Orbit Post Sitemap
狗狗币是"机构最不待见的资产" 主流机构研报几乎从不覆盖DOGE，ETF申请名单里也没有它。这种"被机构无视"的状态，反而成了散户的共识标的？
机构不待见$DOGE 这事，你得反着看。高盛摩根的研报堆成山，比特币以太坊轮番被覆盖，DOGE连个脚注都混不上，ETF申请名单里更是常年缺席，仿佛这玩意儿不存在。
但恰恰是这种无视，把DOGE变成了散户自己的东西。BTC现在是大资金的玩具，涨不涨看ETF流进多少、看美联储脸色，散户拿着那零点几个币，更像在给机构抬轿子。
DOGE不一样，它没有机构定价权这回事，价格就是纯纯的散户情绪温度计，涨了是大家一起喊上去的，跌了也是大家一起跑出来的，至少规则公平，没有信息差收割。没人写研报意味着没有"目标价"这种紧箍咒，想象空间是敞开的，散户信的是社区、是梗、是十几年攒下来的群众基础，而不是华尔街那张盖章的纸。
更妙的是，机构不进场的后果是筹码结构干净，没有巨鲸式的贝莱德每天砸盘或增持吓你一跳，涨跌逻辑就剩一条：人心还在不在。
所以这种"被无视"不是DOGE的缺陷，是它的护城河——主流金融越看不起它，它就越属于草根，而当哪天机构真开始覆盖它的时候，反而要小心了，因为那通常意味着散户的便宜筹码已经被人惦记上了。Saylor嘴上说着“永不卖币”，身体倒是很诚实
STRC干到94.6了，51天新高。从6月底71左右的低点反弹了差不多30%
听着挺牛逼对吧？但你猜怎么做到的——
卖比特币做到的
7月底到8月初，Strategy卖了1638枚BTC，均价63,957刀。他们持仓成本多少？75,419刀。每枚亏1.1万刀在卖
卖了1个多亿美金，一半拿去付优先股股息，一半拿去回购折价的STRC。还顺便把美元现金储备干到了40亿
这套组合拳打下来——卖币护盘、回购拉股价、12%股息吊着——STRC确实从71被拉到了94
但你想过没有？
STRC反弹30%，BTC还在64000-65000晃，跟6月底差不多。Saylor自己在用BTC储备给自己的优先股做市。他说“跌破100美元绝不发新股”，意思就是：只要股价低于面值，他就卖币回购硬撑
有人说STRC涨是BTC要涨的信号，我觉得你想多了——他在护盘，不是BTC要起飞。STRC的目标是回到100美元面值，跟BTC短期走势关系没那么大
别把STRC的反弹当成BTC要起飞的信号
看BTC还是看BTC自己的结构——65000站不稳就别急着动手Using data to measure individual temperatures for encrypted "falling behind." This month, US stocks hit new highs every day, spot gold hit record highs, and both risk money and safe-haven money surged in; Only $BTC was near 64K, with a 24-hour increase of less than 0.2%, failing both risk-on and hedging logics. This is not randomness—when an asset does not react in both extreme environments, it usually means its marginal buying is being pulled away by other sectors. Flat prices do not mean safety; it means there is no new consensus. Don't fool yourself with the phrase 'sooner or later, catch up'—first ask why incremental funds are coming back. Data won't play along with you."The market doesn't punish bad earnings—it punishes fading expectations."
SanDisk just reminded everyone of that lesson in brutal fashion.
I opened my short around 1337, and with the stock now trading near 1212, the position is up more than 120 points. The thesis wasn't about weak numbers. In fact, the earnings report looked almost perfect:
• Revenue surged 372% year over year to $8.97 billion
• EPS exploded to $39.25
• Gross margin hit a record 84.6%
• Data-center revenue jumped 437%
• Another $14 billion buyback was approved
On paper, everything looked incredible.
So why did the stock collapse from 1480 to nearly 1210?
Because Wall Street doesn't price yesterday's victory—it prices tomorrow's growth.
The real problem was guidance.
Q1 fiscal 2027 revenue came in below expectations, profit growth is beginning to slow, and margins appear to be peaking. Suddenly, the market stopped asking, "How good was the last quarter?" and started asking, "Is this the best it gets?"
That's what triggered the sell-off.
How I'm managing the short
My stop-loss has been moved down to 1300, locking in profits even if we see a sharp rebound.
Profit-taking plan:
• Close 30% between 1150–1180
• Close another 30% between 1080–1100
• Exit the remaining 40% around 1000–1020
If price rebounds into 1250–1270 on weak volume, that's a potential area to add. If buyers reclaim 1300 with strength, the bearish thesis is invalidated.
Why I'm watching 1088 for longs
This isn't blind bottom-fishing.
The 1080–1100 zone sits near major technical support and a dense area of historical buying interest. The stock has already fallen nearly 58% from its highs, and short-term indicators are deeply oversold.
More importantly, the long-term story hasn't disappeared:
• AI storage demand remains strong
• Analysts still see significant upside
• 2026 capacity is reportedly sold out
• NAND prices continue to recover
If we reach 1088, I'll consider building a position gradually with limited leverage and strict risk management.
#DailyOrbit 用数据给加密的「掉队」量个体温。这个月美股天天创新高、现货黄金刷历史新高，冒险的钱和避险的钱同时在涌；唯独 $BTC 焊在 64K 附近，24 小时涨幅不到 0.2%，对 risk-on 和避险两套逻辑同时失灵。这不是随机——当一个资产在两种极端环境里都没反应，通常说明它的边际买盘正在被别的赛道抽走。价格横住不代表安全，代表没有新增共识。别用「迟早补涨」骗自己，先问增量资金凭什么回来。数据不会陪你演戏。$SNDK 闪迪利空带崩存储板块，美光深V反转，抄底时机来了吗？
闪迪这次财报看着赚钱不少，还拿钱回购自家股票，本来是好事。
可它对下个季度的业绩预期说的比较保守，机构觉得AI存储涨速要变慢了，纷纷卖出闪迪，直接把整个存储板块都带崩了。
美光一开始也跟着大盘恐慌下跌，不少人害怕跟着砸盘。$MU
跌到低位之后，一部分资金觉得HBM高端芯片是AI离不开的刚需，现在跌多了性价比很高，就进场抄底，尾盘直接拉回来走出深V行情。
不过这次上涨只是短线资金博反弹，并不是行情彻底反转。
现在整个存储行业，大家对后续能不能一直高速赚钱心里都没底，行情还是会来回波动。
总结
现在并不适合着急抄底。美光的回升只是情绪反弹，行业的顾虑还没消除。
最好等这次利空完全消化，订单和业绩确定还能稳步增长之后再动手，贸然进场还有继续下跌的风险。
$SKHYNIX #闪迪财报双超预期，新增140亿美元回购授权 #ADP就业降温，联储政策分歧加剧 #Polymarket洽谈10亿美元融资，估值超200亿美元 Let me use a thermometer for the credit market. Alphabet's latest ultra-large-scale bond issuance received about $115 billion in orders, more than four times the planned limit, and subscription enthusiasm even surpassed Amazon and SpaceX's recent AI-related bond issuances. The insight of this data is that: after the initial sell-off, institutions have reopened their wallets for 'debt pledged on AI infrastructure.' Credit spreads haven't widened, but are actually rushing to buy, indicating the market doesn't see AI capex as a bubble about to collapse—at least the debt side is still surviving. Money in risk assets often heats up first in the credit market, and this early signal is worth watching. Position size speaks for itself.Let me use a thermometer for the credit market. Alphabet's latest ultra-large-scale bond issuance received about $115 billion in orders, more than four times the planned limit, and subscription enthusiasm even surpassed Amazon and SpaceX's recent AI-related bond issuances. The insight of this data is that: after the initial sell-off, institutions have reopened their wallets for 'debt pledged on AI infrastructure.' Credit spreads haven't widened, but are actually rushing to buy, indicating the market doesn't see AI capex as a bubble about to collapse—at least the debt side is still surviving. Money in risk assets often heats up first in the credit market, and this early signal is worth watching. Position size speaks for itself.Miners and communities are the source of CORE's confidence in navigating bull and bear markets.
Millions of bit miners manage computing power to protect network security, while thousands of veteran miners have remained steadfast on duty for years, spontaneously spreading knowledge and maintaining order.
The vitality of decentralized public blockchains is not driven by capital speculation, but is maintained through long-term joint construction and trust among many participants.
