Orbit Post Sitemap
The biggest uncertainty about SpaceX's future is not the unlocking itself, but the combination of three factors—the pace of the unban, AI capital expenditure, and Starship's progress.
The unlocking will be in multiple batches. The first batch of 912 million shares on August 6 will follow, followed by multiple phases and large-scale shareholder unlocks. By the end of the year, tradable shares may increase from 639 million shares to 5.33 billion shares. The supply shock hasn't disappeared, only delayed. Short sellers have more ammunition—unlocking has added new borrowable shares to the securities lending market. Bears won't retreat just because the market rebounds 6% in one day; they are betting on continued dumping in subsequent unlocking batches.
AI capital expenditure is an even bigger issue. Q2 capital expenditure was 18.3 billion, revenue was 7.8 billion, burning money at more than twice the rate of earning. The company said Q3 and Q4 were similar to Q2. Wall Street was calculating — whether SpaceX could maintain EBITDA growth while continuing to burn money. As long as the pace of burning money outpaced the rate of profit, valuation pressure would not disappear.
Starlink business is currently the only source of profit. In Q2, Starlink revenue was 4.29 billion, with operating profit at 1.66 billion. AI and rockets are burning cash, with most of Starlink's earnings going to support AI and Starship. If Starlink's user growth and ARPU continue to exceed expectations, SpaceX's cash flow pressure will be eased. If Starlink's growth slows, SpaceX's cash burn will become the market's biggest concern.
#财报观察员: After the lock-up lifts, prices rebounded—what do you think about SpaceX's future? Altcoins only go long, not short; going long can at most double your money, shorting is chaotic for life, shorting is easy to profit from daily, but encountering a pump-and-dump coin basically ruins everything...
For example, if you buy $HEL at 0.1 with $1000 and it goes to zero, you lose $1000; going long $1000 at 0.1 to 0.5 means you earn $5000. Long positions have unlimited profit potential, short positions have limited profit.
When altcoin hype comes, a few pump-and-dump coins always appear. In crypto, a rags-to-riches story is not impossible, after all, there are far more people shorting than in any other market. Everyone knows altcoins will eventually go to zero; theoretically, as long as the capital is large enough, shorting can always make money, but you and I are just ordinary people without infinite bullets.
Most altcoins have hidden whales; you just need to observe the position size. The contract shell of small market cap altcoins is roughly $5 million, which is why many altcoins, as long as they are not delisted, almost always have about $5 million bought.
So with a $5 million market cap, no hype, no liquidity, why are there two to three million dollars in positions? Bidirectional positions equal the market cap. Would you say such an altcoin has whales?
For these altcoins, you just need to ambush with $1000; your cost is lower than the whales, your advantage is quick in and out, whales can't escape, a big ship is hard to turn around. If the price keeps falling, whales lose more than you, so it's not that the road is uneven.
In short, to make a comeback, initially go long; shorting can indeed make money in the short term, but if you encounter a pump-and-dump coin, just wash and sleep. Shorting is chaotic for life. #联储鹰派信号升温，弱就业能否压过通胀？ #联储鹰派信号升温，弱就业能否压过通胀？
⚠️个人观点交流，不构成任何投资建议
美联储主席凯文·沃什放话：通胀要是继续走高，9月就会考虑加息。
ADP就业数据已经变差，但就因为主席这番偏鹰的发言，市场直接把9月加息的预期拉满。
💡我的核心观点：主席一句话能搅乱短期盘面，但我不会听讲话就去开单。依旧反弹高空，老老实实等通胀、就业实锤数据出来再说。
沃什随口一席话就能把盘面搅得天翻地覆，但这只是带条件的口头说法，不等于已经确定要加息。
只有通胀再度超预期，9月加息才会真正拿出来聊。
美联储两头都要顾：通胀涨，就想加息压制；可要是就业崩得快，又不敢随便加息。
$BTC 很容易被消息砸一下拉一下，但中期怎么走，还是看美债利率和真实宏观数据，不是看官员怎么说。
实操上，看到这种鹰派新闻别头脑一热直接重仓空。消息只当吃瓜参考，数据出来才是动手的信号。
我自己优先看CPI/PCE，通胀是我最看重的指标；非农就用来看看就业扛不扛得住。继续拿反弹高空的思路，一旦通胀数据不及预期，该认错就果断跑。
现在宏观来回反复，很容易被一条新闻带歪，做交易守住自己原本计划最重要。 非农数据到底有没有用？如果数据偏弱，9月加息概率可能跌破50%，大饼可能迎来喘息窗口。
#联储鹰派信号升温，弱就业能否压过通胀？
如果数据偏强，美元走强、加息预期升温，风险资产可能再次承压。但安永说即便非农数据出来，美联储年底前大概率维持利率不变。
理由是“劳动力市场依旧平稳，不太可能改变美联储的政策重心”。只有通胀范围扩大且持续走高，才可能迫使美联储改变立场。
问题在于——就业降温会不会最终传导到通胀上？如果就业持续走弱但通胀粘在3%以上，美联储就会陷入两难：加息会加剧就业恶化，不加息通胀压不住。这就是“滞胀”风险。沃什现在的策略是——通胀不下来就不松手。就业数据可以弱，但通胀不下来就什么都不会变。非农数据影响的是短期情绪，不是美联储的中期路径。 The contradiction between weak employment and strong inflation is becoming the Fed's biggest challenge.
#联储鹰派信号升温, can weak employment outpace inflation?
Employment data is cooling down, but wages are still rising, and service prices are not coming down.
Structural inflation remains, but it has shifted from the demand side to the wage side. Walsh's hawkish stance is clear—price stability is the top priority, employment second.
The market expects 83,000 new nonfarm payrolls in July. If the data meets expectations, the Fed will have more room to continue focusing on inflation. If the data falls significantly short of expectations, the market may briefly trade the logic of "weakening employment → Fed easing." But EY's judgment is: as long as inflation does not come down, even weak employment data will not change the Fed's policy focus.
Bitcoin's current price is around 64,400, and the market is very quiet before the non-farm payroll data comes out. Don't hold too much position before the direction is released. If the data is weak, Bitcoin could rebound to 65,000 or even 66,000. If the data is strong, the 63,000-64,000 price will most likely be tested. Make judgments after the data comes out—this whole night won't make a difference. @币圈超短王马大帅 US July Nonfarm Payroll (NFP) Full Analysis | Crypto Market Reference (Published 2026.08.07 20:30) #存储股财报后下挫, Is the AI Memory Bull Market Still Stable?
1. Basic Core Information
1. Announcement time: August 7, 20:30 Beijing time (tonight)
2. Issuing Authority: U.S. Bureau of Labor Statistics
3. Three core indicators (all indispensable: salary > new employment)
(1) Nonfarm Payroll Additions (Core Header Data)
(2) Unemployment rate
(3) Average hourly wages (inflation core, directly affecting Fed policy bias)
4. Benchmark values
◦ Previous value (June): 57,000
◦ Market consensus expectation: 80,000
◦ Unemployment rate forecast: 4.3% (previous value 4.2%)
◦ Hourly wage expectations rose slightly year-on-year
2. The underlying logic behind price movements (common in the crypto community)
The strength of nonfarm payroll data → expectations for Fed rate cuts → the dollar index, fluctuations in US Treasury yields→ and pricing of cryptocurrencies like BTC and ETH
1. Non-farm payrolls far exceed expectations (>100,000 + rising wages) = negative for the crypto sector
Employment data is booming, inflation is hindering a decline, the market lowers the probability of a rate cut in September, the dollar strengthens, capital flees into risky assets, BTC and ETH are rapidly declining, and contracts are prone to chain liquidations.
2. Nonfarm payrolls below expectations (<70,000 + weakening wages) = positive for the crypto sector
Weakening employment reflects economic pressure, with the market betting on the Fed accelerating rate cuts, rising expectations for liquidity easing, and major currencies surging rapidly in the short term.
3. Data matches expectations (75,000–90,000) = wide fluctuations
The divergence between bulls and bears is obvious, with the market oscillating back and forth, and the one-sided trend is weak.
Key misconceptions
Don't focus solely on new jobs: overall employment is stable, but wages have risen sharply, which is still bearish. Wage inflation is a key target for the Fed's control. At the same time, last month's revised employment data will directly reverse the short-term market trend.
3. Market Prediction for Three Scenarios (Current ETH Price about 1900 USDT, BTC Synchronized Linkage)
Scenario 1: Strong data (100,000 new ≥, bearish downtrend)
• BTC first support: 62,200; if it breaks below 60,800, strong support is needed
• ETH short-term support: 1870, 1840
Scenario 2: Weak data (70,000 new ≤, positive news for upward movement)
• BTC resistance: 65,300, 66,100
• ETH short-term resistance: 1930, 1960
Scenario 3: As expected (78,000–88,000 RMB, range-bound)
BTC fluctuation range: 62,500–65,000
ETH volatility range: 1875–1935
4. Forward-looking reference
Earlier, ADP small nonfarm payrolls and weekly initial jobless claims data were slightly strong, maintaining a narrow consolidation before the nonfarm payroll. If the actual data and expectations deviate significantly, market fluctuations will significantly amplify. 📊 $RE合约爆仓速递（8月7日）
根据爆仓数据，短周期多头被按在地上疯狂摩擦，但中长周期空头直接血崩了。。。
过去1小时爆仓金额约2455.89美元
多单爆仓约2455.89美元
空单爆仓约0美元
过去4小时爆仓金额约5625.23美元
多单爆仓约4598.13美元
空单爆仓约1027.10美元
过去12小时爆仓金额约7.43万美元
多单爆仓约2.36万美元
空单爆仓约5.07万美元
过去24小时爆仓金额约9.46万美元
多单爆仓约3.94万美元
空单爆仓约5.52万美元
从$RE爆仓数据看，1小时多头爆仓碾压空头，空头为零，杀多闪击开局即猛烈；4小时多头优势持续但收窄，比例约4.5倍，杀多全面爆发；12小时方向彻底逆转，空头爆仓碾压多头，空头是多头的2.15倍，逼空全面爆发；24小时空头优势持续，比例约1.4倍，狗庄在RE上完成了从杀多到逼空的凶狠转身——短周期做多的被定向爆破，中长周期做空的被一锅端，累计爆仓突破9.4万美元。大家控制好仓位，别被来回收割。
🔥 市场风向标 | 8月7日
今日三条热点，指向同一主题：市场已进入"预期拉满，瑕疵必杀"的阶段——"超预期"只是及格线，任何增速放缓的信号都会被放大。
💾 存储股财报后下挫：业绩越炸，跌得越狠
闪迪交出"封神级"财报：Q4营收89.65亿美元，同比暴增372%；西部数据同期营收37.47亿美元。SK海力士二季度营收79.32万亿韩元，同比大增557%。
然而，闪迪盘后一度重挫近8%。元凶是指引——下季营收中值105.5亿美元，低于市场预期的108.2亿美元。市场对存储股的定价逻辑，已从"业绩好不好"升级为"增速够不够快"。
AI内存牛市还稳吗？瑞银预测2026年存储行业总收入将达9920亿美元，2027年几乎翻番至1.76万亿美元，HBM是核心驱动力。但短期回调同样真实——截至7月底，AI存储龙头平均回撤约40%；7月SK海力士韩股最大回撤54%，三星电子42%，闪迪单月暴跌47%。超级周期的长期逻辑未破，但估值已跑在基本面之前，任何瑕疵都会被放大。
🏛️ 联储鹰派信号升温：弱就业压不住通胀焦虑
7月FOMC会议出现2016年以来首次三张方向一致的反对票——三位地区联储主席主张加息25个基点。票委卡什卡利甚至表示年内加息三次"并非不可能"。
弱就业能压住通胀吗？7月ADP新增就业仅4.4万人，创1月以来最弱。但薪资增速维持4.4%高位，ISM服务业PMI价格分项飙至70.3，为四个月最高——"就业弱、价格强"形成经典滞胀信号。市场对9月加息的概率仍维持在54.9%。
🚀 SpaceX解禁后反涨：利空出尽的经典剧本
8月6日，SpaceX首批9.115亿股限售股解禁，潜在释放市值约1000亿美元。此前市场普遍预期将触发抛售潮。
结果股价不跌反涨6.14%，收报114.92美元。周三13.6%的财报后暴跌已提前释放了解禁压力；新增卖盘被抄底资金和空头回补有效吸收。市场演绎了"利空出尽即利好"的经典剧本。不过警报未解除——8月20日另有3.19亿股可能解禁，9月预计再释出约7亿股。
💎 总结
闪迪用372%的增长换来暴跌，证明存储股的估值已跑到基本面之前；美联储在弱就业与高通胀之间左右为难，滞胀信号正在浮现；SpaceX用解禁日的逆势反涨，演绎了"利空出尽"的经典剧本。当业绩超预期已成标配，指引的每一分偏差都会被无限放大——旧逻辑正在崩塌，新定价权正在形成，而它惩罚的是所有"不够完美"的答案。#存储股财报后下挫，AI内存牛市还稳吗？
#联储鹰派信号升温，弱就业能否压过通胀？
#财报观察员：解禁后反涨，SpaceX后续怎么看？ On August 6th, the earnings reports from SanDisk and Western Digital caused a collapse in the entire storage sector. #存储股财报后下挫, is the AI memory bull market still stable?
