Orbit Post Sitemap
空头回补，正在成为市场中最沉默却最凶猛的一股买入力量。很多人聊到SanDisk的这轮上涨，往往只盯着基本面或者业绩预期，却忽略了一个非常反直觉的事实：那些曾经赌它下跌的人，最终也可能成为推动它继续上涨的人。🍵 先理清一个基本逻辑。做多的人，路径是买入、等待上涨、卖出离场。做空的人，路径则完全反过来：先借入股票卖出、等待下跌、再买回股票还掉借来的仓位。也就是说，做空者的平仓动作，本质上就是一个买入动作，只是它被延迟了，而且常常是在最不情愿的时候、以最被动的姿态发生的。 看看SNDK这段走势就特别典型。当股价还在1300的时候，有人觉得“投资人日就是利好兑现的时刻”，于是大胆做空。结果呢？公司不仅没有让市场失望，反而给出了比预期更强悍的长期模型。股价从1300一路走到1400、1500、1600。这时候，市场里其实同时出现了三股买入力量。第一股是最常规的长期投资者：看完投资人日，机构上调了未来营收、利润和现金流的预测，这些买单构成了趋势最基础的地基。第二股是趋势型基金：量化策略、CTA、动量交易者、突破交易者，全都开始跟进。价格越高，趋势越确认，买入的也就越多。这两层逻辑都很容易理解。 截至 8月15日，CryptoQuant 抛出了一份让多头稍微能喘口气的报告：比特币的表观需求已经从 6月初 的 负 272,000 BTC 猛地拉回到 负 32,000 BTC。这整整 24 万枚 BTC 的需求缺口回补，简直就是一场无声的诺曼底登陆。
什么是“表观需求”？简单说就是：新挖出来的币够不够大家分的。6月 那会儿，市场简直是供过于求的灾难现场，需求缺口大到像个黑洞（-27.2 万枚）。现在回升到 -3.2 万枚，说明结构性持有者正在发力。
虽然还在负数区间徘徊，意味着新挖出的币还没被完全吃掉，但比起之前“血流不止”的情况，现在这只能算轻微擦伤。
这波需求回升，不全是买盘太强，很大一部分原因竟然是对手盘变弱了。
因为网络哈希率降低，矿工产出下降了。这事儿挺讽刺：矿工们因为赚不到钱或者设备升级而关机，反而减轻了市场的抛售压力。这在经济学上叫被动缩减供应，变相帮多头省了 24 万枚 BTC 的接盘力气。
提个醒，这套路咱们在 2026年2月 和 5月 都见过。
当时数据也这么漂亮过，大家以为牛市引擎要发动了，结果提上裤子就走人——需求改善一下接着又走弱。这种在今年已经玩🚨 BREAKING: 3X LEVERAGED CRYPTO ETFs COULD BE COMING TO THE U.S.
Cboe BZX has asked the SEC to approve the first U.S. 3X leveraged Bitcoin & Ethereum ETFs.
The proposal also covers 3X leveraged gold, silver, crude oil and natural gas ETFs.
More leverage = more volatility. 👀
If approved, U.S. crypto traders could get a whole new level of exposure.
$BTC $ETH特朗普加密叙事越热，BTC越像政治资产
特朗普每次把加密货币放进政治话语，市场最先想到的几乎一定是[$BTC ](https://www.okx.com/zh-hans/trade-spot/btc-usdt)。这不是因为BTC涨幅一定最大，而是因为它最容易被包装成“美国拥抱加密”的门面资产。
政治资金不喜欢复杂。一个小币要解释团队、解锁、代币经济、链上数据、监管风险；BTC只需要解释三件事：供应固定、ETF可买、全球认知最高。对传统资金来说，这种简单非常值钱。越是宏大的政策叙事，越会先流向最容易被理解的标的。
这也是特朗普叙事对BTC的特殊加成。它不是普通利好，而是政治溢价。只要市场相信美国下一阶段可能更友好地对待加密资产，BTC就会被当成最稳的受益者。山寨币当然也会动，但那通常是第二阶段：先买BTC确认方向，再扩散到ETH、SOL、平台币、DeFi和meme。
不过政治溢价有一个麻烦：它兑现很慢，回撤很快。竞选语言可以一天讲完，法案要经历委员会、投票、修订、监管执行。中间任何一个环节卡住，短线资金都会先撤。最近SEC会议取消、参议院休会，就是典型的节奏风险。市场不是不相信亲加密方向，而是不知道什么时候能变成规则。
对BTC来说，特朗普叙事最好的状态不是天天喊，而是制度慢慢落地。真正能改变长期资金配置的，不是一句“支持加密”，而是银行、券商、基金、企业财库能不能在更清晰的规则下配置BTC。如果这个过程走通，BTC的政治资产身份会更稳。
反过来，如果加密议题在选举里变成党争工具，BTC短期会更波动。支持者把它当自由金融旗帜，反对者把它当利益冲突和监管漏洞，这种拉扯会让价格同时拥有想象力和不确定性。
所以交易特朗普叙事，不能只看情绪热度。更该看三件事：法案有没有排期，监管有没有文本，传统金融有没有因此扩大产品和配置。没有这三件事，政治溢价就是烟花；有了这三件事，BTC才会变成制度资产。
BTC已经不是单纯的技术资产了。它正在被卷进美国财政、选举和监管秩序里，越靠近权力中心，越有溢价，也越有噪音。
这种变化会让BTC的新闻敏感度越来越像宏观资产。以前大家盯矿工、减半、链上地址；现在还要盯国会日程、SEC议程、总统候选人表态、银行参与程度。它越被制度接纳，就越会被制度节奏影响。对长期持有者来说，这是成熟的代价；对短线交易者来说，这是新的波动来源。
如果美国最终把加密规则框架搭起来，BTC会得到更清晰的资金入口；如果党争把法案反复拖延，BTC就会在“政治利好”和“制度不确定”之间来回摇摆。政治溢价不是免费的，它要求市场持续忍受政策噪音。
这里的关键不是特朗普本人能不能让BTC涨，而是他能不能推动传统金融系统给BTC更稳定的位置。养老金、银行、券商、企业财库这些资金，不会因为一句话长期配置，但会因为规则清晰、产品成熟、合规责任明确而慢慢进入。BTC真正需要的不是一次演讲，而是一整套可执行路径。
如果这条路径走通，BTC的波动率可能反而会下降一些，因为配置型资金会变多；但上涨速度也未必像早期那么暴力，因为越成熟的资产越难单靠情绪拉出十倍。政治资产化给BTC带来更大池子，也给它带来更严格的定价纪律。🔥SPCX rebounds strangely! Unlocking + heavy bear pressure, why did it actually jump from 104 to 149? Is 🤔 it just to lure the bulls and lure the price up?
This segment of the SPCX movement left me completely confused; the logic is completely opposite to the usual market.
