate expansion.
If prices keep rising, inventory hasn't gotten out of control, and production expansion is being restrained, this round of decline feels more like sentiment-driven valuation killing. If supply gradually catches up, the so-called "wrong selling" now may just be a rebound midway through the decline.
So I won't rush to take this first cut.
Once Korean chip stocks stabilize and continuous capital inflows in, consider participating in batches. The most common disadvantage in semiconductors is assuming the stock price has bottomed out once the industry is still profitable.
This event also has reference value for the crypto world. If Korean stocks and chip stocks stabilize, it at least indicates that risk appetite in Asia is returning, which will help BTC and altcoin sentiment.
But if memory prices continue to rise while related stocks continue to be sold, it indicates that funds are actively moving away from highly volatile assets.
Instead of rushing to guess the truth, it's better to first see the money sold and when they're willing to pay it back.SKHYNIX showed a clear divergence in the storage sector today: SKHYNIX and SanDisk saw significant declines, while Micron remained relatively resilient and the gap widened. This differentiation usually corresponds to two situations:
First: Micron has already completed its valuation correction and is currently receiving effective support. Once peers finish catching up on the decline, it is expected to lead the sector's rebound;
The second type: resistance to decline appears on the surface, but in reality, the main force uses the liquidity advantage of leading stocks to create a "strong illusion," luring bottom-fishing funds into the market. After accumulating enough counterparts, they concentrate selling pressure to drive the market down.
The two scripts point to completely opposite operational strategies. Currently, there is a lack of sufficient signals to rule out either possibility, so further verification is needed based on subsequent trading volume and changes in order books. Personally, I remain cautious and not rushing to make a definitive judgment.
[The above views are for personal reference only and do not constitute investment advice]#闪迪财报双超预期, $14 billion new share buyback authorization. Sandisk released its Q4 2026 financial report after the U.S. market closed on August 5. Revenue and earnings per share significantly exceeded institutional consensus expectations. The board approved an additional $14 billion share buyback quota. Combined with existing quotas, the total repurchase authorization now reaches $15.5 billion, making it the biggest highlight of this report.
On the performance side, the company posted quarterly revenue of $8.96 billion, a year-on-year surge of 372%, with non-GAAP earnings per share of $39.25, significantly above the market expectation of around $33. AI computing power demand continues to drive the data center business, with segment revenue growing more than 12 times year-on-year. The shortage of NAND chips supports product pricing, pushing the quarterly gross margin up to 84.6%. The company has secured eight long-term supply agreements with leading cloud vendors, locked in a total of $93.9 billion in long-term revenue, and has locked in orders for more than half of its capacity in fiscal years 2027 and 2028 to smooth out storage cycle volatility risks.
The 10-billion-yuan buyback has boosted management's confidence, and ample free cash flow supports ongoing buybacks, which can dilute circulating shares and increase earnings per share in the long term. However, the stock price continued to decline after hours, with the core negative factor being a slowdown in revenue guidance for Q1 2027. The market is concerned that the storage price increase dividend is nearing its end, and that previous sector gains have been overdrawn.
Short-term buybacks can only provide bottom support and are unlikely to reverse the sentiment of capital liquidation; The medium- to long-term market depends on the progress of HBF technology implementation and the fulfillment of long-term contract orders, with the storage sector continuing its high-level volatility pattern. $BTC $ETH $SNDK The foundation is the final judgment seat of this building.
After twenty-six years of architectural design, I developed a professional habit: not looking at renderings, only checking slot inspection records. The CLARITY Act is set to be pushed to a Senate vote this week, the proposed rules for the GENIUS Act have been drafted, and will take effect in January 2027—to outsiders who only watch candlesticks, these are two legal terms; For me, there are two sets of "Unified Code for Building Structure Design" that change industry load standards.
The CLARITY Act is the market framework blueprint for the superstructure. It specifies which materials of columns can be used to climb towers and which beams cannot cross red lines, and the entire capital market structure is redefined as fire compartments like high-rise complexes. Advancing the motion procedure now is equivalent to passing the regulatory approval, but it is still some distance from the construction permit. What truly determines whether this building can stand is the GENIUS Act—stablecoin regulation is not just for facade decoration, but for the foundation base and prestressed pipe piles. There are no strict rules on reserves, leverage, or settlement paths, like using 12-diameter steel bars to support a 60-layer core tube—a gust of wind can tell something is about to happen.
January 2027 is the key point that structural engineers will pay close attention to. The completion of the proposed rules means the design institute has released an officially approved blueprint; By 2027, the implementation means that the 28-day cycle for concrete test blocks in the standard curing room must be completed. You watch the market rise today and fall tomorrow, but I focus on key dates and concealed project acceptance records in the construction log. For a real project, you can see whether it can be topped out at the pile foundation stage.
As for $XTSM, when I checked its overall plan, I saw the impact cycle was "monthly or above." What does it mean to be above the monthly grade? It is the stage for foundation pit excavation, cushion pouring, and full adhesion of the waterproof layer. For any ongoing project, in the past two weeks, regulatory statements have only completed the rebar tying inspection for the second-floor slab. If it hadn't previously embedded compliance, auditing, and independent custody at the bottom of the gravel cushion, then once CLARITY and GENIUS officially issue their execution orders, it would be a high-cost era of main structural reinforcement and renovation.
I've seen too many builders rush to the drawing review company before approval, because the core tube wasn't arranged according to load codes from the start. The white paper is the sales office's sand table; the true design intent is hidden in the reinforcement diagram of each shear wall. What Jeremy Allaire is doing now is calling the heads of each discipline into the meeting room, tapping the structural model, and saying: "The function can be discussed again, but the force system must be recalculated according to the final specification." This action is more qualified than any market analysis to be included in the construction organization design.
Before January 2027, all projects claiming to be stablecoin foundations will be revealed in the data of subsidence observation points. Whether the drawings behind the tower crane are real or not, the construction site is the most honest.
#影响周期· Monthly Grade and above #全球监管· U.S. Legislation #CLARITY· GENIUS #clarityactaug2026 January 2027