SanDisk's revenue was $8.965 billion, a year-on-year surge of 372% and a quarter-on-quarter increase of 51%, with non-GAAP earnings per share of $39.25.
Western Digital reported revenue of $3.747 billion, up 44% year-on-year, with earnings per share of $3.56.
Both data sets would be legendary in any normal year. Then SanDisk fell 6.8%, Western Digital dropped 13%, and Micron was hit hard.
The problem lies in the guidance. SanDisk's Q1 revenue guidance median was 10.55 billion, below market expectations. Quarter-on-quarter growth dropped from 51% to around 17%. The market wants "continued acceleration," while SanDisk is asking for "slowing growth." This is the essence of the "earnings sell-off"—not poor performance, but market expectations have been pushed to the ceiling.
Wu Hao, fund manager at Founder Fubon Fund, put it bluntly: "The negative factors for the two companies are not in their performance, but in the gap between expectations." Goldman Sachs also said that when market expectations are excessively inflated, even bland guidance can be interpreted as a negative signal. 存储板块现在分歧极大。#存储股财报后下挫，AI内存牛市还稳吗？
花旗说供应链库存低、产能不能满足需求，维持乐观预测。
华夏基金张景松说本轮存储周期跟以前不一样——AI需求持续性极强，价格有望高位企稳。
摩根士丹利说内存合约价Q4见顶，但价格不会随即大跌
仁桥私募夏俊杰说存储行业很可能已经见顶，极度暴利阶段必然短暂。
多空双方都不否认存储巨头业绩强劲，分歧在于这种强劲的持续性。Pave Finance首席投资官说：“盈利爆炸式增长不可持续，更大的问题是利润率能否维持”。
闪迪的NBM长期协议模式是最大的变数。截至Q4签署了8份长期协议，覆盖8家客户，最低合同收入939亿美元。闪迪正在从周期股变成有长期收入可见性的基础设施公司。如果NBM模式跑通，存储周期的逻辑会被改写。但长约在上行周期是护城河，在下行周期可能是枷锁。 亚太市场8月6号同步崩了——铠侠、SK海力士都跌超10%。#存储股财报后下挫，AI内存牛市还稳吗？
全球存储板块同频共振，问题已经推到了台前：AI驱动的存储景气上行，究竟是“在路上”还是“已近终点”？
本轮存储周期的核心变量是HBM。HBM产能基本售罄至2026年，英伟达、微软、亚马逊、Meta持续扩建数据中心。HBM4正在进入批量发货阶段，存储供需紧张的局面短期不会改变。但市场已经提前把2026-2027年的预期打进了股价里。闪迪年内涨幅超460%，西部数据约200%。股价已经预支了未来一两年的增长。闪迪的财报证明AI存储需求真实存在，但需求存在和股价继续涨是两回事。存储板块的这轮回调不是基本面出了问题，是估值和预期的重新校准。AI内存牛市还没结束，但“闭眼买存储股就赚钱”的阶段已经过去了。Real-time Bitcoin and BTC price movements
Daily chart: Price stands above 64,000, above the 50-day (≈63,275) and 200-day moving average (≈61,480), but is being pushed down at the lower edge of the middle Bollinger band (≈64,350); RSI 14≈48–50 is neutral, MACD has a very short golden cross and red bars below the zero axis, oversold recovery is ongoing without a reversal. The large box range at 62,200–64,500 remains intact.
4H: Pushing back to 64,300–64,500, Bollinger closing (upper band ≈ 64,500 / middle band ≈63,500), MACD golden cross histogram shortening; 1H repeatedly above EMA20-50 (63,900–64,100), bulls tried 64,300 but failed, returning to 64,180 for high-level consolidation.
Drivers: U.S.-Iran easing + falling oil prices + lower inflation expectations, Nasdaq helped BTC rebound from 62,250; Spot ETFs saw net inflows of $170 million (IBIT +$111.4 million) on 8/4, ending a three-day withdrawal, but the price rose and positions closed (holding slightly down in 24h), indicating short-term + ETF covering rather than new leveraged long positions, with continuity questionable; Fear and greed 27–31 remain in the fear zone.
Relative to ETH: ETH 1866 is holding flat, ETH/BTC remains at a low of 0.0291, with BTC's dominance unchanged.
Technical level (short-term, not recommended)
Resistance: 64,300–64,500 (4H upper band + previous three-day resistance band; only discuss 65,000 if volume stabilizes) → 65,000–65,300 (psychological barrier + previous daily high) → 66,200 (stronger cyclical pressure)
Divide between bulls and bears: 63,500 (4H mid-band + intraday strength line, also the last line before defending the 50-day moving average at 63,275)
Support: 63,275–63,000 (50-day moving average + round number level) → 62,500 (previous low concentration zone) → 62,200 (if broken, daily candle turns bearish at 61,000)
Rhythm Judgment (Not Recommended)
Hold 63,500 + break 64,300 on increased volume→ test 64,500–65,000, daily chart recovery continues;
64180 repeatedly fails to hold, falling back to 63500 below → pullback to 63275, break further to 62500;
Tonight, 20:15 US ADP + 22:00 ISM Non-Manufacturing is the first repricing window of the week. The 64,000 integer level marks pure sentiment, and stop-losses on both sides of 64,300/63,500 are easily swept by inserted needles.@币圈超短王马大帅 US July Nonfarm Payroll (NFP) Full Analysis | Crypto Market Reference (Published 2026.08.07 20:30) #存储股财报后下挫, Is the AI Memory Bull Market Still Stable?
1. Basic Core Information
1. Announcement time: August 7, 20:30 Beijing time (tonight)
2. Issuing Authority: U.S. Bureau of Labor Statistics
3. Three core indicators (all indispensable: salary > new employment)
(1) Nonfarm Payroll Additions (Core Header Data)
(2) Unemployment rate
(3) Average hourly wages (inflation core, directly affecting Fed policy bias)
4. Benchmark values
◦ Previous value (June): 57,000
◦ Market consensus expectation: 80,000
◦ Unemployment rate forecast: 4.3% (previous value 4.2%)
◦ Hourly wage expectations rose slightly year-on-year
2. The underlying logic behind price movements (common in the crypto community)
The strength of nonfarm payroll data → expectations for Fed rate cuts → the dollar index, fluctuations in US Treasury yields→ and pricing of cryptocurrencies like BTC and ETH
1. Non-farm payrolls far exceed expectations (>100,000 + rising wages) = negative for the crypto sector
Employment data is booming, inflation is hindering a decline, the market lowers the probability of a rate cut in September, the dollar strengthens, capital flees into risky assets, BTC and ETH are rapidly declining, and contracts are prone to chain liquidations.
2. Nonfarm payrolls below expectations (<70,000 + weakening wages) = positive for the crypto sector
Weakening employment reflects economic pressure, with the market betting on the Fed accelerating rate cuts, rising expectations for liquidity easing, and major currencies surging rapidly in the short term.
3. Data matches expectations (75,000–90,000) = wide fluctuations
The divergence between bulls and bears is obvious, with the market oscillating back and forth, and the one-sided trend is weak.
Key misconceptions
Don't focus solely on new jobs: overall employment is stable, but wages have risen sharply, which is still bearish. Wage inflation is a key target for the Fed's control. At the same time, last month's revised employment data will directly reverse the short-term market trend.
3. Market Prediction for Three Scenarios (Current ETH Price about 1900 USDT, BTC Synchronized Linkage)
Scenario 1: Strong data (100,000 new ≥, bearish downtrend)
• BTC first support: 62,200; if it breaks below 60,800, strong support is needed
• ETH short-term support: 1870, 1840
Scenario 2: Weak data (70,000 new ≤, positive news for upward movement)
• BTC resistance: 65,300, 66,100
• ETH short-term resistance: 1930, 1960
Scenario 3: As expected (78,000–88,000 RMB, range-bound)
BTC fluctuation range: 62,500–65,000
ETH volatility range: 1875–1935
4. Forward-looking reference
Earlier, ADP small nonfarm payrolls and weekly initial jobless claims data were slightly strong, maintaining a narrow consolidation before the nonfarm payroll. If the actual data and expectations deviate significantly, market fluctuations will significantly amplify. #财报观察员: After the lock-up lifts, prices rebounded—what do you think about SpaceX's future?
Don't be fooled by the positive results from SPCX. Although revenue increased by 92%, the massive capital expenditure of 18.37 billion yuan (especially AI infrastructure) shocked the market—a classic case of "good news is all bad news."
The real danger is the unlocking event. The first batch of 911.5 million shares has circulated, but the negative news is far from over. Internal chip costs mostly range from 1 to 5 USD, and the current price of 114 is still dozens of times the profit. Never believe in "all negative news has been released"; if you don't sell the tokens with unlocking qualification, you can sell them anytime on the second or third day. That vague phrase "I think so" has caused countless people to hold their shares above 120, and the lesson remains.
The current 114 level is not the bottom; it is a relay of the game where short spots cover on the unlocking day. As soon as there is any sign of movement (such as a macro weakening or the next batch of unlocking announcements), the wait-and-see sentiment within these 900 million stocks will instantly turn into a rushing position, triggering snowball panic. It's not wrong if 105 fails, and 114 is not in a hurry, but the logic must be clear: it's not "positive news for protection," but rather "selling pressure hasn't been fully released." Keep a close eye on the 108-110 support zone; if it breaks, it will rush to previous lows. There will be phased unlocking in late August and September; before that, all rebounds are escape windows provided by insiders, not reversal signals $SPCX $XSPCX Terafab didn't come out of nowhere. #特斯拉SpaceX投建168亿美元AI芯片厂
In April this year, Tesla launched the ATCF wafer fab R&D in Texas as a pre-production phase for Terafab.
Aerial footage of ATCF's construction was revealed in early August, and progress is rapid. The two factories are about 100 miles apart.
Terafab is located in Grimes County, near SpaceX's key operational base in Texas.
Musk said on X: "The number of chips Tesla and SpaceX will need in the future will far exceed what current and future global production capacity can provide." This means—there aren't enough external suppliers, so they have to build themselves.
But $16.8 billion is just the beginning. SpaceX's listing documents show Terafab is still in the overall planning framework, with cumulative investment possibly reaching $119 billion over multiple phases. The financial report just dropped 13.6% due to excessive capital expenditure, but the stock price rebounded 6% on the lock-up day. The market can accept a $16.8 billion Phase I investment, but the $119 billion long-term plan will force the market to recalculate SpaceX's cash-burning speed. If AI chip manufacturers continue to require massive capital injections, SpaceX's free cash flow may remain negative for the long term. This is a short-term narrative catalyst, while the medium- to long-term is a balance sheet pressure.Another signal from Terafab is that Musk is deeply integrating the supply chains of SpaceX and Tesla.