A stable community foundation is CORE's strongest wall. The purpose of war is to eliminate war. After a period of tug-of-war, both sides in the market will stabilize. We have always insisted on $CORE not to pursue short-term speculative profits, but because we believe it aims to rebuild the Bitcoin ecosystem, break the value limits of Bitcoin's single storage, establish decentralized finance and payment systems without intermediaries, and use tangible value to eliminate endless speculation and confrontation in the market.
When discussing issues, we should base our judgments on actual situations, not just definitions. $CORE Traditional public chain standards cannot be used to judge. Consensus, native Bitcoin empowerment, and decentralized payment ecosystems are all industry firsts, with no precedents to follow. When evaluating their value, we should start from real real-world scenarios, on-chain data, and ecosystem scale, without stereotypes or emotional views. People are the only force capable of creating world history.
The main feature of Web3 is decentralized autonomy. Every technical update and ecosystem breakthrough in CORE is proposed and jointly completed by community users, while the team is solely responsible for implementation, fully aligning with the core concept of DAO autonomy. It is precisely the persistence of many ordinary holders that has enabled CORE to gradually develop into a leader in the BTCFi field.
All reactionary $UB are like paper tigers. Those who deliberately smear, bad, or ride traffic may have great momentum, but in reality, they are very weak. Such public opinion lacks technical logic, holds faith, and long-term vision, only trying to create panic to attract attention, but in the end, it will only be fleeting.
Only if $CORE sticks to its original intention of decentralization and continuously strengthens the value of its ecosystem will all rumors dissipate on its own. Looking at the entire crypto market, hotspots keep turning, and projects that are just air-conditioned will soon disappear. Only projects with solid technology, real implementation, a foundation of decentralization, and deep community beliefs can survive the cycle. $CORE carries the original intentions of practitioners across many industries, relying on BTCFi's unique innovation logic to possess strong and lasting value.
Let go of restless emotions, face reality objectively, stick to your beliefs, and time will always give every loyal $CORE participant a satisfactory answer.
#MSTR再卖1638枚比特币, scale halved #财报观察员: Mixed results, unlocking imminent! What is SpaceX's outlook going forward? #ADP就业降温, Fed policy divisions intensify SPCX单日暴跌13.61%、收于108.27美元后，期权市场出现了更值得警惕的信号——
超5万份本周五到期的看跌期权（Put），集中堆在了100美元行权价。
100美元附近，市场已经堆满了火药，它会成为下跌加速的突破口，还是利空出尽的反弹地板？
1.为什么是100美元？这不是巧合
最新期权数据显示，本周五到期的100美元Put未平仓合约已达50158份，同价位Call仅4655份，单一行权价的Put/Call比直接拉到10.78:1。
5万份Put，对应约500万股SPCX现货头寸。
以昨日收盘价计算，100美元距离现价只剩约7.6%的空间。
这意味着100美元早已不是普通的整数心理关口。
它是保护仓位、裸空交易、做市商对冲盘三方集中碰撞的价格：
- 持股股东在这里买Put做下跌保险
- 空头在这里下注破位大跌
- 做市商在这里对冲风险敞口
但要明确：Put多≠所有人都在裸做空。
未平仓量只能告诉我们筹码集中在哪，不能直接证明所有人都看跌——里面也可能有机构挂单接货、做组合策略。
2.真正的风险：下跌会自我实现
期权不是预言，但在极端持仓下，它能直接拉动现货价格。
假设这批Put主要由投资者买入、做市商卖出，那么SPCX越靠近100美元，Put对股价的敏感度就越高（Gamma快速上升）。
为了控制风险，做市商必须卖出更多股票对冲敞口。
于是形成一条负反馈链条：
股价下跌 → Put价值飙升 → 做市商加卖现货对冲 → 股价被进一步砸向100美元
这才是这组数据最核心的看点：
市场不只是在押注SPCX下跌，它还可能通过对冲机制，亲手把下跌变成现实。
当然，这种效应的强度取决于做市商真实净头寸，仅靠公开未平仓量无法100%确认。
3.双重杀：期权压力撞上千亿解禁
单靠5万份Put，未必能撬动一家万亿市值公司的走向。
但这次不一样——它刚好撞上了美股史上最大规模的IPO解禁。
8月6日起，SPCX首批限售股解禁，最多9.12亿股获得出售资格，是原有公开流通盘的1.4倍以上，对应市值超千亿美元。
解禁不等于全都会立刻卖出，但它彻底改变了市场的供给预期：
原本就脆弱的情绪，叠加天量潜在抛压，已经足够让资金提前避险。
本周的局面是两股力量叠加：
解禁放大潜在卖盘，期权对冲加快卖盘释放速度。
市场怕的早已不是昨天跌了13%，而是股价一旦摸到100美元，会不会触发机械性的踩踏抛售。
4.反过来：100美元也可能是反弹起点
事情永远有两面。
如果SPCX跌到100美元附近，却始终无法有效跌破，那么随着周五到期日临近，这批Put的时间价值会快速归零。
此前为了对冲而卖出的现货空头，也会逐步买回平仓。
届时交易链条完全反转：
100美元守住 → Put快速贬值 → 对冲盘回补买入 → 股价反向拉升
所以5万份Put 也不是“一定会跌破100美元”。
它只意味着：100美元附近，一定会有人被迫交易。
- 跌破时，做市商和止损盘被迫卖出
- 守住时，对冲盘和短线空头被迫买回
结论
本周SPCX的核心矛盾，已经不是“市场看不看好SpaceX”。
而是：当股价来到100美元，谁会先被迫出手？
放量跌破100美元，它会成为下跌加速器，负Gamma效应会把跌势放大；
盘中触及后迅速收回，它就是利空出尽的起点，空头回补能拉出快速反弹。
期权不会告诉你结局
它只告诉你，这一周的战场，已在 100 刀这里
Ready ～
#财报观察员：业绩喜忧参半，解禁将至！SpaceX后续怎么看？ This is the nuance between the Japan (yen) and US economics,
The yen has fallen to its weakest level in decades because US interest rates are much higher than Japan’s.
To support the yen,
Japan may need to sell some of its huge holdings of US Treasury bonds and use the dollars to buy yen.
That is a problem for the US.
Japan is the largest foreign holder of US debt, with about $1.2 trillion.
If Japan starts selling Treasuries at scale, US bond yields could rise sharply, making borrowing more expensive across the economy.
This is why the US just sold every euro to help the yen survive
but this does not fix the main cause
this large gap between US and Japanese interest rates is the "problem" for usd
and this is why longer-term currency-debasement risk comes from the liquidity backstop and the willingness to protect bond markets from forced selling
They will try every measure to save the yen because they have to and it only ends by devaluing the usd in the process
and that is why i continue to buy high desirable scarcity assets like bitcoin
#DailyOrbit 记一个容易被币圈忽略的宏观变量：美元指数今天向上触及 100，日内涨 0.31%。别小看这条线——DXY 走强通常意味着美元流动性收紧，对所有以美元计价的风险资产（包括 $BTC）都是隐性逆风。过去两个月加密「跟跌不跟涨」，一部分答案就藏在美元这条曲线里：当美元有更好的确定性回报，边际资金没理由跑来接高波动资产的盘。不是看空谁，是提醒你把美元指数放进你的仪表盘。数据不会陪你演戏。上一个周期是第一次 $BTC 牛市，伴随着：
- 没有延长的推动/扩张/抛物线/狂热
- 没有山寨币季节或 BTC.D 崩盘
- 在黄金牛市期间，且 RTY 对黄金下跌
以及更多其他首次。
当人们抱怨这是一个多么糟糕的牛市时，在我看来，我不禁好奇，在如此不同的宏观背景下——考虑到 RTY 对黄金自 2022 年以来一直处于熊市——它是如何支撑住并甚至达到 12 万的。你不觉得这让 BTC 看起来更加强大吗？
历史上，BTC 的推动通常在 RTY 对黄金反转后的第一次回调之后开始，而我们现在似乎正迎来那个第一次回调。让我们拭目以待 ！
#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？ 记一个容易被币圈忽略的宏观变量：美元指数今天向上触及 100，日内涨 0.31%。别小看这条线——DXY 走强通常意味着美元流动性收紧，对所有以美元计价的风险资产（包括 $BTC）都是隐性逆风。过去两个月加密「跟跌不跟涨」，一部分答案就藏在美元这条曲线里：当美元有更好的确定性回报，边际资金没理由跑来接高波动资产的盘。不是看空谁，是提醒你把美元指数放进你的仪表盘。数据不会陪你演戏。Nvidia's next-generation Rubin Ultra plans to reduce memory capacity due to insufficient supply of advanced HBM, shifting the AI industry's bottleneck from computing chips to storage chips, becoming the core driver of the current integration between the tech sector and crypto assets in the US stock market.