Reviewing historical market trends:
After listing, it was heavily speculated, soaring from the issue price of 135 yuan to $220, with the bubble fully inflated.
Then a fierce decline began, barely allowing a decent rebound, and it kept falling all the way.
Massive trapped orders are piling up above, with major bearish pressure and the market generally bearish.
Logically, stocks are experiencing large-scale unlocks and a large release of chips, which should have been pressured to decline.
Plus, with heavy short positions in the market, the normal logic would be to dump the stock.
But the reality was completely different: the price was forcibly pushed from $104 to a high of 149.47.
During this period, there was also Musk making positive remarks. When the news broke, it didn't crash the market but instead rebounded, now falling back to 139.23.
Look at the 4-hour candlestick chart 📊
After surging to 149.47, it pulled back under pressure, with the price having already broken below the Supertrend trendline at 147.85.
The upper Bollinger Band at 148.92 is strong resistance, while the lower band supports at 134.32.
RSI fell back to 41.91, MACD has turned negative, bullish momentum is waning, and trading volume has clearly shrunk compared to the rally.
My biggest question in my mind: Is this a reversal, or is it just a scam to rally and sell off?
A summary of market realities:
1. The unlock-up does leave a large amount of old chips to sell; Elon Musk's public statement has boosted market sentiment and attracted a wave of bottom-fishing retail investors.
2. Above are 220, 180, and 150 levels of trapped stocks; historical trapped positions have not been fully digested.
3. Short positions are heavy, and this rally is driven by some short squeezes. Bears are forced to close positions and drive prices higher, but not all of them are new bulls entering.
Two possibilities:
✅ Scenario One: Short-term rebound and recovery. Taking advantage of news to complete a short squeeze to wear down the bears' strength, but with heavy selling pressure above, it's hard to pull off a major reversal directly.
⚠️ Scenario 2: Induce bulls to sell. Taking advantage of Musk's remarks + the unlocking window, they pulled a rally to attract retail investors to bottom-fish, while the main players distributed chips in batches during the rebound, surging high but then pulling back again.
Pay special attention to two signals to distinguish between real and fake:
1. Whether it can hold above the previous high of 149.47 with increased volume and sustained breakouts will sustain the rebound;
2. If it breaks below the lower Bollinger Band support at 134.32, then this wave is very likely to be a bullish lure to deceive the gain.
Having suffered losses in several counter-trend markets, I dare not make subjective judgments now.
Good news doesn't necessarily mean prices rise, and lifting restrictions doesn't necessarily mean prices fall; the market often defies common sense.