#特斯拉SpaceX投建168亿美元AI芯片厂
The chip needs of both companies have been consolidated into one factory, covering everything from logic chips to DRAM storage to packaging and testing, all under one roof.
An annual capacity of 1 terawatt is 50 times the current global capacity, according to Musk. This target means that if Terafab reaches full production, it could disrupt the current chip supply landscape.
But for Tesla and SpaceX, the real strategic value lies in no longer competing with others for capacity. TSMC and Samsung's capacity schedules have been extended for several years; although building their own factories is expensive, they can ensure supply. Spending 16.8 billion yuan for self-supplied chips is clearly a good deal for Musk.
After the announcement, SPCX's stock price rebounded 6.14% on the unlocking day. The market's pricing is: burning cash is fine, as long as you prove you have long-term competitiveness. Terafab is part of this proof. But the capital market doesn't just look at the narrative; it also looks at cash flow. After spending 16.8 billion, what changes will occur on SpaceX and Tesla's balance sheets is something the market will repeatedly calculate in the coming months.#存储股财报后下挫, is the AI memory bull market still stable?
I'm the mid-term intelligence bro. I saw the drop after the recent financial reports of storage stocks—Micron, SK Hynix, and SanDisk posted explosive earnings but were hit by "earnings blows"—not because the AI memory bull market ended, but because "valuations were pulled back after expectations were too high."
The medium-term logic remains unbroken: the HBM supply-demand gap remains, cloud manufacturers' capital expenditures will remain high year-on-year in 2026, DRAM contract prices are expected to rise more than 10% in Q3, 70% of advanced original capacity is locked in HBM and enterprise-grade, and traditional DRAM is being squeezed out. What is falling is the luxury expectation of "further upward revisions in gross margin," not the underlying demand for "AI needs storage." Morgan Stanley said that the price increase slope slowed in Q4 2026 and the profit revision peaked, which is a feature of the mid-to-late stage of the cycle, not a crash.
My medium-term outlook: the AI memory supercycle isn't over yet, but the blind-eyed, broad-surge phase has passed. Coming up is divergence—if there really are leaders with HBM capacity or long-term contract lock-ups, pullbacks are buying opportunities; Module manufacturers purely speculating on DRAM spot elasticity, beware of Q4 guidance turning hostile. Watch three signals: cloud vendor Capex is downgrading, HBM long-term contract loosening, DRAM contract prices have fallen for two consecutive quarters, and so far none have appeared.
$SNDK
$MU #Uniswap进军发射台, can UNI open up a new narrative?
Robinhood Chain has only been online for a little over a month, and Uniswap has already matched $12.8 billion in trading volume on this chain, making it the second largest market after Ethereum mainnet.
The launch of Pools.trade means Uniswap is no longer just a post-issuance trading venue, but will personally participate in the entire issuance process.
For UNI holders, the most critical factor is the fee switch. The V4 fee switch burns about $90 million of UNI annualized. The larger the trading volume on Pools.trade, the more UNI is burned, and the stronger the expectation of supply deflation. But the question is—how much of the trading volume on Robinhood Chain is sustainable? The launch pad's business is essentially "new token issuance→ short-term hype → replacing the next one." If the issuance slows down, Pools.trade's trading volume will also drop.
The real competition Uniswap faces is—can it establish a launchpad network effect on the Robinhood Chain that surpasses Flap and Pons? On its first day of launch, it launched 6,000 tokens, which is indeed more than the combined volume of three competitors. But the core competitiveness of the launchpad lies in whether it can continuously attract high-quality projects. If it's just a place for launching a patch, its long-term value is limited.$BTC $ETH $SNDK Both SanDisk and Western Digital exceeded expectations in their earnings, but both plunged—one dropped nearly 7 points, the other plunged 13 points. SanDisk's revenue guidance was weak, gross margin showed signs of peaking, Goldman Sachs directly lowered its target price, and Western Digital did the same. The same goes for Western Digital. If there's a slight flaw, capital pours in. The fundamental reason for this round of decline is that market confidence in the AI hardware sector is weakening. Beating expectations is standard; only if it exceeds expectations does it pass. If guidance loosens a bit, it's a sign of respect for the first drop. This isn't a healthy adjustment It's the systematic withdrawal of funds. Storage is one of the core links in AI hardware. The sector remains under pressure, indicating that the overall AI infrastructure narrative is cooling down. Funds are shifting from highly elastic assets toward more certain directions. When risk appetite tightens, all highly elastic assets won't have an easy time. If storage continues to fall, after funds withdraw from AI hardware, they will most likely flow into US Treasuries or cash. The chain reaction of selling off highly elastic assets is not over yet. Is the AI memory bull market stable? Long-term demand has indeed not stopped. HBM is still expanding, and NBM contracts are still being signed But the market doesn't look at the long term; it focuses on marginal changes. Gross margins have peaked, and guidance is weak. These signals are already enough to make funds rush in first. The earnings season isn't over yet; there are still a few more companies to release. It's not too late to wait until the impact of this round of earnings and guidance has been absorbed before watching. Reaching out now is easy to catch halfway up the mountain. #Storagestocks plunged after earnings, is the AI memory bull market still stable? #联储鹰派信号升温, can weak employment suppress inflation? #财报观察员: Rebound after lifting restrictionsPools.trade的零手续费模式很激进。
#Uniswap进军发射台，UNI能否打开新叙事？
竞品收1%-5%的发射费，Uniswap直接打到零。
这不是在做慈善，是在抢市场份额。
Pools.trade不直接在发射环节收费，而是通过V4费用开关捕获价值。项目代币上线后，在Uniswap上交易的手续费部分会被销毁。这意味着Uniswap不需要靠发射费赚钱，靠的是后续的DEX交易量。
这招对竞品来说是降维打击。Flap和Pons依赖发射费收入，Uniswap可以用零手续费抢走他们的项目方和用户，然后靠后续交易量变现。如果这套模式跑通了，UNI的价值捕获逻辑会被重写——不只是DEX交易量，还有发射台的流量入口价值。UNI现价4.14美元，EMA50支撑4.01，MACD金叉0.01维持看涨，RSI 57.24仍有上行空间。市场正在观察Pools.trade能不能把UNI从“治理代币”变成“有真实收入支撑的资产”。 @币圈超短王马大帅 US July Nonfarm Payroll (NFP) Full Analysis | Crypto Market Reference (Published 2026.08.07 20:30) #存储股财报后下挫, Is the AI Memory Bull Market Still Stable?
1. Basic Core Information
1. Announcement time: August 7, 20:30 Beijing time (tonight)
2. Issuing Authority: U.S. Bureau of Labor Statistics
3. Three core indicators (all indispensable: salary > new employment)
(1) Nonfarm Payroll Additions (Core Header Data)
(2) Unemployment rate
(3) Average hourly wages (inflation core, directly affecting Fed policy bias)
4. Benchmark values
◦ Previous value (June): 57,000
◦ Market consensus expectation: 80,000
◦ Unemployment rate forecast: 4.3% (previous value 4.2%)
◦ Hourly wage expectations rose slightly year-on-year
2. The underlying logic behind price movements (common in the crypto community)
The strength of nonfarm payroll data → expectations for Fed rate cuts → the dollar index, fluctuations in US Treasury yields→ and pricing of cryptocurrencies like BTC and ETH
1. Non-farm payrolls far exceed expectations (>100,000 + rising wages) = negative for the crypto sector
Employment data is booming, inflation is hindering a decline, the market lowers the probability of a rate cut in September, the dollar strengthens, capital flees into risky assets, BTC and ETH are rapidly declining, and contracts are prone to chain liquidations.
2. Nonfarm payrolls below expectations (<70,000 + weakening wages) = positive for the crypto sector
Weakening employment reflects economic pressure, with the market betting on the Fed accelerating rate cuts, rising expectations for liquidity easing, and major currencies surging rapidly in the short term.
3. Data matches expectations (75,000–90,000) = wide fluctuations
The divergence between bulls and bears is obvious, with the market oscillating back and forth, and the one-sided trend is weak.
Key misconceptions
Don't focus solely on new jobs: overall employment is stable, but wages have risen sharply, which is still bearish. Wage inflation is a key target for the Fed's control. At the same time, last month's revised employment data will directly reverse the short-term market trend.
3. Market Prediction for Three Scenarios (Current ETH Price about 1900 USDT, BTC Synchronized Linkage)
Scenario 1: Strong data (100,000 new ≥, bearish downtrend)
• BTC first support: 62,200; if it breaks below 60,800, strong support is needed
• ETH short-term support: 1870, 1840
Scenario 2: Weak data (70,000 new ≤, positive news for upward movement)
• BTC resistance: 65,300, 66,100
• ETH short-term resistance: 1930, 1960
Scenario 3: As expected (78,000–88,000 RMB, range-bound)
BTC fluctuation range: 62,500–65,000
ETH volatility range: 1875–1935
4. Forward-looking reference
Earlier, ADP small nonfarm payrolls and weekly initial jobless claims data were slightly strong, maintaining a narrow consolidation before the nonfarm payroll. If the actual data and expectations deviate significantly, market fluctuations will significantly amplify. 8月7号，Alphabet干了一件大事——发了250亿美元的债。
认购规模直接干到1150亿美元，超了四倍多。
这是什么概念？比亚马逊和SpaceX近期发债的认购规模还大。投资者嘴上喊着“AI资本开支太高了”，手上却在疯狂抢AI相关债券。
这不是Alphabet第一次发债了。2月份刚发了200亿，加上瑞郎、英镑、欧元等各种计价债券，上半年债务发售超500亿，股票还发了近850亿。这次10个期限，从2年到40年全覆盖。最长期限的债券收益率比国债高130个基点，初始指导价是155个基点，需求上来之后直接收窄了。
市场情绪很矛盾。一方面，Alphabet刚把资本开支预测上调到2050亿，是2025年的两倍多，直接导致上个季度自由现金流转负。另一方面，投资者对科技债的需求正在回暖，因为Alphabet这次给出了比一般发债高出15个基点的收益率溢价。利息给够了，钱就来了。
Alphabet还明确告诉市场：以后每年固定发两次美元债券。这招是故意安抚市场对科技债供给过剩的担忧。但债务规模越滚越大，最终会侵蚀利润率。AI到底能不能赚回这些钱，才是真正的核心问题。#谷歌母公司发债250亿美元，AI投入压力升温 亚马逊一个月前刚发了250亿，Meta通过贝莱德发了125亿数据中心SPV债券。
而谷歌在7月刚把2026年AI基础设施支出预期上调至2050亿美元。
问题的核心在于——这些钱花出去之后，能赚回来多少？Alphabet上半年已经发了超500亿债务和近850亿股票。以谷歌的信用评级，借钱成本确实很低，250亿的利息支出对每年近2000亿营收的体量来说不算什么。但投资者的疑虑在于：资本开支翻了两倍，营收能翻两倍吗？
亚马逊3月发债时市场反响平平。Alphabet这次给了更高的收益率溢价才把需求推上去。市场对AI债务的态度是——你借钱可以，但得给够利息。投资者不是不信任AI叙事，是开始要求更高的风险补偿了。Alphabet这次的成功发债说明AI债务市场还没有关闭，但融资成本正在上升。#谷歌母公司发债250亿美元，AI投入压力升温 Debt is only part of it; the real pressure lies off the balance sheet.
Last week, BlackRock issued $12.5 billion in bonds linked to Meta's Texas data center SPV.
This structure bypasses the balance sheets of tech companies themselves, shifting risk to special purpose vehicles and bond investors.
Tech companies are using various methods to finance AI infrastructure—on-balance-sheet debt, off-balance-sheet SPV, equity financing. Alphabet's oversubscription of 25 billion yuan bonds shows that the market's confidence in the AI narrative remains. But confidence is fragile. If AI revenue growth can't keep pace with capital expenditure at this time next year, debt pressure will backfire on stock prices. In 2026, AI capital spending is no longer about "whether to invest," but about "whether you can earn it back." Alphabet has played the 25 billion card now, and the market is waiting for it to prove it's not burning cash.