Physical supply chains suppress memory stacking, and advanced HBM supply is in short supply, slowing the pace of computing power expansion and propagating upstream memory chip factories along the DRAM chain. In the ranking of driving factors, upstream pellet manufacturers have the highest supply and demand pricing power, followed by rising end-user computing costs leading to a valuation recalibration of tech stocks, and finally a resonance of risk sentiment across the market due to high correlation.
The correlation between BTC and the Nasdaq index has exceeded 0.8, and fluctuations in tech heavyweights in US stocks directly translate into liquidity pressure in the crypto market. Storage bottlenecks raise the operating costs of AI models. If U.S. stocks trigger risk asset deleveraging due to supply chain concerns, BTC, with its high beta attributes, will find it difficult to break out of a one-sided rally.
The scenario triggered by upstream storage giants like $MU is that the ongoing capacity shortage of upstream storage giants translates into price elasticity, and the US valuation logic successfully absorbs the negative impact of memory memory adjustments. If the U.S. tech sector stabilizes and rises and U.S. Treasury yields fall, BTC will follow the Nasdaq upward, and the decentralized computing power concept will also gain a valuation premium due to persistently high computing costs. This script fails as a signal: the industry's expansion rate exceeding expectations, resulting in short-term oversupply.
The trigger for the downside scenario is the continued escalation of Nvidia's supply chain issues, raising institutional concerns about the overall slowdown in growth among U.S. tech stocks. If the Nasdaq experiences a sharp correction, US Treasury yields remain high, suppressing liquidity, BTC will fall in tandem with US stocks due to deleveraging pressure, and pure concept crypto projects will face liquidity exhaustion. This scenario failed signal: upstream pellet deliveries exceeding expectations or macro liquidity easing again.
If the correlation between the U.S. tech sector and crypto assets rapidly decouples and falls below 0.5, cross-market coordinated predictions will immediately fail. At that point, asset pricing will return to their internal capital logic, rather than being dominated by supply chain events in the US-based tech sector.
In the next 7 days, focus on Micron $MU's stock price absorption of the HBM transmission chain, the overall volatility of U.S. tech weights, and changes in BTC's correlation coefficient with Nasdaq.
#MSTR再卖1638枚比特币, Scale Halved #黄金重返4200美元—Why Has BTC Not Risen Accordingly?#闪迪财报双超预期，新增140亿美元回购授权
SanDisk's earnings exceeded expectations, but the stock price dropped 15%, which is confusing to me.
Last night, while lying on the sofa scrolling through my phone, I saw SanDisk released its earnings report. Revenue was 8.97 billion, surpassing the market expectation of 8.48 billion. EPS was $39.25, also beating estimates. They also approved a $14 billion buyback, and with the remaining from before, the total buyback can reach $15.5 billion. The numbers look pretty good, yet the stock fell more than 15% after hours.
I stared at it for a while but still didn’t fully understand. After checking around, the market is talking about the Q1 guidance midpoint of 10.3 billion, which is lower than Wall Street’s expectations. Also, the stock had risen too much, expectations were set too high, and when the results came out and didn’t seem strong enough, it got sold off. Last year it rose more than sevenfold, climbing all the way to 2300, so any slight disturbance causes people to run. Additionally, the Changxin incident is also weighing on the whole sector sentiment.
The performance itself is fine; AI storage demand is still supporting it. But the valuation is too high, expectations too elevated, so any slight disappointment leads to a sell-off. Retail investors cut losses, institutions unload shares.
Good performance, stock price drops,,,,
#DailyOrbit Here's a hard footnote to this storage supercycle: NVIDIA is reportedly considering cutting HBM memory capacity for the next-generation Rubin Ultra—not because they don't want to stack, but because advanced high-bandwidth memory simply can't be supplied. Even the hash power leader is giving way to memory, indicating the bottleneck has shifted from GPU to memory chips themselves. Data is honester than narrative—HBM shortages are carried up the entire DRAM chain, and the most price elasticity is often among the upstream chip manufacturers. Storage stocks pulled back today, but the fundamentals curve is still moving upward. Speak by supply and demand, not by sentiment $MUSanDisk SNDK is an on-chain token mapped from the US-listed SanDisk stock, not a native project token. It leverages the popularity of the storage chip industry to attract speculative funds from the crypto community. This token lacks an actual project implementation and token economic model; its value is entirely pegged to the US-listed SanDisk stock price, while also amplified by leverage and speculative sentiment in the crypto market, resulting in double magnified price volatility.
From the market characteristics, the token has generally poor depth and weak liquidity, with holdings highly concentrated in the hands of a few major players. The market often experiences spikes, sudden pumps, and dumps; the token price does not fully follow the US stock’s movement, leading to severe decoupling between the token price and the stock price, causing abnormal premiums and discounts. Contract trading carries extremely high liquidation risks. Major players can manipulate expectations through news to lure retail investors into buying at high prices, then quickly dump their holdings causing the token price to sharply retract.
On the risk side, this mapped token has no official endorsement and is a third-party on-chain derivative product, with potential black swan risks such as the mapping institution running away, contract vulnerabilities, and liquidation failures. Virtual currency trading and speculation are banned domestically, and related trades are not legally protected with no channels for rights protection. Even if the US stock performs well, the crypto version may independently decline due to crypto market sentiment.
Ordinary participants are easily misled by the storage sector theme and overlook the inherent risks of the mapped token, suffering significant losses after chasing price increases. It is not recommended for ordinary investors to engage in spot or contract trading of this token. Be wary of thematic speculation traps and stay away from on-chain stock-mapped speculative assets. "SanDisk just delivered blockbuster earnings—and still got punished. Welcome to today's market."
Revenue beat. Profit exploded. Earnings crushed expectations.
Yet the stock fell 8% after hours.
Why? Because in 2026, investors don't pay for what you achieved yesterday—they pay for what you promise tomorrow.
SanDisk guided next quarter's revenue to $10.3–$10.8 billion, with a midpoint of $10.55 billion. Wall Street was looking for $11.16 billion.
That's it. A roughly $600 million expectation gap erased billions in market value overnight.
We've seen this movie before. A few days ago, Palantir faced the same reality: earnings are just the ticket to enter the game; guidance determines the price of admission.
The market's message is brutal:
Good results are expected. Great results are priced in. Anything less than perfection is a disappointment.
SanDisk didn't suddenly become a weaker company. Expectations simply ran ahead of reality. After the previous quarter raised the bar to impossible levels, even "very good" was no longer good enough.
And the newly announced $14 billion buyback? Investors barely reacted.
In the past, buybacks were automatically bullish. Today, they compete against one thing: future growth expectations. In a market obsessed with guidance, even a massive repurchase program can't offset forecasts that come in slightly below consensus.
But zoom out for a moment.
The long-term story is still intact.
AI inference demand continues to expand. Customer orders are reportedly locked in for years. New storage standards are being adopted, and the global NAND market could approach $500 billion by 2027.
One earnings call doesn't change those trends.
What happened to SanDisk wasn't a collapse in fundamentals—it was a collision between reality and expectations.
The market isn't saying, "Your business is bad."
It's saying, "You're great, but we expected even greater."
And in today's market, that small difference can cost you 8% in a single night.
$SNDK $XAU $SPCX
#SanDisk #EarningsSeason #AI #Semiconductors
#DailyOrbit $XUSAR – Rally Incoming
Fresh capital is rotating into quality assets, and xUSAR is holding a healthy uptrend with higher targets ahead.
EP: 17.20 – 17.50
TP: 18.20 | 19.20 | 20.50
SL: 16.70$KNC – Momentum Heating Up
Market sentiment continues improving, and KNC is attracting fresh buying pressure. Watch for continuation above support.
EP: 0.1015 – 0.1035
TP: 0.1090 | 0.1150 | 0.1230
SL: 0.0980You might think the Bitcoin is moving sideways, but in fact, it is pricing stagflation
Let's start with two sets of battle data.
Last week, initial jobless claims in the US were 199,000, less than the expected 202,000. Judging by this figure, employment seems quite strong. But when the July services PMI came out, the employment sub-item weakened, and the price sub-was pushing higher. In the same week and the same country, the two tables say the exact opposite.
BTC has been hovering around below 65,000 all day, neither surging up nor falling. Many people watching this horizontal line think there is no market activity, but in fact, the market is seriously doing one thing: pricing with a word, which is stagflation.
Stagflation, frankly, means things are getting more expensive and jobs harder to find. For people like us holding positions for one or two months, this term is more important than any candlestick. It simultaneously removes two pillars: high inflation makes rate cuts difficult; weak employment, making it hard for companies to make money. Risk assets are pleasing on both sides.
What does the plate look like now?
The 200-week moving average is at 63,657, which means the average cost of all buyers over the past four years. The price has just risen above the previous level, but the trading volume hasn't followed. Breaking above low volume is not much different from not reaching the top at all; this kind of positioning is easily knocked back by a large bearish candlestick.
The Coinbase Premium Index has been negative for 80 consecutive days, with the latest reading of -0.0978—the longest losing streak since the index was recorded. This figure is more reliable than any news; it says that spot buyers in the US haven't truly returned. The narrative of stagflation should be beneficial for BTC and fight inflation, but the price difference on the real money side is obvious. No matter how much you talk about it, the money hasn't come.
Now let's look at liquidations. In the past 24 hours, total liquidations across the network amounted to $234 million, long positions totaled 108 million, and short positions hit 126 million—both sides were getting hit. This is a classic case of sideways grinding of the flesh—no matter which direction you go in, it's easy to get stabbed.
On the liquidation chart, it fell below 61,456, with a long liquidation strength of $1.527 billion; it rose to 67,341, pressing down on $1.437 billion in short positions. The current price is stuck in the middle, with the upper wall near over 2,000 points, the lower wall near over 3,000 points. The upper wall is closer, which also explains why every recent surge is only to be pushed back.
There's another figure that's easy to overlook: BTC's 30-day implied volatility has dropped to 36%, at the bottom of long-term support. Low volatility doesn't mean low risk; it's just about accumulating risk, and once you've accumulated enough, you pay it off all at once.
In the short term, this structure is a wall above and below, with a hollow middle space. The most expensive cost isn't the wrong direction, but frequent inflows and outflows. Fees plus slippage points are deducted daily, and over a month, that's a substantial amount.
The long-term line is a different matter. If stagflation really rises, the market's view of non-yielding assets will gradually change, but that will take time and US domestic funds to enter again. The double confirmation I personally watch is still the same two: ETF weekly net inflows turning positive, plus Coinbase premiums turning positive. Before these two appear simultaneously, I prefer to treat every rally as short covering.
One last question: with weakening employment and rising inflation facing you at the same time, do you think this is good or bad news for BTC? Which side are you currently positioning?But why have BTC and ETH been weak recently? Why have the US tech sector started to fluctuate? Why are some semiconductor stocks under pressure? The reason is actually not complicated—the market has already priced in a large amount of positive news. Taking the semiconductor sector as an example, some companies have released solid earnings reports and continued to advance stock buyback plans, but the market is expecting more aggressive earnings guidance and future growth expectations. When expectations far exceed reality, even if good results are delivered, a trend of "positive news being realized and capital taking profits" is likely, cooling overall sentiment in the chip sector. Asian markets were also affected, with adjustments in Korean chip stocks and short-term deleveraging in tech stocks further suppressing risk appetite. However, judging from the demand for AI servers, high-bandwidth storage (HBM), and data centers, the long-term growth logic of the AI industry chain has not changed. The crypto market is in a similar situation. Spot ETF funds continue to flow in, institutional allocation demand has not disappeared, and Bitcoin and Ethereum continue to attract long-term capital attention. However, after the previous rally, more and more funds have begun to lock in profits, waiting for new catalysts such as Federal Reserve policies, economic data, and regulatory progress. The current market is no longer a broad rally; funds are paying more attention to quality and certainty. Hotspot rotation is accelerating, with funds more inclined to flow into assets that truly have fundamentals and growth expectations. In the US stock market, there was also capital rotation. Some funds have flowed from highly valued AI tech stocks into defensive industries, financials, and value sectors, thus technologyBitcoin continues to consolidate within the $62,000 to $65,000 range, pulling back after multiple tests of the upper boundary of the range, with mid-zone buying remains strong. Technical structure shows the market is still accumulating directional momentum, with no clear breakout signals.
Spot ETF capital inflows remained positive by early August, with recent cumulative net inflows exceeding $600 million, effectively absorbing selling pressure from long-term holders. This indicates that institutional allocation remains intact, providing bottom support for the market.
In terms of funding rates in the perpetual contract market, BTC is near neutral to slightly negative, ETH slightly positive, with overall moderate leverage levels and no obvious overheating. Open interest remains below the previous cycle high, and the total supply of stablecoins has shrunk to about $299 billion, indicating that incremental liquidity available to drive market expansion is decreasing.
From the perspective of capital structure, smart money is more concentrated in large-cap assets. Institutional funds are clearly biased toward BTC and ETH, with ETH showing greater resilience compared to BTC. SOL, XRP, BNB, and LINK are mostly consolidating within a narrow range; AAVE and some selected Layer-2s like ARB still have low trading volumes. HYPE and RWA sectors like ONDO have selective attention but have yet to form a widespread sector rotation effect.
🟢 On the bullish side, if ETF funds continue to flow steadily and BTC closes above $65,000 on high volume, the current range will be repriced, opening up upward room for further growth.
🟡 On a cautious scenario, if stablecoin supply continues to shrink or prices fall below the $62,000 support, combined with the ongoing policy uncertainty in September, the market may face deeper correction risks.
Core risks also include thin order book liquidity, slower regulatory clarity than expected, and an accelerated pace of whale addresses transferring funds to exchanges. All these variables could amplify instantaneous market volatility.
📊 Next, focus on the metrics to track$ONE – Breakout Loading
Volume continues to increase as buyers step in. ONE is building strength and looks ready for another move higher.
EP: 0.00124 – 0.00126
TP: 0.00134 | 0.00142 | 0.00150
SL: 0.00119杠杆的代价：存储三巨头狂欢之后
2026年8月7日，全球存储芯片市场再度迎来一场“业绩杀”。闪迪$SNDK 盘前一度暴跌超8%，西部数据跌近12%，SK海力士跌近6%，美光$MU 跌近3%。周二收盘，闪迪暴跌14%，股价较一个月前的历史高点已跌去逾50%，直接腰斩；SK海力士跌8.98%，美光跌8.85%。费城半导体指数跌超3%。
业绩依然炸裂。闪迪Q4营收89.65亿美元，同比暴增372%，毛利率84.6%，EPS是去年同期的135倍；西部数据净利润同比暴增1215%。问题出在指引——闪迪下一季营收指引103亿至108亿美元，不及市场预期的111.5亿美元。差了不到6%，股价崩了9%。
“不是因为存储不行了，而是因为市场对存储的期待已经到了一个'你必须完美'的地步。”高盛如此总结。股价已经提前Price-in了“持续加速”的完美剧本，任何回归正常的信号都被解读为见顶前兆。美光较年内高点回撤35.4%，闪迪跌幅达53.6%，SK海力士和三星分别跌48.1%和41.3%。
韩国股市是这场风暴的震中。KOSPI盘中一度跌超5%触发熔断。韩股自6月高点已累计下跌15%，市场一级熔断3次。韩国杠杆去化尚未完成——券商融资余额7月31日单日缩水10%，创年内最大降幅，散户单日爆仓强平金额仍超千亿韩元，爆仓率高达7%-8%，远高于稳定状态下的1%。融资余额已抹平2月至6月半导体暴涨潮堆积的全部增量杠杆。
韩国监管层紧急出手。财政部长具润哲在内阁会议上表示将努力缓解股市波动，并为此前仓促引入单股杠杆ETF致歉；韩国总统办公室政策室长因此面临刑事指控。监管部门正推动引入“紧急行动权限”，允许快速调降杠杆产品倍数。但机构抛售虽有放缓，指数上行空间依旧有限。
加密市场同样承压。比特币当前在65,000美元附近震荡，较年初12.6万美元的历史高点近乎腰斩。美国7月服务业PMI数据显示就业走弱而通胀压力上升，引发市场对滞胀风险的讨论。以太坊$ETH 现价约1,910美元区间，整体属于联动行情，受大饼牵制。在美元强势和高利率环境下，风险资产上行空间依然受限。