#谷歌母公司发债250亿美元, pressure to invest in AI is intensifying 🐋 Big fish swallow small fish: The first Bitcoin ETF in the US is also not immune to this rule.
Hashdex has just announced the closure and dissolution of its Bitcoin ETF, stopping trading from August 17 and is expected to complete asset liquidation on August 28.
At first glance, many people would think this is a bad sign for Bitcoin. But I see a different story.
This is not a Bitcoin failure, but a fund that is too small to compete in the game of the giants.
Since BlackRock, Fidelity and the big players joined, cash flows have increasingly focused on ETFs with high liquidity, low fees and a scale of tens of billions of dollars. Funds that only manage a few tens of millions of dollars have almost no land to live.
It's also the rule that's playing out across the crypto market:
Big fish attract liquidity.
Liquidity creates more confidence.
Confidence continues to attract more cash flows.
Eventually, the big fish got bigger and bigger, and the small fish gradually disappeared.
In my opinion, the closure of Hashdex is not the end of Bitcoin ETFs, but rather a sign that the market is entering a consolidation phase where only those who are strong enough can survive.
In crypto, your biggest competitors aren't the bear market... it's the "whales" that are sucking up the entire flow of money. 🐋📊 $OKB Contract Liquidation Express (August 7)
According to liquidation data, this wave of short positions was frantically crushed by the Dog Traders...
The liquidation amount in the past hour was about $8,989.30
Long positions were liquidated at about $0
Short positions were liquidated at about $8,989.30
The liquidation amount in the past 4 hours was about $12,000
Long positions were liquidated at about $0
Short positions were liquidated by about $12,000
The liquidation amount in the past 12 hours was about $12,100
Long positions were liquidated at about $0
Short positions were liquidated by about $12,100
The liquidation amount in the past 24 hours was approximately $13,200
Long liquidation is about $1,073.15
Short positions were liquidated by about $12,100
Looking at $OKB liquidation data, short liquidations at 1-hour, 4-hour, and 12-hour levels crushed the bulls, with zero in all longs. The short squeeze was purely one-sided throughout the short-to-medium cycle, with short liquidation steadily rising from $8,989 to $12,000; 24-hour short liquidations still led by a wide margin, but bulls began to resist ($1,073), with bears outnumbering bulls by 11.3 times. Dog Broker completed a full-cycle slaughter of short sellers on OKB—short, medium, and long-term bears were targeted and destroyed from all directions. The only resistance from bulls appeared briefly in the long cycle, but it was just a drop in the bucket, with cumulative liquidations exceeding $13,000. Although the scale is limited, the direction of short pressing is highly consistent. Everyone should control their positions carefully to avoid being bought back.
🔥 Market Barometer | August 7th
Today's three hot topics point to the same theme: the market has entered a stage of "fully loaded expectations, flaws must be eliminated"—"exceeding expectations" is just the passing line, and any signal of slowing growth will be amplified.
💾 Storage stocks plunged after earnings reports: the more explosive the earnings, the harder the drops fell
SanDisk delivers a "legendary" financial report: Q4 revenue reached $8.965 billion, a year-on-year surge of 372%; Western Digital's revenue for the same period was $3.747 billion. SK Hynix's Q2 revenue was 79.32 trillion KRW, a year-on-year increase of 557%.
However, SanDisk's stock plunged nearly 8% after the session. The culprit is the guidance—next quarter's median revenue was $10.55 billion, below the market expectation of $10.82 billion. The market's pricing logic for storage stocks has shifted from "whether the performance is good" to "whether the growth rate is fast enough."
Is the AI memory bull market still stable? UBS forecasts that total storage industry revenue will reach $992 billion by 2026, and nearly double to $1.76 trillion by 2027, with HBM as the core driving force. But short-term pullbacks are equally real—as of the end of July, leading AI storage companies had an average drawdown of about 40%; In July, SK Hynix's Korean stock market saw a maximum drawdown of 54%, Samsung Electronics 42%, and SanDisk plunged 47% in a single month. The long-term logic of the supercycle remains unbroken, but valuations have already outpaced fundamentals, and any flaws will be magnified.
🏛️ Fed hawkish signals heat up: weak employment cannot suppress inflation anxiety
The July FOMC meeting saw three unanimous opposing votes for the first time since 2016—three regional Fed chairs advocated for a 25 basis point rate hike. Voter Kashkari even said that raising interest rates three times within the year is "not impossible."
Can weak employment suppress inflation? In July, ADP added only 44,000 jobs, the weakest since January. However, wage growth remained high at 4.4%, and the ISM Services PMI price sub-index surged to 70.3, the highest in four months—"weak employment, strong prices" signaled classic stagflation. The market's probability of a rate hike in September remains at 54.9%.
🚀 SpaceX rebounds after unlocking restrictions: a classic scenario where all negative news is released
On August 6, SpaceX unlocked its first batch of 911.5 million restricted shares, potentially unlocking a market value of about $100 billion. Previously, the market generally expected a wave of selling to trigger a sell-off.
As a result, the stock price rose 6.14% instead of falling, closing at $114.92. After Wednesday's 13.6% earnings report, the sharp drop has already released the pressure to lift the lockdown; New selling orders were effectively absorbed by bottom-fishing funds and short covering. The market has played out the classic scenario of "when the negative news is gone, good news follows." However, the alert has not been lifted—another 319 million shares may be unlocked on August 20, and about 700 million more shares are expected to be released in September.
💎 Summary
SanDisk's 372% growth led to a plunge, proving that storage stock valuations have moved ahead of fundamentals; The Fed is torn between weak employment and high inflation, with stagflation signals emerging; SpaceX used the unlock-up day to reverse the trend, playing out the classic scenario of "all bad news being gone." When earnings beating expectations becomes standard, every point of guidance deviation is magnified infinitely—the old logic is collapsing, new pricing power is forming, and it punishes all the "imperfect" answers. #存储股财报后下挫, is the AI memory bull market still stable?
#联储鹰派信号升温, can weak employment outpace inflation?
#财报观察员: After the lock-up lifts, prices rebounded—what do you think about SpaceX's future? Don't be scared by ADP's 44,000 — American employment is not cooling off at all; it's just "not hiring, but not firing."
A few days ago, ADP private employment only increased by 44,000, and the whole internet was shouting "job cliff cut, interest rate cuts coming." But the next day, initial jobless claims came out—199,000, lower than the expected 202,000, and the four-week average of 198,800, the lowest since September 2022. Looking at these two together, it's clear: companies aren't laying off employees, they're just too lazy to hire new hires. The continued jobless claims slightly rose to just over 1.8 million, meaning "people who lost their jobs are finding new jobs more slowly," but initial claims were below 200,000, indicating almost no dismissal.
This is the Fed's real trump card for continuing to play hawkish: unemployment is still at 4.2%, layoffs haven't started, wage pressure hasn't eased, inflation stickiness remains—so why should it relax?
So now, the logic of the game has completely shifted course
Previously: weak employment → fear of recession, → easing expectations → risk assets rose.
Right now: slight cooling in employment (on the hiring side) + inflation not dead + initial hiring as steady as old dogs → can't afford a rate cut, let alone a rate hike in September. CME is still pricing in a 55%–64% chance of a 25bp hike in September, not a rate cut.
In the short term, the market is trading not "easing has arrived," but rather a "easing hasn't come, but not to the point of further aggressive increases" — a state of unconditional eating. ADP dropped the probability of a rate hike from 67% to 55%, but the initial policy supported it again—offsetting both sides, with rate expectations stuck at high levels.
A direct translation of the crypto world
BTC has been trading back and forth in the 62,000–65,000 range these days, not because the fundamentals have changed, but because macro expectations have repeatedly shifted between "slowing rate hikes" and "hawks not retreating." Weak employment cannot suppress high inflation, so it cannot hold back the Federal Reserve; If the Fed does not retreat, there is no unilateral dumping of water buffalo.
So don't treat every rebound as a reversal. Before the non-farm payroll (August 7 evening) and next week's CPI release, all rallies are sentiment-driven arbitrage + leveraged turnover, not the starting point of a trend. If the non-farm payroll < 60,000 and the unemployment rate breaks 4.3%, then the probability of a rate hike will truly collapse; If the 80,000 level is mild, the market will continue to be in a box range; If 100,000+, rate hike pricing will pull back, and BTC will test 61,000 support.
How should we proceed at this stage?
Not betting on one-sided or heavily betting on non-farm payrolls is the most expensive clarity today.
• Bullish strategy: Wait until both nonfarm payrolls + CPI are implemented, then buy in batches if the 61,000–62,000 price rebound holds, rather than chasing 64,000+ now
• Bearish strategy: The 65k–65.6k resistance zone is a hedging point with shrinking volume and an upward move; avoid leaving a bare short position in the middle of the box
• Knockoffs: High-beta HYPE/SOL will only get wilder before the macro trend is set. Use small positions for sentiment arbitrage, don't treat them as trend positions
Before macroeconomic conditions take root, all "good news" is a bullish inducement in a shakeout, and "negative news" is also a bearish inducement in a shakeout. Waiting quietly for the non-farm payroll to set a direction is more useful than guessing the Fed's lip movements. #联储鹰派信号升温, can weak employment beat inflation? $BTC $ETH @币圈超短王马大帅 美国7月非农（NFP）全解析｜币圈行情参考（2026.08.07 20:30公布）#存储股财报后下挫，AI内存牛市还稳吗？
一、基础核心信息
1. 公布时间：北京时间8月7日 20:30（今晚）
2. 发布机构：美国劳工统计局
3. 三大核心指标（缺一不可，薪资＞新增就业）
①非农新增就业（核心表头数据）
②失业率
③平均时薪（通胀核心，直接影响美联储政策倾向）
4. 对标基准数值
◦ 前值（6月）：5.7万人
◦ 市场一致预期：8万人
◦ 失业率预期：4.3%（前值4.2%）
◦ 时薪同比预期小幅上行
二、涨跌底层逻辑（币圈通用）
非农就业数据强弱→美联储降息节奏预期→美元指数、美债收益率波动→BTC、ETH等加密货币定价
1. 非农大幅超预期（＞10万+薪资走高）=利空币圈
就业数据火爆，通胀回落受阻，市场下调9月降息概率，美元走强，资金出逃风险资产，BTC、ETH快速下探，合约容易出现插针连环爆仓。
2. 非农低于预期（＜7万+薪资走弱）=利多币圈
就业走弱映射经济承压，市场押注美联储加速降息，流动性宽松预期升温，主流币短线快速拉升。
3. 数据贴合预期（7.5万–9万）=宽幅震荡
多空分歧明显，行情来回插针震荡，单边趋势较弱。
关键误区
不要只看新增就业人数：整体就业平稳，但薪资大幅上涨，依旧属于利空行情，薪资通胀是美联储重点管控目标。同时上月就业数据修正值，会直接扭转盘面短期走势。
三、三种情景盘面预判（现价ETH约1900USDT，BTC同步联动）
情景1：数据强劲（新增≥10万，利空下行）
• BTC第一支撑：62200，跌破下看60800强支撑
• ETH短线支撑：1870、1840
情景2：数据疲软（新增≤7万，利多上行）
• BTC压力：65300、66100
• ETH短线压力：1930、1960
情景3：符合预期（7.8–8.8万，区间震荡）
BTC震荡区间：62500–65000
ETH震荡区间：1875–1935
四、前置前瞻参考
前期ADP小非农、周度初请失业金数据小幅偏强，盘面在非农前维持窄幅整理，一旦实际数据和预期偏差较大，行情变盘波动会明显放大。#联储鹰派信号升温, can weak employment outpace inflation?
Hawkish Fed Meets Weak Employment: This macro showdown is what determines the next wave of crypto rebound
The global market is currently stuck in a very difficult phase—two forces are locked in a fierce battle.
On one hand, employment data is starting to show signs of weakness: in July, ADP private employment only increased by 44,000, compared to the expected 75,000, which was halved to a six-month low, clearly signaling that companies' recruitment engines are stalling.
On the other side, the Fed's mouth is harder than a stone: Moussalem said inflation is more likely to be above target and leans toward rate hikes, Schmid insists "AI investment drives up inflation, policy is not tight," and Washh is reportedly preparing to raise rates in September. CME is now pricing in a 55% probability of a 25bp hike in September, not a rate cut.
When these two things are combined, investors' previous muscle memory of "weak employment = immediate liquidity injection = mindless bulldosing" becomes ineffective.
Previous scenarios: Employment is weak → recession is feared, → rate cut expectations → risk assets rise.
Current scenario: Hiring is cold (ADP 44,000) + layoffs remain stable (initial 199,000 is below expectations, continued hiring is stable) + job-hopping salaries are at a high 7.0% year-on-year + hawkish sentiment from the Federal Reserve → rate cut gates are welded shut, and the probability of rate hikes is actually above 50%.
In other words, weak employment currently cannot suppress high inflation or the Fed's hawkish stance; it can only pull back the "probability of rate hikes" from 66% to 55%, then be rejected by initial hiring and wage data.
So what does this mean for the crypto world? Don't automatically translate "weak employment" as "BTC rises."
If we really want to reach the "next round of rebound" you mentioned, we have to meet this chain:
Subsequent CPI + non-farm payrolls will remain soft (nonfarm < 60,000, unemployment rate breaks 4.3%, hourly wages fall below 0.2%)→ rate hike pricing collapses, short-term US Treasury yields fall, dollar weakness → global liquidity expectations recover, → institutions will reclaim BTC as a "scarce + highly liquid" macro hedge, ETH will be more flexible with DeFi + RWA + ecosystem funds, and SOL/LINK/AI will rotate with increased risk appetite.
This chain is currently not connected. Currently, the Fed is in a tug-of-war period of "continued hawkish Fed stance vs. marginal weakening employment." BTC is stuck between 61,000 and 65,000 units washing leverage, and ADP had a 24-hour amplitude of 9.65% for two-way liquidation. This is a typical "macro expectations surge and chip swaps" volatility.
Conversely, if next week's CPI rebounds, tonight's nonfarm payrolls reach 80,000+ units, and hourly wages remain firm, the Fed will set prices for "higher and longer" again. Nasdaq will be the first to fall, BTC testing the 61,000 support, ETH relatively resilient but still unable to escape a range, and counterfeit high-beta (HYPE/SOL/TAO) volatility will only get wilder.
So your original statement that "the bull market outlook is far from certain" is correct, but you need to add one more thing: the current situation is not a bull-bear turning point, but a compression spring before direction choice. Historically, a major Fed pivot (the turning point from hawkish to dovish) was indeed the starting point of a new cycle, but before the turning point appeared, all rebounds were "sentiment arbitrage + leveraged turnover."
• Do not chase bulls at 64,000+; wait for intangible cultural heritage + CPI to set the tone and then pull back to 61,000–62,000 before breaking below 61,000 or 62,000 before splitting them up;
• ETH is stronger than BTC because ecosystem funds haven't withdrawn, but don't treat it as an independent bull—macro conditions haven't loosened, it can rotate and outperform at best;
• SOL/LINK/AI tokens are the new money trendsetter, but only if the Fed's pricing really shifts from "55% rate hikes" to "rate cuts and repricing"; otherwise, it's high volatility grinding.
In short: the showdown between the hawkish Fed and weak employment—tonight is the first round of the nonfarm payrolls, next week's CPI is the second round, and the September FOMC will be the final whistle. Before that, BTC's next major move isn't 'whether it will rise,' but 'which support level will be washed out first, then choose the direction for macros.' Don't use every needle-shaped rebound in the middle as a reason for holding positions.
$BTC $ETH @币圈超短王马大帅 US July Nonfarm Payroll (NFP) Full Analysis | Crypto Market Reference (Published 2026.08.07 20:30) #存储股财报后下挫, Is the AI Memory Bull Market Still Stable?
1. Basic Core Information
1. Announcement time: August 7, 20:30 Beijing time (tonight)
2. Issuing Authority: U.S. Bureau of Labor Statistics
3. Three core indicators (all indispensable: salary > new employment)
(1) Nonfarm Payroll Additions (Core Header Data)
(2) Unemployment rate
(3) Average hourly wages (inflation core, directly affecting Fed policy bias)
4. Benchmark values
◦ Previous value (June): 57,000
◦ Market consensus expectation: 80,000
◦ Unemployment rate forecast: 4.3% (previous value 4.2%)
◦ Hourly wage expectations rose slightly year-on-year
2. The underlying logic behind price movements (common in the crypto community)
The strength of nonfarm payroll data → expectations for Fed rate cuts → the dollar index, fluctuations in US Treasury yields→ and pricing of cryptocurrencies like BTC and ETH
1. Non-farm payrolls far exceed expectations (>100,000 + rising wages) = negative for the crypto sector
Employment data is booming, inflation is hindering a decline, the market lowers the probability of a rate cut in September, the dollar strengthens, capital flees into risky assets, BTC and ETH are rapidly declining, and contracts are prone to chain liquidations.
2. Nonfarm payrolls below expectations (<70,000 + weakening wages) = positive for the crypto sector
Weakening employment reflects economic pressure, with the market betting on the Fed accelerating rate cuts, rising expectations for liquidity easing, and major currencies surging rapidly in the short term.
3. Data matches expectations (75,000–90,000) = wide fluctuations
The divergence between bulls and bears is obvious, with the market oscillating back and forth, and the one-sided trend is weak.
Key misconceptions
Don't focus solely on new jobs: overall employment is stable, but wages have risen sharply, which is still bearish. Wage inflation is a key target for the Fed's control. At the same time, last month's revised employment data will directly reverse the short-term market trend.
3. Market Prediction for Three Scenarios (Current ETH Price about 1900 USDT, BTC Synchronized Linkage)
Scenario 1: Strong data (100,000 new ≥, bearish downtrend)
• BTC first support: 62,200; if it breaks below 60,800, strong support is needed
• ETH short-term support: 1870, 1840
Scenario 2: Weak data (70,000 new ≤, positive news for upward movement)
• BTC resistance: 65,300, 66,100
• ETH short-term resistance: 1930, 1960
Scenario 3: As expected (78,000–88,000 RMB, range-bound)
BTC fluctuation range: 62,500–65,000
ETH volatility range: 1875–1935
4. Forward-looking reference
Earlier, ADP small nonfarm payrolls and weekly initial jobless claims data were slightly strong, maintaining a narrow consolidation before the nonfarm payroll. If the actual data and expectations deviate significantly, market fluctuations will significantly amplify. Crypto Daily · 2026.08.07 Friday
1. One-sentence summary today
BTC is sideways around 64K, neither bulls nor bears have strength, this kind of market is the most exhausting.
2. Market thermometer
Neutral to bearish
Bulls are facing a paper loss of nearly 40 million, and volume can't support any story.
3. Core market today
BTC: $64,366 | 24h -0.28% | No drop or rise, but bulls are overall losing money, I can't say this level is stable
ETH: $1,902 | 24h +0.18% | Barely in the green, but sticking above the 1900 integer level, not exactly strong
Strongest sector today: US tokenized assets | SPCX | +5.7%
Weakest sector today: Korean tech stock tokens | SKHYNIX | -3.1%~-4.5%
4. Most important news today
[Wintermute US business obtains broker-dealer license]
Confirmed by WSJ, Wintermute's US entity has obtained a broker-dealer license, allowing compliant trading of crypto ETFs, commodities, and tokenized stocks.
[Impact] Positive for market sentiment in the short term, indicating institutional compliance channels are opening; mid-term, this type of market maker entry will bring more liquidity to the tokenized asset sector.
[My judgment] Market reaction is insufficient. This news is more important than most realize—Wintermute is a leading market maker, not a small player; obtaining the license means the infrastructure for compliant market making is being completed. But today's market showed little reaction, possibly because everyone is waiting for Friday's close.
[KMNO listed on Upbit + Bithumb spot, price surged 20% in short term]
Kamino Finance token launched simultaneously on South Korea's two major exchanges this morning, OI rose 71.8% synchronously, price soared 20% within an hour.
[Impact] Pure short-term event-driven, clear Korean premium effect, but market cap only 148 million, liquidity limited.
[My judgment] This kind of listing pump is typical; retail investors chasing in will likely be left holding the bag. Good news realized is bad news, an old saying, but people always forget.
[Iran strikes "hostile targets" in the Strait of Hormuz]
According to Fars News Agency, Iran took military action in the Strait of Hormuz, details pending confirmation.
[Impact] Hormuz is a global oil transport choke point; if the situation escalates, crude oil tokens (CL/BRENTOIL) will move first, then risk-off sentiment will transmit to BTC.
[My judgment] Currently, the source is single and market reaction is mild. I tend to observe first, not to rush into positions because of this news, but it’s worth monitoring.
5. Signals worth attention today
Signal: BTC bulls’ paper loss exceeds $63 million, net PnL -$38.9 million, but bull-to-bear ratio remains at 1.43:1
Why it matters: Bulls haven’t massively withdrawn despite losing money, meaning either they are holding hard or someone is quietly accumulating at low levels—these two scenarios have very different outcomes
Tracking period: short term
Signal: SPCX (S&P 500 synthetic token) today OI $238 million, price +5.7%, while the actual S&P 500 US stock fell 0.2% in the same period
Why it matters: On-chain S&P synthetic asset gains far exceed the underlying stock, indicating leveraged long exposure to US stocks; this divergence is somewhat odd
Tracking period: short term
6. Key events tomorrow
🕐 Tonight (20:30 Beijing time) US Nonfarm Payroll data → Expected impact: highly uncertain, if nonfarm exceeds expectations strongly, it will suppress rate cut expectations and be bearish for crypto; if weak, the opposite. Today is Friday, nonfarm day, BTC will likely continue sideways before the data release.
📅 Next Monday follow-up on Iran Hormuz situation → Expected impact: neutral to bearish, if there is escalation news over the weekend, crude oil and safe-haven assets will move first at Monday open.
7. Today's view
What I want to emphasize most today is the Wintermute news. Everyone is watching whether BTC goes up or not, but the truly valuable information is that infrastructure is being built piece by piece. Market makers getting licenses, tokenized stock compliance channels opening—these won’t make you money today, but in two years you’ll know what happened today. Adults only look at pros and cons; short-term sideways is exhausting, but the direction hasn’t changed. Can Tim Scott really turn the tide? Can the Republicans really eliminate dissenters and then find enough Democratic votes to start voting?
As the Clarity Act arrived on Friday and the Senate summer recess window approached, Tim Scott offered a new perspective: he said Republicans would definitely hold a procedural vote on the Clarity Act before the recess, and even suggested the Senate would decide to delay the adjournment
Tim Scott is considered Trump's temporary mouthpiece, as he expressed this week that he could delay the adjournment to allow a procedural vote on the Clarity Act, aiming to buy enough time for the bill. However, whether the Senate can delay the recess will depend on Speaker Thune's stance
So far, Thune has not made a clear statement on the matter, so whether the Senate can delay the recess remains unknown. The market $BTC has not optimally priced in prices, with prices remaining volatile, and the probability of the pre-recess procedure passing on Polymarket is as low as 17%
Clearly, Tim Scott's viewpoint seems optimistic, but its actual driving force is still lacking verification. The current challenge for the Republicans is to first reach a high point within the party, and then persuade certain Democrats to initiate the process, but for now, the possibility remains very low
At this stage, the focus is not on the Clarity Act, but on whether Senate Majority Leader Thune announces a delayed recess. If this issue is not resolved, the Clarity Act will be difficult to enter the Senate process!
#联储鹰派信号升温, can weak employment outpace inflation? Potential demon coin: AOB all holding addresses and D0–Dx change analysis.