存储三巨头的故事揭示了一个朴素道理：当杠杆成为主旋律，任何不及预期都可能引发踩踏。业绩可以炸裂，但杠杆堆出来的估值经不起风吹草动——无论是在首尔的股市，还是在全球的加密市场。
#闪迪财报双超预期，新增140亿美元回购授权
#ADP就业降温，联储政策分歧加剧
#黄金重返4200美元，BTC为何没跟涨？ 你盯的成交量是机器人刷的 他只被罚了1万
加密圈有一项明码标价的服务，叫交易量支持。项目方付钱，服务商用机器人在好几家交易所之间来回对敲下单，把成交量和活跃度做出来。一家叫 MyTrade 的公司就是干这个的，旗下平台叫 MyTrade MM。
现在这门生意被判了。美国司法部宣布，MyTrade 创始人兼主要运营者 Liu Zhou 因为参与加密市场操纵，在波士顿联邦法院被判处罚款1万美元。案子由美国检察官办公室和FBI波士顿分部一起推进的。
1万美元。这个数可能还不够很多人一天的浮亏。但我觉得真正贵的不在罚款上，而在认罪协议那几条：永久关掉刷量机器人，停止提供这项服务，还要在自己网站上写明白一句话，交易量支持属于洗售交易，在美国法律下属于非法行为。等于自己给自己门口挂了块牌子。这句话以后会被无数人拿去引用。
洗售交易说人话就是左手倒右手。同一伙人同时挂买单和卖单，币在自己几个账户之间转圈，K线动了，成交量上去了，榜单排名也爬上去了，但背后没有一分钱是真实需求。为什么有项目方愿意花钱买？因为新币要上榜单、要够格去谈交易所、要让点进来的人觉得这地方有人气。成交量就是门面。
这事对咱们的意义比判决本身大得多。成交量是大部分人唯一觉得客观的东西，价格能被喊，故事能被编，唯独成交量看着像铁证。结果它才是最好造的那个。
给几个笨办法。看挂单簿两边一个点以内有多厚，能不能一口吃下你手上的仓位，吃不下的，成交量再漂亮也是纸糊的。看24小时成交额和市值的比例，一个市值1000万的币天天成交好几千万，多半有猫腻。再看持币地址数有没有跟着量一起涨，只有量在涨人没多，那量就是空的。
把它跟最近这一串事放一起看更清楚。中国香港那边 Fun Coffee 骗案已经收到255宗报案，损失约1.04亿港元；Chainalysis 说今年上半年光是暴力抢劫逼人转币就损失超3000万美元。手法从链上做到线下，处罚力度和危害之间经常差着好几个数量级。
落到盘面，这案子不影响今天的K线。BTC还在6.4到6.5万这段横，200周均线63657刚站上去，量一直没跟上，Coinbase溢价连着80天为负。它改的是三个月后的合规成本：做市这行会被重新划线，中小项目再想买便宜的量不容易了，新币上线的热闹劲儿多半要掉一档。往长了说这是好事，量变实了，咱们看到的数字才值得信。
所以问一句，你上次买一个小币之前，除了看K线和成交量，还看过别的吗？Before people even understand self-custody, the wallet has already given the keys to AI
In the early hours of August 7th, MetaMask launched something called Agent Wallet. Simply put, it means equipping AI agents with a self-custody wallet, connecting proxy frameworks like Claude Code, Codex, OpenClaw, and setting the rules, and letting the machine handle the rest on-chain.
It supports many chains, including Hyperliquid, Robinhood Chain, Monad, and other EVM chains. Proxies can use ERC-7821 for bulk exchanges or one-time transactions. The most interesting part is that MetaMask directly settles network fees from the assets you transfer.
They also laid a layer of security. Before transaction execution, there are simulations, threat scans, and MEV protection. If a transaction passes all security checks and you still lose money, the maximum monthly compensation is $10,000. The official stance is very restrained, saying the core is to have agents work within user-set permission boundaries, not to hand over wallets without limits.
It sounds pretty comprehensive. But if you rewind the timeline a few days, you'll find this matter a bit surreal.
Last week's Coldcard incident was still unresolved. A flaw in the firmware was hidden at the junction of two unrelated submodules—neither in Bitcoin's code nor in the encryption algorithm, but stuck in that unclaimed gap. On-chain, 1,596 BTC were confirmed to have been taken away, resulting in losses exceeding $100 million.
Coinkite added an even more chilling comment during their own review afterwards. They re-examined all the code using three cutting-edge large models and found nothing. This vulnerability not only evaded years of manual auditing but also evaded AI.
And it was mentioned at least twice to my face. In April 2021, some users questioned that implementation and were dismissed. In May 2025, James O'Beirne directly pointed out that random number generation was suspicious, but still no one took it. Three opportunities, missed once.
Glassnode's data from those days was written like this: BTC daily active addresses surged to 980,000, hitting a new high since December 2024. It sounds like the market is picking up, but in reality, it's all people moving—changing mnemonic phrases, custody methods, and moving coins to places they think are safer.
So the current picture is like this. On one side, the core group of Bitcoin believers just proved unable to understand the code running in their wallets; on the other, the industry immediately handed over the signing rights to AI agents. One moment they said AI audits weren't reliable, the next they let AI place orders for you.
This isn't just MetaMask alone. Cloudflare recently launched an AI proxy wallet, allowing agents to use their own payment identity and consumption API. Everyone is heading in the same direction; no one wants to fall behind in this round.
I'm not saying this path shouldn't be taken. When it comes to agents performing on-chain operations, efficiency is on another level. The issue is who will bear the boundaries of responsibility. A monthly protection quota of $10,000 is enough for small transactions, but for a proper on-chain exchange, it might not even count as a fraction.
There's a more practical issue. Did you really understand the permission boundary you set, or did you just click a few times to confirm by default? The Coldcard group are the most self-managed, yet they still got stuck in an invisible crack in code. Ordinary people face a proxy that can sign themselves—why should they think they can keep an eye on that line?
The term self-custody used to mean the key is in my hands. Now it is turning into the key in the program I configured, and that program I don't understand.
Would you hand over your wallet to an AI agent to manage? If so, how much would you be willing to give it?我是不是又错过了？现在上车$ETH还来得及吗？还没等美国国会那帮老爷们把法案磨出来，俄罗斯的新加密监管法昨天直接就生效了。说干就干啊，这速度比我们追涨还快。讲真这件事我很意外我们天天盯着某机构盯某机构，盯着鲍威尔每个表情做阅读理解结果北极熊一声不吭把框架搭好了不发推不预告没有听证会上议员表演式发言就这么静悄悄地落地了法案细则我扫了一遍，矿工被正式承认为合法经济活动交易所要拿牌照但门槛不低稳定币可以用卢布以外的资产发行，但必须全额抵押最狠的是跨境结算那一块明显是为绕开SWIFT准备的地缘政治我不多聊，但这对加密世界意味着什么你品你细品一个国家把加密写进法律体系，不是当成投机工具是当成战略基础设施在用这和萨尔瓦多当年买$BTC的性质完全不同有人问我是不是利好我觉得短期别指望拉盘，这种消息不是拿来炒的但长期看，这就是硬生生把饼做大以后别的国家跟上，合规资金才能真正进来稳扎稳打，别被一根阳线骗进去又割在最低点 #世界杯收官：西班牙夺冠 #特朗普将决定是否扩大对伊战事 #芯片股反弹，美股空头仓位创历史新高 🔥 Nvidia's chips have "shrunk"—how much longer can the AI narrative in the crypto world continue?
I just saw a piece of news that NVIDIA's next-generation Rubin Ultra chip may have to reduce memory capacity. The reason is simple—high-end memory chips are running low and supply exceeds supply.
On the surface, this matter seems unrelated to the crypto world, but if you think about it a bit, the impact is actually quite significant.
🔴 First impact: AI computing power will become increasingly expensive
Memory reduction means AI companies need to call more chips to run large models to achieve their original results. Chip demand only grows, not decreases, but supply is stuck at memory bottlenecks. Computing power costs cannot fall in the short term and may even continue to rise.
This is a double-edged sword for the crypto AI sector. On one hand, tight computing power gives the concept of "decentralized computing power" more storytelling, and projects like Render and Akash could theoretically benefit. But on the other hand, if Nvidia itself can't handle its supply chain, shouldn't those crypto AI projects claiming to "disrupt Nvidia" also reconsider their technical capabilities?
Don't just focus on the hype—many AI token underlying technologies don't even have demos, relying solely on concepts to rally the market. Nvidia's news serves as a wake-up call to the market: AI computing power isn't something you can just do if you want to.
🟡 Second layer of impact: Nvidia's stock price will fluctuate, and BTC is likely to shake as well
Currently, BTC's correlation with Nasdaq has exceeded 0.8, and Nvidia is a heavyweight on the Nasdaq. If the market fears Nvidia's growth slowing due to chip downgrades and the stock price pulls back, the entire U.S. tech sector will come under pressure.
BTC is now positioned as a "tech stock PLUS," and when US stocks fall, it's hard for it to remain unaffected. So don't be fooled by this news being about chips; in reality, it's planting a seed of uncertainty in the entire risk asset market.
🟢 My judgment
In the short term, AI-themed crypto tokens may be speculated again on the "computing power shortage," but the divergence will be severe. Only those with real technology implementation and real computing power networks can emerge; those purely riding the hype will likely show their true colors.