Data was pulled through a crawler, cleaned, and analyzed, revealing some interesting data changes. No more words, take a look yourself~ Brothers, now is the critical moment
1. Employment data has clearly weakened
Expectations for rate cuts have been ignited, and risk assets are theoretically benefiting from positive developments. But the biggest trap lies here: if the data is too bad to reach a limit, the market will start trading for a "recession," and instead, the market will see a plunge as positive news materializes. Good data doesn't necessarily mean prices will rise, and bad data won't necessarily fall—this is the easiest place for non-farm payrolls to exploit people. #联储鹰派信号升温, can weak employment outpace inflation?
2. #Circle财报后押注Arc, can USDC experience new growth? Circle is trying to break free from reliance on single interest income, transforming into a platform-based infrastructure layout by launching Arc blockchain and payment networks, but whether this new growth engine can truly support valuations still requires further market validation.
3. Although the storage giant's earnings report was impressive, the stock price plunged—not due to a collapse in AI demand, but because previous gains overloaded expectations, with funds taking profits.
Although the gap between bulls and bears is intensifying, the structural demand for high-end memory in AI computing power remains strong, and the shortage pattern is difficult to reverse in the short term.
The era of broad growth in the storage sector has ended, and the next phase will enter a period of differentiation validation. Only manufacturers truly bound to high-end AI orders can continue to enjoy premiums. #存储股财报后下挫, is the AI memory bull market still stable?
Practical advice for ordinary traders
(1) Don't gamble on large positions at the moment data is released; in the first few dozen minutes, most market moves are inserted to lure bulls and bears, making it hard to distinguish truth from falsehood. Patiently wait 15-30 minutes; only after the market has digested the noise will the true direction emerge.
(2) Do not mistake short-term data fluctuations for changes in medium- to long-term trends. A month's employment report can't change the big cycle.
(3) Stock selection > the market now. Even if the index doesn't move, some main themes will still show market trends; Conversely, even if the index rebounds, many weak stocks still cannot outperform.
Personal Market Thoughts:
🥇 $BTC — Control overall market liquidity and determine the overall long sentiment on the market
🏧 $ETH — Chips are continuously accumulating, moving steadily to build up strength
🚀 $SOL — The Layer 1 sector represents high elasticity, showing explosive momentum when the market arrives
🧠 $TAO & $WLD — The AI story continues to ferment, repeatedly receiving funding support
📊 $HYPE — Used to observe the market's overall willingness to take on risk
🐾 $DOGE & $ZEC — Intuitively reflects the bullish and bearish sentiment of retail investors
💵 Aggressive Directions for Concentrated Capital:
$JTO • $JELLY • $BTC • $OPG • $BTCSLX • $LAB • $BSB • $ALLO • $CHIP
🇺🇸 Key U.S. stock tracking observations:
$MU • $SPCX • $SNDK • $SKHY • $CL • $XAU • $NITC • $AMD
📉 Funds retreat, upward momentum exhausted Stocks:
$BEAT • $EDGE • $COAI • $TRUMP • $RAVE • $SPACE • $SOPH • $IP • $AVNT • $ZAMA • $OFC • $PIEVERSE • $VIRTUAL • $ACU • $H • $MEGA
🔎 Waiting for the signal to confirm the alternative pool:
$MEME • $EDEN • $HUMA • $ZKP • $METIS On the day SpaceX's 100 billion yuan unlock was lifted, it actually rose 6%. The real "unlocking shipments" may have happened yesterday.
On August 6, SpaceX's first batch of 911.5 million restricted shares was unlocked, bringing the potential market value release close to $100 billion.
The market originally expected a large number of internal shareholders cashing out → the stock price would come under pressure.
The result was completely the opposite.
SpaceX opened up about 5% and closed up 6.14%, with a trading volume of 251 million shares, the highest since June 18.
Why?
Because real selling pressure may have already been released early on Wednesday.
On Wednesday, SpaceX's first earnings report showed over $18 billion in AI capital investment, causing market concerns. Its stock price plunged nearly 14% in a single day, with a trading volume of about 200 million shares.
On this day, many funds that had originally planned to be sold on the lock-up day completed their trades ahead of schedule.
In other words:
Wednesday may be the real "release date."
The lifting of the ban on August 6 feels more like a reverse stress test for the bears.
Currently, SpaceX's short position accounts for about 36% of the outstanding shares, with cumulative book gains exceeding $9 billion.
Many bears are betting not on SpaceX's fundamentals collapsing, but on:
"After the lock-up is lifted, will employees and early investors concentrate on selling?"
But the reality is:
Unlocking ≠ will definitely sell.
Early investors still hold huge unrealized gains, and many may choose to hold on.
What's even more noteworthy is that during Wednesday's crash, retail investors actually bought in large quantities.
Vanda data shows that retail investors net bought about $22.7 million in SpaceX in the first hour of trading on Wednesday, three times the average.
Institutions are seeing:
AI capital spending surges, putting short-term profit pressure on the market.
What retail investors see is:
Musk is turning SpaceX into a player in AI infrastructure.
But risks remain.
This unlock is only the first step in the nine-stage unlock mechanism.
The real test will come in June 2027—when about 6.4 billion Class A shares of Musk will be released.
So, in the short term:
The negative news from the unlocking has not materialized, which may instead trigger short covering.
In the long term:
The market ultimately returns to one question:
SpaceX invests tens of billions of dollars in AI—can it create a new profit growth curve?
That's what determines how much it's worth.
$SPCX $RKLB $LUNR $ASTS $RDW #美股📊 $XAUT Contract Liquidation Express (August 7)
According to liquidation data, short-cycle bears are being crushed wildly, but long-term bulls have collapsed instantly...
The liquidation amount in the past hour was about $38.52
Long positions were liquidated at about $0
Short positions were liquidated at about $38.52
The liquidation amount in the past 4 hours was approximately $7,741.48
The long position liquidation was about $181.16
Short liquidations were about $7,560.32
The liquidation amount in the past 12 hours was approximately $9,238.47
Long position liquidation was about $1176.52
Short positions were liquidated at about $8,061.95
The amount of liquidation in the past 24 hours was approximately $58,600
Long positions were liquidated by about $34,400
Short positions were liquidated by about $24,300
From $XAUT liquidation data, within 1 hour, short liquidations crushed the bulls, with short sellers monopolizing everything, starting with a short squeeze flash; The 4-hour short advantage continued, with bears outnumbering bulls 41 times, leading to a full burst of short squeezing; 12-hour bears still far ahead, with a ratio of about 6.85 times, with short squeezes running through the short to medium cycle; 24-hour direction completely reversed, with long liquidations crushing bears, bulls 1.42 times the shorts. On XAUT, Dog Maker completed a fierce turnaround from short squeezing to long selling—short-term short chases were targeted and destroyed, long-term long chasers were wiped out at once, with cumulative liquidations exceeding $58,000. As a gold stablecoin, XAUT switches direction very decisively, making it a textbook-level short squeeze turning long selling. Everyone should control their positions to avoid being recovered and cut off.
🔥 Market Indicator | August 7
Today's three hot topics point to the same theme: the market has entered a stage of "fully loaded expectations, flaws must be eliminated"—"exceeding expectations" is just the passing line, and any signal of slowing growth will be amplified.
💾 Storage stocks plunged after earnings reports: the more explosive the earnings, the harder the drops fell
SanDisk delivered a "legendary" financial report: Q4 revenue was $8.965 billion, a year-on-year surge of 372%; Western Digital reported $3.747 billion in revenue during the same period. SK Hynix's Q2 revenue was 79.32 trillion KRW, a year-on-year surge of 557%.
However, SanDisk's stock plunged nearly 8% after hours. The culprit is the guidance—next quarter's median revenue is $10.55 billion, below the market expectation of $10.82 billion. The market's pricing logic for storage stocks has shifted from "good performance" to "whether growth is fast enough."
Is the AI memory bull market stable? UBS forecasts total storage industry revenue to reach $992 billion in 2026 and nearly double to $1.76 trillion by 2027, with HBM as the core driving force. But short-term corrections are equally real—as of the end of July, leading AI storage companies had an average drawdown of about 40%; In July, SK Hynix's Korean stock market saw a maximum drawdown of 54%, Samsung Electronics 42%, and SanDisk plunged 47% in a single month. The long-term logic of the supercycle has not been broken, but valuations have already outpaced fundamentals, and any flaws will be magnified.
🏛️ Fed hawkish signals heat up: weak employment cannot suppress inflation anxiety
At the July FOMC meeting, three unanimous votes were cast against the same direction for the first time since 2016—three regional Fed chairs advocated for a 25 basis point rate hike. Voting member Kashkari even said that raising rates three times within the year is "not impossible."
Can weak employment suppress inflation? In July, ADP added only 44,000 new jobs, the weakest since January. However, wage growth remained high at 4.4%, and the ISM Services PMI price sub-index surged to 70.3, the highest in four months—"weak employment, strong prices" forms a classic stagflation signal. The market's probability of a rate hike in September remains at 54.9%.
🚀 SpaceX rebounds after unlocking restrictions: a classic scenario where all negative news is released
On August 6, SpaceX unlocked its first batch of 911.5 million restricted shares, potentially releasing a market value of about $100 billion. Previously, the market generally expected a wave of sell-offs.
As a result, the stock price did not fall but instead rose 6.14%, closing at $114.92. After Wednesday's 13.6% earnings report, the plunge had already released the pressure to lift the lock; New selling orders were effectively absorbed by bottom-fishing funds and short covering. The market played out the classic scenario of "when all the negative news is gone, good news." However, the alarm was not lifted—another 319 million shares may be unlocked on August 20, and about 700 million more are expected to be released in September.
💎 Summary
SanDisk traded 372% growth for a plunge, proving that storage stock valuations have moved ahead of fundamentals; The Fed is torn between weak employment and high inflation, with stagflation signals emerging; SpaceX played out the classic scenario of "all negative news being exhausted" by rallying against the trend on the unlocked day. When earnings beating expectations becomes standard, every point of guidance deviation is magnified infinitely—the old logic is collapsing, new pricing power is forming, and it punishes all the "not perfect" answers. #存储股财报后下挫, is the AI memory bull market still stable?
#联储鹰派信号升温, can weak employment outpace inflation?
#财报观察员: After the lock-up lifts, prices rebounded—what do you think about SpaceX's future? $BTC #存储股财报后下挫, is the AI memory bull market still stable? There are three core reasons behind the storage crash
1. It's not that prices are stopping, but that the price isn't as strong as before, and the market has flipped and dumped the market
Previously, the surge in memory chips was due to the market assuming AI would face a constant shortage of capacity, prices would soar monthly, and profits would explode endlessly.
But the data SanDisk provided for the next quarter is, to put it bluntly, that they're still making big money, but the pace of price increases has slowed.
Consumer demand has long been weak, relying entirely on AI to keep things going.
Now, AI can't support expectations of an "unlimited surge." As soon as growth slows, the capital market completely disregards the original high valuations and plunges.
Simply put: you can make money, but you can't make money faster and the funds will disappear.
2. The industry is fine, but the hype logic has completely changed
To be honest, long-term demand for AI storage remains strong, and the industry fundamentals have not collapsed at all.
But Wall Street's mindset has completely changed:
In the past, prices could rise with empty promises; now they only look at real profits and do not accept any expected cooldown.
Previously, prices rose too much and expectations were overhyped; now, as long as future growth isn't as dramatic as imagined,
Funds directly "crashing expectations and crashing valuations" is simply unreasonable.
3. This major drop is not a global crash, but purely a case of high-level funds switching positions and fleeing
Don't get me wrong:
It's not that the entire U.S. stock market is struggling, nor is the economy dooming.
It's just that storage and AI hardware have skyrocketed before, doubling several times, and profits are too full.
Major funds are cashing out at high levels, fleeing from AI hardware, and some blue-chip assets are stabilizing as safe havens.
Some are preparing to find new, highly elastic tracks.
It's just a rotation in the track, not a market panic.
The ultimate one-sentence summary
The storage industry is still profitable, and AI demand remains,
But the era of explosive growth has ended and the era of steady growth has arrived,
The capital market does not accept "slowing growth" and directly restores all previously overdrawn bubbles.
Let's still focus on the macro employment data: this week's true crossroads, three scenario simulations
1. Scenario (1): Rising initial jobless claims and sharp weakness in nonfarm payrolls (positive for cryptocurrency)
With employment cooling and the market betting on an earlier Fed rate cut, falling US Treasury yields are positive for Bitcoin and growth assets, with BTC having the opportunity to test the resistance range between 66,000 and 67,000.
2. Scenario (2): Employment data beats expectations (obviously bearish for crypto)
Labor force remains hot, rate cut expectations have been directly delayed, liquidity tightening expectations are being repriced, and there is a risk of breaking below the 64,000 support level.
3. Scenario (3): Data is unbiased, mixed (shock script)
The market continues to play in place, with BTC oscillating between 63,000 and 65,500 for a long time, waiting for the next round of news to trigger the market.Circle invited Wall Street into the node seat, and the new throne for stablecoins is set
Revolutions always start with a beautiful word.
Open, permissionless, and decentralized.
When Qian actually entered, the entrance was already packed with people in suits.
Circle's stablecoin public chain Arc is approaching mainnet launch, with traditional financial institutions such as BlackRock, Visa, Mastercard, DTCC, Standard Chartered Bank, and Intercontinental Exchange beginning to rally around this chain.
It looks like Wall Street is embracing blockchain.
Looking at it from another angle?
It could also be that Wall Street is preparing to take over the settlement layer of blockchain.
Arc is not designed to help retail investors find the next hundredfold coin. It targets stablecoin payments, forex, tokenized securities, and institutional clearing. The network uses stablecoins to pay gas fees, with transaction outcomes emphasizing predictability, and permissioned validator nodes will be used in the early stages. According to Arc's announced deployment plan, the network is expected to have only about 20 permissioned validators at launch.
In the past, when Circle issued USDC, it had to borrow public chains like Ethereum and Solana to complete circulation. Every transfer and every financial application generated value was partly left on someone else's land.
After Arc launched, Circle could print money and build roads; It set access rules and sold tickets to institutions to enter the new financial network.
#Circle财报后押注Arc, can USDC experience new growth? 英伟达下一代 Rubin Ultra 显存传出因产能挤压而缩减配额，算力前端的冲锋最终卡在了上游存储的物理墙前。
美股存储板块在消息落地后出现回调，存储核心标的 $MU 随大盘科技股高位震荡，完成了一轮短期获利盘的清理。
宏观层面上美联储利率预期反复扰动高估值资产，美元走势趋缓之际，资金开始在加密资产与美股之间寻找抗波动的确定性底座。
高级 HBM 产能短缺向全 DRAM 供应链的溢价传导，将宏观利率环境带来的短期估值压力转化为上游芯片厂商的实质定价权。
若美股科技板块风险偏好回升且下游 AI 资本开支继续兑现，DRAM 价格上涨将推动存储标的重构向上通道，而利率高企导致的整体折现率抬升会推迟突破时间。
若下游硬件采购预算整体收缩，存储芯片溢价能力将面临修正，但在高级存储产能彻底释放前，供给端瓶颈会限制整体下行幅度。
美股硬件链与加密市场的跨市场联动表明，当宏观流动性受阻时，拥有实体产能瓶颈防线的资产往往展现出更强的价格韧性。
未来七天最值得观察的变量，是 DRAM 现货价格是否在科技股整体波动中保持独立冲高态势。
#闪迪财报双超预期，新增140亿美元回购授权 #黄金4200美元拉锯，BTC为何没跟涨？Guys, I understand this round of storage sell-offs. The old logic was "chip shortage = price hikes = good news," but now it's "too tight = stuck downstream = bad news." SanDisk's earnings report was clearly terrible, but because it didn't do so badly, it was completely crushed. Citibank even decisively slashed the target price. This is called the "high expectations prison"—if you want to break free, you have to get full marks. Now Nvidia is even considering downgrading HBM, which shows the price is really hurting demand. Hold on in the short term, don't be a hero, wait until valuations squeeze out the specs before seeing. #存储股财报后下挫, is the AI memory bull market still stable?
$SNDK $XSNDK 📊 $NEAR Contract Liquidation Express (August 7)
According to liquidation data, this bull market was frantically rubbed by the bull market...
In the past hour, liquidations amounted to about $55,400
Long positions were liquidated at about $55,400
Short liquidation is about $0
The liquidation amount in the past 4 hours was about $78,800
Long positions were liquidated by about $78,800
Short liquidation is about $0
The liquidation amount in the past 12 hours was approximately $109,400
Long positions were liquidated at about $109,400
Short positions were liquidated at about $18.43
The liquidation amount in the past 24 hours was approximately $324,000
The long position liquidation was about $322,100
Short positions were liquidated at about $1936.57
According to $NEAR liquidation data, 1-hour and 4-hour long liquidations crushed the bears, with zero bears, and the long sell-off rally unfolded with nuclear explosion-level intensity in the short cycle; The 12-hour bullish advantage persists, with bulls at 5,937 times the bearish position, and selling long positions runs through the short-to-medium cycle; In 24 hours, long positions were liquidated to $322,100, 166 times that of short positions. Dog Farm completed a full-cycle slaughter of bulls on NEAR—short, medium, and long-term bulls were targeted and destroyed from all directions. The only resistance from bears appeared slightly in the long cycle, but it was just a drop in the bucket, with cumulative liquidations surpassing $320,000. Everyone should control their positions carefully to avoid being bought back.
🔥 Market Barometer | August 7th
Today's three hot topics point to the same theme: the market has entered a stage of "fully loaded expectations, flaws must be eliminated"—"exceeding expectations" is just the passing line, and any signal of slowing growth will be amplified.
💾 Storage stocks plunged after earnings reports: the more explosive the earnings, the harder the drops fell
SanDisk delivers a "legendary" financial report: Q4 revenue reached $8.965 billion, a year-on-year surge of 372%; Western Digital's revenue for the same period was $3.747 billion. SK Hynix's Q2 revenue was 79.32 trillion KRW, a year-on-year increase of 557%.
However, SanDisk's stock plunged nearly 8% after the session. The culprit is the guidance—next quarter's median revenue was $10.55 billion, below the market expectation of $10.82 billion. The market's pricing logic for storage stocks has shifted from "whether the performance is good" to "whether the growth rate is fast enough."
Is the AI memory bull market still stable? UBS forecasts that total storage industry revenue will reach $992 billion by 2026, and nearly double to $1.76 trillion by 2027, with HBM as the core driving force. But short-term pullbacks are equally real—as of the end of July, leading AI storage companies had an average drawdown of about 40%; In July, SK Hynix's Korean stock market saw a maximum drawdown of 54%, Samsung Electronics 42%, and SanDisk plunged 47% in a single month. The long-term logic of the supercycle remains unbroken, but valuations have already outpaced fundamentals, and any flaws will be magnified.
🏛️ Fed hawkish signals heat up: weak employment cannot suppress inflation anxiety
The July FOMC meeting saw three unanimous opposing votes for the first time since 2016—three regional Fed chairs advocated for a 25 basis point rate hike. Voter Kashkari even said that raising interest rates three times within the year is "not impossible."
Can weak employment suppress inflation? In July, ADP added only 44,000 jobs, the weakest since January. However, wage growth remained high at 4.4%, and the ISM Services PMI price sub-index surged to 70.3, the highest in four months—"weak employment, strong prices" signaled classic stagflation. The market's probability of a rate hike in September remains at 54.9%.
🚀 SpaceX rebounds after unlocking restrictions: a classic scenario where all negative news is released
On August 6, SpaceX unlocked its first batch of 911.5 million restricted shares, potentially unlocking a market value of about $100 billion. Previously, the market generally expected a wave of selling to trigger a sell-off.
As a result, the stock price rose 6.14% instead of falling, closing at $114.92. After Wednesday's 13.6% earnings report, the sharp drop has already released the pressure to lift the lockdown; New selling orders were effectively absorbed by bottom-fishing funds and short covering. The market has played out the classic scenario of "when the negative news is gone, good news follows." However, the alert has not been lifted—another 319 million shares may be unlocked on August 20, and about 700 million more shares are expected to be released in September.
💎 Summary
SanDisk's 372% growth led to a plunge, proving that storage stock valuations have moved ahead of fundamentals; The Fed is torn between weak employment and high inflation, with stagflation signals emerging; SpaceX used the unlock-up day to reverse the trend, playing out the classic scenario of "all bad news being gone." When earnings beating expectations becomes standard, every point of guidance deviation is magnified infinitely—the old logic is collapsing, new pricing power is forming, and it punishes all the "imperfect" answers. #存储股财报后下挫, is the AI memory bull market still stable?
#存储股财报后下挫, is the AI memory bull market still stable?
#财报观察员: After the lock-up lifts, prices rebounded—what do you think about SpaceX's future? $XAU 4000 is supported by the central bank, $BTC Strategy continues to sell coins
The current macro environment has amplified the differentiation,
The real interest rate and liquidity environment in 2026 will not be friendly to risk assets, so gold will mainly be independently supported by central bank gold purchases and safe-haven demand. The result is: gold can be held at high levels through "institutional buying," while BTC must wait for liquidity easing to ease again or risk appetite to pick up before it can truly catch up.
Simply put, gold has sovereign-level, price-insensitive buying support, while gold's "muscle memory" and institutional status remain stronger. Historically, BTC has significantly outperformed gold during liquidity expansion periods but often lost in the early stages of tightening or risk events.
Currently, short-term pricing power lies with risk appetite and liquidity.
Image: Comparison of BTC and gold trends over the past year
#黄金4200美元拉锯, why hasn't BTC risen in tandem? Talking about some solid stuff, those who know, know. $SPCX is up 6% today against the trend, and behind it is a bigger narrative: SpaceX is building its own gas power plant and a massive battery array for the semiconductor giant Terafab in Texas, generating and storing its own electricity. This is no longer just about making rockets; Musk is moving the entire "energy—chip—AI" chain into his own backyard. With vertical integration to this extent, the valuation anchor changes. Don't just focus on the daily ups and downs; watch how this chain develops.#AIMemoryBullTest #DailyOrbit #Alphabet25BBond #Storage stocks plunge after earnings reports, is the AI memory bull market still stable?
Last night, I checked the earnings reports of the storage sector, and my blood pressure shot up immediately. How good do you think the performance has to be to cause a rise? The result tells you: even if it's explosively good, as long as it’s not out of this world, it will still be crushed down to the ground.
First, look at these numbers, they’re simply outrageous. Sandisk just released its earnings report, with revenue hitting $8.97 billion, nearly a fivefold increase year-over-year! Gross margin at 84.6%, this money printer is running slower than it. Western Digital isn’t bad either, with revenue of $3.75 billion, up 44%. Logically, shouldn’t these stocks hit the daily limit up? But what happened? Sandisk dropped 7% after hours, and Western Digital was even worse, down 11%.
This is called “expectation killing.” The funds are very sharp now; earnings reports are yesterday’s story, who cares? Everyone is focused on next month’s guidance. Sandisk’s median revenue guidance for next quarter is $10.55 billion, sounds like a lot, right? But Wall Street’s expectation is $11.1 billion. Because of this less than 6% gap in expectations, hundreds of billions in market value just vanished. Citibank also acted swiftly, cutting the target price from 2500 to 2100. This market is like this: if you score 99 but don’t get 100, it’s considered a fail.
This drop dragged the whole sector down the drain. It’s even scarier in South Korea, with SK Hynix and Kioxia all down over 10%, Samsung also down more than 6%, dragging the entire market down. Why such panic? Mainly because three forces are pushing together.