In the medium to long term, if Nvidia's supply chain issues persist, the valuation logic of US tech stocks may readjust, which is not good news for BTC. After all, BTC's pricing power is now in the hands of institutions, who watch the US market's mood.
💡 Advice for the community brothers
1. If you have AI sector coins in hand, pay close attention to the actual progress of project teams recently—don't just listen to stories. If you say a small project can handle something even Nvidia can't handle, do you believe it?
2. Pay attention to NVIDIA's upcoming earnings reports and stock price movements, as these are early indicators of risk asset sentiment.
3. August is inherently a month of volatility; don't hold too much position; keep some cash for certainty.
Nvidia's chip "shrinkage" is not just about memory memory, but may also be driven by the market's enthusiasm for AI's unlimited growth. The AI narrative in the crypto world should move from the "painting promises" stage to the "inspection" stage.
👇 Do you hold any coins related to the AI concept? How much longer do you think this narrative can last? Let's talk in the comments."The market doesn't punish bad earnings—it punishes fading expectations."
SanDisk just reminded everyone of that lesson in brutal fashion.
I opened my short around 1337, and with the stock now trading near 1212, the position is up more than 120 points. The thesis wasn't about weak numbers. In fact, the earnings report looked almost perfect:
• Revenue surged 372% year over year to $8.97 billion
• EPS exploded to $39.25
• Gross margin hit a record 84.6%
• Data-center revenue jumped 437%
• Another $14 billion buyback was approved
On paper, everything looked incredible.
So why did the stock collapse from 1480 to nearly 1210?
Because Wall Street doesn't price yesterday's victory—it prices tomorrow's growth.
The real problem was guidance.
Q1 fiscal 2027 revenue came in below expectations, profit growth is beginning to slow, and margins appear to be peaking. Suddenly, the market stopped asking, "How good was the last quarter?" and started asking, "Is this the best it gets?"
That's what triggered the sell-off.
How I'm managing the short
My stop-loss has been moved down to 1300, locking in profits even if we see a sharp rebound.
Profit-taking plan:
• Close 30% between 1150–1180
• Close another 30% between 1080–1100
• Exit the remaining 40% around 1000–1020
If price rebounds into 1250–1270 on weak volume, that's a potential area to add. If buyers reclaim 1300 with strength, the bearish thesis is invalidated.
Why I'm watching 1088 for longs
This isn't blind bottom-fishing.
The 1080–1100 zone sits near major technical support and a dense area of historical buying interest. The stock has already fallen nearly 58% from its highs, and short-term indicators are deeply oversold.
More importantly, the long-term story hasn't disappeared:
• AI storage demand remains strong
• Analysts still see significant upside
• 2026 capacity is reportedly sold out
• NAND prices continue to recover
If we reach 1088, I'll consider building a position gradually with limited leverage and strict risk management.
The plan is simple:
#DailyOrbit 65K附近已经盘了多久了？动都不带动一下的。$BTC这几天的走势就像被按了暂停键，各种上下插针，急跌完又拉回来，拉回来又继续磨。昨晚PMI数据一出来，市场直接懵了。数据超预期，但价格结构很差——成本在涨，订单在缩，妥妥的滞胀前兆。美股弱，黄金强，比特币在中间犹豫。这个分化已经不是一天两天了，资金都在做防御性配置，没人敢all in风险资产。滞胀这个事，老韭菜应该都有肌肉记忆。2022年那波怎么跌的？不就是通胀没下来，经济先下来了。当时$BTC从48K一路滑到16K，连个像样的反弹都没有。现在倒不至于重演那种级别的崩盘，但情绪上已经有那味儿了。先别慌，我不是说要去16K。现实情况是机构手里还有大把筹码没出完，ETF的流入量虽然下来了但也没转负，说明大资金也在等。$BTC这个位置，我觉得关键看两点：一是68K能不能有效站稳，过不去就别谈什么趋势反转；二是下面56K-58K的支撑带，那是近三个月的筹码密集区，跌破的话短期情绪就真的崩了。我自己现在是半仓观望，不加也不减。这种行情追涨杀跌就是送钱，横盘越久，变盘越猛，等方向出来了再动也不迟。别听那些大V喊单，说什么逆势梭哈的鬼话。滞胀周期里的资Google Trends has reached its lowest level since November 2020
This marks the lowest level in nearly six years
The last time such a low level of attention occurred was in September 2023 and October 2022;
As everyone knows, these two months have basically been the start of a bull market
The bottom of the next cycle is expected to emerge this year
It should be fine if it's within 2 months or 3-4 months at the fastest
It is estimated that BTC will see at most one or two more waves of decline, which would give us the chance to enter the best all-in spot trading in the past five or six years
I wonder if the market will give this opportunity?
What do you think? #现货黄金站上4300 The expansion of AI infrastructure has extended storage chip order backlogs to the long term, and capital in the U.S. tech sector has once again concentrated and flowed back to storage leaders represented by $MU.
After the earnings report was released, the related stocks corrected at market highs, but contract prices still adjusted valuation benchmarks by 50% to over 90% each quarter.
Capital rotates between high-risk sectors such as U.S. tech hardware and crypto assets, and the tight demand for high-profit capacity in data centers has locked in multiple contracting contracts.
When large manufacturers prioritize server-grade product capacity supply, the clearing out of traditional markets directly translates high prosperity into sustained growth in leading companies' gross margins.
If enterprise-level storage orders continue to be backed until 2027, the highly certain earnings release will drive prices upward to digest high valuations; If the macro interest rate environment tightens and liquidity retreats, the upward path will be cut off early.
If the pace of capital expenditure by major companies slows, the highly volatile chip structure may trigger cross-market safe-haven selling pressure; As long as quarterly contract prices remain strong, the room for deeper declines will be limited.
If the assumption of a continued supply gap is shattered by the accelerated pace of new capacity launches, the current market-based double revaluation premium will be completely disproven.
The most noteworthy variable to watch in the next seven days is the capital expenditure plans of major companies and the realization of new storage contract prices.
#西联稳定币卡落地, Visa further advanced its payment scenario by #意大利大行减IBIT普通股94%, increasing its staked ETH positionBitcoin's victory was not due to technology but to the choice of capital. On the surface, competitors like BCH, ETC, FTT, and ADA seem to have collapsed one by one, but what the market has actually repriced is not the superiority of surviving protocols, but rather the process of liquidity and development resources concentrating in one direction. What happened? The cases listed in the original text are not just price comparisons, but the result of competition for capital, developers, and users in the crypto market since 2017. Bitcoin absorbed BCH's hashrate and mining capital, and Ethereum replaced ETC's smart contract ecosystem. BNB became the benchmark for trust among centralized exchanges after the collapse of the FTX ecosystem, to which FTT belongs, while Solana surpassed ADA in attracting developers and real-world transactions. The cases of Avalanche, Uniswap, Aave, and Curve are specific examples of latecomers replacing established giants in subnets, DEX liquidity, lending protocols, and stableswaps, respectively. The most decisive event was the collapse of FTX. This is because competition一家卖存储的公司，为什么市场开始用AI基础设施公司的眼光重新审视它？
这个问题放在闪迪（SNDK）身上，现在比财报数字本身更值得关注。
闪迪最新财报交出了一份超出市场预期的成绩单，营收和利润双双击败预期，同时宣布新增140亿美元股票回购授权。表面看，这是一份漂亮的财报，但市场真正交易的并不是这一季度赚了多少钱，而是一个更大的问题：
存储行业，真的只是周期反弹吗？
过去很多年，闪迪和整个NAND行业一直摆脱不了“周期股”的标签。
需求上涨，价格改善，厂商扩产；随后供给增加，库存压力上升，利润快速回落。
所以过去投资者看存储公司，更多关注库存周期，而不是长期成长。
但AI时代正在改变这个逻辑。
以前市场关注GPU算力。
后来开始关注HBM。
现在越来越多人意识到，AI基础设施不仅需要计算能力，也需要大量高性能存储。
模型越来越大，数据越来越多，企业对于存储容量、速度和稳定性的要求都在提升。
这也是为什么闪迪这份财报的重要性，不只是利润超预期。
它证明了一件事：
市场可能低估了AI时代对于存储的长期需求。
但这里也有一个不能忽略的问题。
回购140亿美元，说明公司现金流改善，也代表管理层对于未来信心增强。
但回购无法解决存储行业最核心的问题：
供给。
因为存储行业最大的敌人，从来不是没有需求。
而是需求增长速度，能不能长期跑赢产能扩张。
当利润越来越高，三星、SK海力士、美光、闪迪都会增加投入。
如果未来AI建设速度放缓，市场对于存储周期的担忧可能再次回来。
所以现在看SNDK，不能简单理解成：
“财报超预期，所以继续上涨。”
真正需要观察的是：
AI带来的存储需求，到底是一轮新的产业周期，还是一次被市场情绪放大的阶段性行情。
我觉得闪迪现在最大的变化，是市场正在重新定义它。
$SNDK
以前买SNDK，是赌存储价格上涨。
现在买SNDK，是赌AI时代的数据增长。
如果这个逻辑成立，它的估值体系会发生变化。
但如果AI资本开支降温，它也会成为最先被重新定价的资产之一。
回购只能影响短期情绪，真正决定闪迪未来高度的，是AI时代到底需要多少存储。
财报证明的是现在，市场下注的是未来。
#闪迪财报双超预期 #SNDK #AI基础设施 #半导体 #美股CLARITY法案被踩刹车！
美国加密监管似乎离落地只差一步，参议员沃伦却明确反对当前的法案。
她认可立法的必要性，但认为法案的内容规则在利益冲突、消费者保护、国家安全和经济风险上留下了太多空白。
据美国国会公开记录，CLARITY法案在2025年以294票对134票通过众议院，今年6月又经参议院银行委员会修订，进入参院立法日程。
到了这一步，说明两党对“需要一套加密规则”已有共识，分歧集中在规则的清晰程度。
法案试图划清SEC与CFTC的监管范围，解决加密项目长期面对的身份认定难题。
行业当然欢迎，因为边界越清楚，融资、上市和机构入场越容易。
但沃伦担心，如果只给代币分类、给平台发通行证，却没有同步补上反洗钱、关联交易和客户资产保护，所谓“监管清晰”可能变成监管套利。
所以接下来看法案能否获得足够的民主党支持才是关键。
保护条款加码，企业合规成本必然上升；维持现稿，参院过关又会更难。
#CLARITY法案推进受阻，参议院分歧扩大 Wall Street's Q2 revenue forecast of $86.4 billion for $NVDA clearly differs from the company's official median guidance of $91 billion, with the core conflict being whether the Q3 guidance can support a high valuation premium.