First, everyone’s mentality collapsed. Originally, you thought you were a top student scoring full marks every time, but you scored 99. Even if you’re still first place, everyone will think you’re not good enough. Once this sentiment spreads, no one can escape.
Second, the logic has changed a bit. Previously, when we speculated on storage, the logic was “shortage = price increase = easy win.” Now? I heard Nvidia is considering downgrading the next-gen Rubin Ultra graphics card memory from the highest-grade 12-layer HBM4e to 8 layers. Why? Because it’s too expensive, production capacity can’t keep up, and yields are uncertain. This is interesting; before it was “you have money but can’t buy,” now it’s “you’re too expensive for me to buy” or “I have to cut specs to ship.” Once this shortage chokes the downstream, it’s not good news for the shovel sellers.
Third, profit-taking is heavy. This sector has risen for a year, with many stocks up several times. Valuations have long flown to the sky. Now with a little turbulence, those who made a killing are rushing to exit. If they don’t run, are they waiting for the New Year?
So some ask, is the AI storage rally completely over?
I don’t think it’s that pessimistic. Morgan Stanley and others are still saying this is a fake fall; data center shortages remain severe, and next year’s capacity is already sold out. This makes sense since the AI mega trend is there. But that doesn’t mean stock prices have to rise every day. The market is shifting from “listening to stories” to “verifying quality.”
We need to separate two things:
Short term (next few months): Don’t rush to catch a falling knife. Sandisk has set a rule this time: in future earnings seasons, unless guidance is “mind-blowing,” expect to get hammered. The key now is to be quick, exit when needed, don’t fight the trend.
Long term (looking at next year): As long as AI keeps developing, demand for storage remains. After this wave of sentiment vents and valuations deflate, those with real technology and who can handle HBM capacity will still be valuable.
Overall, this storage adjustment is a “killing the donkey after unloading the millstone” targeting high valuations. Don’t just focus on earnings anymore; that’s just the entry ticket, guidance is the key to whether you stay. For now, let’s just hold on and let the dust settle.
$SNDK $ETH $BTC After Google's four founding members left, they suppressed Loop: letting AI handle research automatically
---
The next stage of AI will truly be industrialized may not be "intelligence," but "trial and error."
In the past, scientists could only advance limited experiments per day; in the future, AI can simultaneously propose tens of thousands of hypotheses, run tens of thousands of parallel routes, automatically eliminate losers, and move on to the next round. Once this cycle is completed, scientific research will, for the first time, have scale effects similar to industrial assembly lines. This is the most noteworthy change behind the collective departure of Google's four founding members.
In August 2026, DeepMind's former CEO became chairman, Jeff Dean and three others left to found Discovery Loop, with Alphabet both an investor and a provider of computing power. This isn't a split; it's more like Google is actively launching top talent to bet on a more aggressive direction—letting AI "discover" on its own. The Loop in the company's name is the answer.
In the past, humans would ask questions and AI would answer; later, agents would execute goals. Loop Engineering goes further: humans only set goals, rules, evaluation criteria, and termination conditions, and the rest is automatically looped by AI—proposing hypotheses, writing experiments, executing, evaluating, eliminating, and generating the next round. The key is parallelism. Humans conduct several experiments a day, and AI can run thousands or tens of thousands simultaneously, sharply compressing trial and error time. Instead of continuing to build the upper limit of intelligence, it's better to let existing intelligence try and error ten thousand times a day, which is also an order of magnitude change.
This is not an isolated event. Currently, AI Scientist is divided into three layers: the most mature code model closed loop, clear feedback from Loss and benchmarks, and AlphaEvolve already in use; Theoretical closed loops are being experimented with, but the answer is no longer just numbers; The physical closed loop is the hardest, requiring automated labs and robotics to cooperate. The Discovery Loop approaches AI optimization precisely because it meets the premise of repeatable experiments and quantifiable results.
The stronger the cycle, the more critical the evaluation system. AI is good at optimizing the metrics you give it, but if the metrics are wrong, it will cleverly deviate. So the real barrier isn't the model, but who defines "what success is." When AI can run experiments infinitely, the rarest people are no longer those who find answers, but those who ask the right questions and design evaluation criteria.
In the past decade, computational costs have been reduced; in the past three years, generation costs have been reduced; the next step is to reduce discovery costs. In the future, computing power will no longer be measured by tokens per second, but by how many effective discoveries are generated per unit of computing power. Companies will be buying R&D speed. This path will ultimately encounter recursive self-improvement—AI design experiments to find better training methods, obtain stronger AI, and then move on to the next round.
So I don't see this resignation as a simple talent drain. When several people who have been at the forefront for over twenty years bet on the same direction, the signal is clear: after model capabilities, the next battle will be over who has the fastest discovery cycle. The day AI truly changes scientific research isn't the first time it gives astonishing answers, but it begins tirelessly trial and error.
#存储股财报后下挫, is the AI memory bull market still stable? $GOOGL
$SNDK $BTC
📌 实盘声明
当前仓位：BTC现货长持 + 指数基金定投。不碰合约、杠杆、期货。
本批次新闻不改变当前仓位计划——基建叙事在加速，但还没到动手的时候。
---
一、Base成为金融上链的默认入口
jessepollak放话"全球每一家金融机构都将把业务迁移到加密轨道上"，数据确实在支撑：2025年以来其他所有生态TVL下降，Base逆势涨35%；全球第二大借贷市场；BTC和ETH在Base上的现货交易量超过任何其他链；x402上90%的代理金融交易发生在Base。
Base正在吃下"传统金融进链上"这波红利。不是叙事，是资金流。
交易判断： ETH生态内的Base值得长期关注，短期不追高，等回调。
#Base #Layer2 #DeFi
---
二、Agent经济的支付层正在成型
本周多条消息指向同一个方向——AI代理正在获得独立的经济能力：
- Browser Use与Coinbase合作支持x402协议，代理可直接用USDC钱包购买积分，无需账户和API密钥
- Arkham API新增x402支持，代理可按需用USDC付费调用链上数据
- x402标准由Add a hard footnote to this storage supercycle: Nvidia is reportedly considering reducing the HBM memory capacity for the next-generation Rubin Ultra, not because they don't want to stack more, but because advanced high-bandwidth memory is simply unavailable. Even the computing power leader has to yield to memory, indicating the bottleneck has shifted from the GPU to the storage chips themselves. Data is more honest than narrative—HBM shortages will propagate up the entire DRAM supply chain, and the most price-elastic are often the upstream chip manufacturers. Storage stocks pulled back today, but the fundamental curve is still trending upward. Let supply and demand speak, not sentiment. $MU
#AIMemoryBullTest
#FedHawksVsWeakJobs
#SpaceXUnlockRebound 🔥 With the US suddenly engaging in nuclear blackmail, should the crypto world panic?
When I first saw this message, to be honest, my heart skipped a beat.
The U.S. Department of Defense is drafting a new nuclear strategy, with a single core statement: if regional conflict breaks out with China or Russia, the use of tactical nuclear weapons cannot be ruled out. And it's not just talk—it's officially included as a military option.
How serious is this matter? Let me break it down for you.
Tactical nuclear weapons aren't the kind that destroy cities like Little Boy's, but they're powerful enough to wipe a small city off the map. Previously, the U.S. took a "deterrence focus, not used lightly" attitude toward nuclear weapons, but now it has suddenly lowered the threshold for "use," targeting the two nuclear powers, China and Russia. This is essentially shattering the unspoken bottom line from the Cold War era.
For financial markets, such news has always been like a nuclear bomb.
Interestingly, however, the crypto community's reactions often differ from traditional markets.
When the Russia-Ukraine conflict first broke out in 2022, BTC dropped 8% in a single day, but rebounded a week later. During the 2023 Israel-Palestine conflict, gold surged, and BTC followed suit. In 2024, Iran's nuclear facility was attacked, causing market panic all night, then going back to business the next day.
Have you noticed the pattern? Geopolitical shocks are significant, but crypto memory's memory is shorter than goldfish. As long as there is no real fighting, after the market digests the panic, it often returns to its original trend.
But this time, what's different is that the U.S. is playing with the word "nuclear."
Nuclear deterrence and conventional warfare are completely different concepts. Once nuclear options are on the table, global capital will systematically decline in risk appetite. The US dollar may strengthen in the short term due to safe-haven demand, and gold is undoubtedly a beneficiary. But what about BTC?
My judgment is divided into two layers.
Layer 1: Short-term negative news.
BTC now has a correlation with Nasdaq exceeding 0.8, essentially making it a "tech stock Pro Max." If nuclear blackmail leads to a rise in global safe-haven sentiment, funds will flow into traditional safe-haven assets like gold, US Treasuries, and the Japanese yen instead of BTC. Don't forget, although BTC is called "digital gold," when it really comes to a critical moment, institutions are the first to sell it.
Layer 2: Medium- to long-term opportunities may be present.
If geopolitical tensions persist, global demand for "decentralized" and "censorship-resistant assets" will rise. A prime example is the surge in cryptocurrency usage after Russia was kicked out of SWIFT. If confrontation between major powers prolongs, the narrative of BTC as a "masterless currency" could be reactivated.
But the premise is—they can't really fight. If it really came to a fight, all assets would be wasted.
💡 Advice for the community brothers
1. Don't heavily invest in the direction at this position. August was already highly volatile, and now with added geopolitical risks, any heavy position is yourself.
2. Pay attention to the movements of gold and the US dollar index. This is a leading indicator for assessing the strength of risk aversion. If gold keeps hitting new highs while BTC remains unmoved, it means funds are not treating BTC as a safe-haven asset at all.
3. Keep stablecoins in hand. A real black swan has appeared—cash (USDT) is the real king. Only when the market panics and sells off will you have the ammunition to pick up bargains.
4. Don't be fooled by ghost stories like "nuclear war benefits BTC." When it really comes to that point, what matters to you is how to survive, not the price of the token.
Finally, a heartfelt thought.
The U.S. is essentially a strategic probe in its nuclear blackmail. It's highly unlikely to actually use it, but the very act of "daring to use it" has already changed the game. For the crypto community, this is not news to be ignored, but there is no need for excessive panic.
The market always rewards those who can think calmly when others panic.
👇 Do you think this nuclear blackmail will have a substantial impact on BTC? Or will the crypto world continue to "eat and drink when it should be"? Let's talk in the comments.There is an interesting point behind the postponement of the bill's vote to September: a proper delay may actually increase the chances of its passage.
Avoid the rushed vote during the August recess, giving both parties more time to reconcile differences and refine clauses, fully communicate and compromise contentious points, and reduce the risk of forced vote failure.
Of course, delay does not mean a safe pass. September also faces a complex parliamentary agenda, with many variables remaining, but the game dynamics have shifted. This bill is an important policy catalyst in the crypto community, and subsequent developments will continue to influence overall sentiment.
$BTC $ETH $BICO
#存储股财报后下挫, is the AI memory bull market still stable?
#联储鹰派信号升温, can weak employment outpace inflation?
#特斯拉SpaceX投建168亿美元AI芯片厂 Western Union connects stablecoins, Solana, and Visa payment scenarios, which is more practical than many public chain narratives. It's not just about issuing another stablecoin. What Western Union really wants to do is integrate on-chain dollars into its original remittance network: user receipts, offline cashing, merchant spending, and cross-border settlement, all trying to avoid being stuck by bank business days and correspondent bank chains. What stablecoins have always lacked is not on-chaiSpaceX is back in the spotlight. 🚀
More than $100B worth of $SPCX shares recently became eligible for trading, yet the stock still moved higher after receiving a bullish analyst price-target upgrade.
The lockup release could lead to increased volatility as early investors gain the option to sell. However, the bigger story is whether the market can continue absorbing the additional supply.
If demand remains strong despite the increased float, it would be a positive signal for investor confidence and the stock's longer-term outlook.
This isn't just about a lockup expiration—it's a real test of market conviction.
Definitely one to keep on your watchlist in the weeks ahead. 📈
#AIMemoryBullTest
#FedHawksVsWeakJobs
#SpaceXUnlockRebound