The focus of trading on the board has shifted from actual fulfillment capacity in Q2 to provisioning guidance and gross margin changes. The factor ranking is, in order, Q3 revenue guidance range, Q2 gross margin performance, and management's statement on computing power demand.
The bullish breakout scenario requires Q2 revenue to be above the $91 billion guidance midpoint, and the Q3 guidance above the $98 billion ceiling. If this condition is met, it will strongly suppress short positions and trigger after-hours buying to accelerate the rally.
The scenario for a volatile correction corresponds to Q2 revenue exceeding Wall Street's $86.4 billion expectation, but Q3 guidance falls below $94 billion. This positive sentiment pattern will repeat past sell-and-trade scenarios, suppressing short-term upward price elasticity.
The bear acceleration scenario occurred when Q2 revenue fell below the $89.2 billion guidance floor, while Q3 outlook and gross margin were revised downward. This would confirm the hypothesis of slowing computing power spending growth and trigger concentrated selling.
When performance indicators fall short of expectations, if CEO Jensen Huang offers a strong outlook for emerging and future AI, market sentiment may recover rapidly, causing short positions built solely on financial indicators to instantly become invalid.
In the next 48 hours, the key focus will be on whether Q3 revenue guidance surpasses $98 billion and whether gross margin data remains stable.
#Circle财报后押注Arc, can USDC see new growth? #内存卖方市场延续, can Korean stocks see a reversal?RESOLV: Liquidity traps lie behind the volatility of small-cap tokens
RESOLV surged 13.73% in 24 hours to $0.0200, with prices fluctuating widely between $0.0169–$0.0211. Despite a market cap of only $9.26 million, it contributed $1.87 million in trading volume, with a turnover rate exceeding 20%, typical of speculative small-cap high turnover.
All social sentiment data is missing (N/A), with both bullish and bearish values at 0%, which is a significant divergence from the price surge—either the market doesn't need public hype, or the rapid and short-term capital moves in and out without sustained discussion. Smart Money Signal reveals the cold trading reality: net short selling, zero positions, zero traders—professional funds not only failed to participate in the rebound, but instead hedged or exited. This 'price rises, volume increases, but smart money shorts' scissors gap is highly likely due to market makers using thin liquidity to push up and sell.
Core judgment: RESOLV's current gains lack fundamental support. Smart money is shorting against the trend and sending clear risk signals. The probability of a short-term pullback to the $0.017 range is very high. Buyers should be wary of one-sided declines when liquidity dries up.风险偏好半场反转？代币化美股内斗：XSOXL吃肉 XSNDK送死
当前美股叙事主线是财报季+AI半导体+risk-on反弹。
但代币化市场直接打架：XSOXL小涨1.51%独苗，XSNDK却暴跌8.4%。
这说明纳指在反弹，半导吃香，3x空头直接被收尸。
XSPY仅跌0.24%稳住，XSKHY却大跌6.4%，XSKHY跌幅是XSPY的26倍。
资金明显从创新股切向大科技+AI，风险偏好还没完全回来。
总成交5351万刀，XSNDK和XSPCX两家占了40%，波动全集中在这里。
crypto在宏大叙事里还是跟屁虫：BTC只涨0.31%，恐惧指数25低位徘徊。
我现在short XSPCX @109.68，浮盈0.11%，TP设101.78，SL 114.07，等反弹继续空。
同时short GRVT浮盈0.81%，等它再跌一波。
这波美股代币化，到底是risk-on启动还是假反弹？ZEC：隐私赛道龙头在合规阴影下寻找支撑
ZEC 现报 $496.28，24 小时跌 4.82%，在 $488.81–$524.20 区间回调，市值 $83.2 亿配合 $864 万成交量，换手率仅 0.1%，大户筹码高度锁定。
社交情绪与聪明钱信号双双「三零」：无讨论热度、无多空分歧、智钱净做空且零持仓。这反映市场对 ZEC 已形成「知名但不参与」的共识——隐私币叙事虽存，但交易所下架潮、监管不确定性导致机构资金系统性规避。价格中枢下移更多源于缺乏增量买盘，而非主动抛压。
核心判断：ZEC 估值锚定已从「隐私叙事溢价」切换至「合规生存预期」，$480–$500 将是中期筑底区间，等待监管风向明确或技术升级（如 Halo 2 普及）带来基本面拐点前，大概率维持弱势震荡。SanDisk's earnings both beat expectations—is the AI storage frenzy just beginning?
The biggest bottleneck in the AI era may be shifting from chips to storage
SanDisk's latest financial report sends a strong signal: the expansion of AI infrastructure is redefining the storage industry.
The company released its FY2026 Q4 financial report, with revenue reaching $8.97 billion, higher than the market expectation of $8.48 billion; adjusted EPS was $39.25, significantly exceeding the market expectation of $34.96. At the same time, the company announced an additional $14 billion share repurchase authorization, raising the total repurchase scale to $15.5 billion.
Behind the data, the core logic remains the demand for AI.
In the past, the market focused more on computing chip companies like Nvidia, but as AI model scale expands and data processing volume grows rapidly, demand for high-performance storage, high-bandwidth memory, and data center storage is becoming a new competitive focus.
SanDisk's better-than-expected performance essentially reflects that AI capital spending is being transmitted deeper into the industry chain.
However, after the earnings report was released, the market did not fully accept it, and the stock price fluctuated, mainly because investors began to focus on the next issue:
Can the growing demand for AI continue to translate into profit growth?
The company's revenue guidance for the next quarter is $10.3 billion to $10.8 billion, with the median below some market expectations, triggering a short-term capital adjustment.
This is also a common issue facing the AI industry chain today.
Over the past year, the market traded on "AI demand explosion"; Entering the next phase, the market will focus more on corporate profitability, price cycles, and returns on capital investment.
My view is that the storage industry may be entering a new cycle, but this time the driving force is different from the past.
In the past, storage prices relied more on supply and demand cycles, but this round is driven by structural demand changes brought by AI. In the coming years, data center expansion, growing demand for AI inference, and widespread high-performance computing may all continue to increase storage value.
But the market doesn't always reward only expectations.
Whether SanDisk can continue to enjoy AI dividends depends crucially on whether demand for high-end storage persists and whether profit margins can keep pace with industry prosperity.
In the short term, post-earnings volatility is more likely to be an adjustment to expectations; In the long term, AI infrastructure is still in an expansion phase, with storage, power, and data centers potentially becoming the next beneficiaries.
The story of AI has shifted from "who owns the model" to "who can support the model's operation."
Storage is becoming an increasingly important part of this industrial transformation.
$SNDK $BICO $GRVT
#闪迪财报双超预期, an additional $14 billion repurchase authorization was added XCRCL：高价低市值的流动性悖论
XCRCL 报 $65.11，24 小时涨 3.55%，波动区间 $60.24–$65.93，但市值显示为 $0.00M、成交量 $672 万，这种「高单价、零市值、有成交」的异常组合，暗示流通极少或数据统计异常。
社交情绪全维度缺失，聪明钱信号同样是净做空、零持仓、零交易员。专业资金的集体缺席，配合极低的有效流通量，构成典型的「筹码锁仓+对敲拉升」特征。高单价往往吸引散户「便宜心理」接盘，实则是流动性陷阱——一旦想大额卖出，滑点将极其惊人。
核心判断：XCRCL 当前价格体系缺乏市场共识支撑，聪明钱做空释放避险信号，建议非项目方内部人士完全回避，此类代币的风险收益比严重失衡。Everyone says that behind Open USD is a coalition of traditional financial institutions, which will impact existing stablecoins, which is why $CRCL plunged 18%.
If you ask me, what impact can it make? Whether it can actually be issued is still unknown. The more people participate and the greater the momentum, the more often the matter falls out due to unequal distribution of interests. You have to believe that traditional finance is all about refined egoists.
Conversely, this is actually a positive for $CRCL. Why? Just one stablecoin project has mobilized so many traditional financial institutions to resist, which precisely shows that the market demand for stablecoins is real and huge. It has already started to shape the traditional financial market.
No matter how much they organize or unite to resist, it will ultimately be in vain. At that point, $CRCL will instead face a new round of surgesSuper wow, Yushu Technology's IPO issue price has been officially set, and due to the difficulty of winning the lottery, netizens joke that if you win, you are the child of fortune.
The final issue price is set at ¥150.8 per share. Calculated at 500 shares per lot, the payment after winning the lottery is about ¥75,400. Based on the estimated net profit excluding non-recurring items of about ¥591 million in 2025, the issue price-to-earnings ratio is about 103 times. This is not cheap compared to traditional manufacturing industries, but considering that the humanoid robot sector is still in a high-growth phase, institutional pricing reflects more the performance expectations for the next 2-3 years.
This IPO is raising about ¥6 billion, issuing about 40 million shares, accounting for about 10% of the total share capital, corresponding to an overall company valuation of about ¥60 billion. For robot companies that have attracted high attention in recent years, many investors believe this valuation still has room for growth, and the market performance after listing is worth continuous attention.
The real difficulty is not the valuation, but the probability of winning the lottery.
Compared with the large-scale issuance of Changxin Storage before, Yushu's issuance quantity this time is more than 100 times less, with a significantly smaller circulating share, which also means the difficulty of winning the lottery has greatly increased. The attached chart shows the AI-simulated winning probability. Some netizens joke that if you win, you are either a Tianlong person or the child of fortune.
Regarding the new share subscription returns, market expectations also differ. Some shout "earn ¥300,000 per lot," while others think this goal is too optimistic. Based on the current on-chain Pre-IPO market transaction price of about $74 per share, which is about ¥500 per share, the theoretical return per lot is about ¥170,000. Many market participants believe that ¥150,000 to ¥200,000 may be a relatively reasonable expected range, while ¥300,000 would require extremely strong capital chasing after listing.
It should be noted that the on-chain Pre-IPO price belongs to over-the-counter market transactions, which fluctuate greatly and can only be used as a reference for market sentiment and expectations, not representing the actual transaction price after listing.
Overall, the biggest highlight of Yushu Technology may no longer be "how much it is worth," but whether the scarce circulating shares combined with the high heat of the robot sector can continue to drive premium performance after listing. What really determines the final return, besides market sentiment, is a more realistic question—whether you can win the lottery.The best way to buy a $BTC bottom has always been through a mix of two strategies.
Time-based accumulation and price-based capitulation.
Yet every bear market, people try to one-shot one exact price.
They watch BTC fall 50%, decide good prices still aren’t good enough, then keep lowering their bids until the market eventually front-runs them.
The better approach is to begin DCAing after the mid-cycle flush, while keeping some dry powder available for a potential final capitulation.
If capitulation comes, deploy that capital into the deeper move.
If it doesn’t, deploy it once price breaks from the lows and begins confirming strength.
Using $100 per week through the highlighted 2018 period would have accumulated 0.369 BTC for $2,100 at an average price of $5,688.
At the next cycle high, that position was worth $25,473.
Repeating the same process in 2022 would have accumulated 0.0685 BTC for $1,400 at an average price of $20,423.
At the next cycle high, it was worth $8,656.
Neither one required you to time the bottom.
Those calculations also exclude any additional capital deployed during the capitulation itself.
Because you don’t need to predict the exact bottom.
You need a plan that reduces your risk across both time and price, keeps capital available for capitulation, and prevents one unfilled target from leaving you completely underexposed.🐋 XRP Whales Dominate 81% of Binance Outflows
Large $XRP holders are once again dominating Binance withdrawals.
📊 On Aug. 3, the 7-day average of whale outflow dominance reached 81%, while retail participation fell to just 18%—nearly identical to levels last seen on June 11.
However, the broader exchange market tells a different story.
🔹 Binance: Whales 81% | Retail 18%
🔹 All CEXs: Whales 72% | Retail 27%
Across all centralized exchanges, whale dominance has actually fallen from 79% to 72% since July 2, while retail activity increased from 20% to 27%.
💡 The divergence suggests that large-holder XRP activity is becoming increasingly concentrated on Binance, rather than representing a market-wide trend.
⚠️ These figures measure the share of outflow activity—not total withdrawal volume or the final destination of XRP.
#XRP #Ripple #Binance #Whales #OnChain #Crypto #CEX #MarketUpdate #CryptoNews$SNDK I actually felt this rebound too—I mentioned it 8 hours ago
The strong rebound the next day was based on previous historical market conditions
But unexpectedly, the rebound was so fierce and the market expert taught it a lesson The BTC Clarity Act has not yet been released, reportedly on the morning of the 7th.
Originally, starting from the 6th, the short-selling Mizhu would have been shorted because there was no clear bill on the 5th, but the bill has yet to yield results. Therefore, it is not moving.
However, the next 7 days were empty events.
Assuming the circulated issue is on the morning of the 7th, there are only two scenarios:
1. Clarify that after the bill is announced and the vote passes, it becomes a push-up and a push.
2. If the bill is directly negative, then it will directly plunge downward.
How to distinguish these two using event nulls lies in the first step interpolation.
Because the first push was hype, he had no idea about the bill's outcome. Therefore, the push meant the event was in vain, and then the accuracy of the bill's outcome news was determined by whether it would be upgraded.
In other words, online users first push down and wait for media release before the rally will rally a few minutes later.
A few minutes later, a release is bearish, and that's when it's a dip followed by a plunge.
$BTC 2026年8月7日 · ETF连吸5天，但今天有个更大的炸弹
先说价格。BTC现在挂在64600附近，24小时基本没动，成交量207亿比昨天缩了13%。但别被这种平静骗了——今天盘面下面压着三股力量，方向随时可能爆出来。
第一股力量：ETF资金翻脸了。 8月1号还在流出2.07亿，从8月3号开始连续5天净流入，累计吸了将近8亿美金。贝莱德IBIT一个人买了4天，8月4号单日就砸了1.7亿。更有意思的是Franklin Templeton——停了一个多月没买，8月3号也回来加仓了。机构从跑到买，这个转向比任何技术指标都真实。
第二股力量：链上数据在说话。 8月初巨鲸地址（>100 BTC）净增1.3万枚，散户小地址（<10 BTC）净减1.4万枚。翻译成人话就是：散户在割肉，巨鲸在接盘。活跃地址71.2万，三个月最高；巨鲸交易6.18万笔，五个月最高。大资金在动，而且方向是吸货。
第三股力量：今天是美国参议院夏休前最后一个工作日。 CLARITY加密法案卡在程序投票，需要60票才能推进，共和党只有53票，还差7张民主党票。今天投不过，今年基本就凉了。这是2026年最大的政策催化剂，盘面在等这个结果。
技术面： 价格正好顶在50日EMA（64500-64600），过去三周每次摸到这根线就被压回来。往上放量站稳65000看到67000，往下62500-63000是买方防线，破了看61400。恐贪指数17，极度恐惧——但历史数据告诉我们，ETF在买+散户在怕=中期底部区域。
今天计划：
🔥放量站稳65000，回踩64600轻仓多，止损64100
🔥跌破63000，反弹63500试空，止损64100
🔥法案投票结果出来前别重仓，今天波动会非常夸张 $BTC