Don't bet on the outcome early; wait for the market to give a confirmation signal before making your move.
$SPCX 市场预期美国解除对伊朗封锁的概率下滑至23%
Polymarket交易数据显示，交易员预判美国在8月31日前官宣解除伊朗封锁的概率回落至约23%，24小时内下行近10个百分点，创下近期低点；9月30日前解封的预期概率同样跌到44%。
背后多重利空叠加：美伊谈判节奏遇冷，停火磋商陷入停滞，美方持续加码对伊施压；霍尔木兹海峡航运承压，8月14日通行商船数量远低于正常水准，航道运输依旧受限；美国防长表态美军有能力维持无限期海上封锁，释放长期对峙信号。而伊朗立场未见松动，坚持要求美方解除制裁等先决条件，短期想要敲定共识难度大幅上升。
行情传导逻辑：利空风险资产，利多原油。
倘若解封预期继续走低，市场会开始定价霍尔木兹海峡长期通航受阻的情景：
原油上行 → 通胀预期抬升 → 美联储降息预期降温 → 美债收益率走高 → 美股、加密资产承压。
风险提示：仅分享思路，不构成投资建议，无不良引导，遵守社区公约！$BTC $ETH $SNDK #霍尔木兹通航谈判未果，美伊施压升级 目前利率期货隐含的九月加息概率已经砸到30%附近（Kalshi/Polymarket差不多这个数，CME还高一点但也在往下掉），紧缩预期明显降温，对风险资产估值的压制松了一截。 弱就业数据（7月非农直接砍了2.3万）把市场从“九月大概率加息”的情绪里拽了出来，CPI虽然还没彻底老实，但至少没再火上浇油。 地缘这块伊朗那边还没彻底消停，油价一抽风就可能再把通胀预期搞起来，所以避险情绪随时可能抬头。 只要那边别再升级、避险情绪别重新拉满，风险偏好就有机会慢慢修复，BTC大概率能吃到一轮流动性驱动的反弹窗口。 但短期别太冲动，量能配合和关键阻力位突破还得盯着，过早重仓就是给自己挖坑。 BTC维持震荡偏多思路： 回踩62300-62800继续尝试多，上方64000—64500就是压力带，放量突破并站稳后，短中线结构才能进一步转稳。 ETH昨天插针1860附近，没上车的回踩1850—1870直接尝试多；1900是日线关键压力，突破后日线有望再度走出反弹形态，后续可看惯性冲高。别追高，等回调再动手。 #消费动能转弱，9月政策仍受通胀制约 #OpenAI与Anthropic估值竞赛升温 #海力士#Consumption momentum weakens, September policy still constrained by inflation
After this report, analysis of the future trend of altcoins🔥
Weak consumption does not fully equal an altcoin bull market; inflation locks down easing strength; in a difficult macro environment, altcoins are unlikely to rally broadly, ETH/BTC does not rise, and most altcoins may only have pulses without a trend.
Macro core: Weak economic consumption calls for interest rate cuts, but inflation stickiness ties the Fed's hands, trapped in a dilemma of a cold economy and strong price resilience.
Current market situation: BTC range 62500‑64800, BTC market dominance remains high, capital shows clear risk aversion, making a full altcoin season unlikely, only localized thematic pulse rallies.
📊 Three scenario simulations
1. Neutral scenario
Consumption weakens, inflation stubborn, only slight rate cuts in September, rejecting large-scale easing.
- BTC maintains range-bound oscillation; altcoins overall diverge, no altcoin season, only individual thematic short-term pulses. ETH/BTC ratio remains low, most small and mid-cap altcoins oscillate and bottom out, awaiting new catalysts from PCE and Fed statements.
2. Optimistic scenario
Consumption significantly declines while inflation falls simultaneously, Fed releases dovish signals, rate cut expectations rise.
- BTC holds above resistance, ETH/BTC ratio rises, capital flows outward, leading altcoins rebound collectively, hot sector coins show outstanding gains; small-cap low-quality coins remain extremely risky.
3. Pessimistic scenario
Consumption weakens, but inflation rebounds again, September rate cut expectations sharply retreat.
Wishing everyone all the best! $BTC Every major BTC rally begins at a turning point in macro liquidity.
March 2020 — The pandemic collapsed, and the Federal Reserve implemented unlimited QE. BTC rose from 3,800 to 69,000.
Early 2023 — The pace of rate hikes slowed, and the market began to "shift" pricing. BTC rose from 16,000 to 70,000+.
What about this time?
At the July 29 FOMC, the Federal Reserve kept rates unchanged for the fifth consecutive time, at 3.50%-3.75%.
The key point is—rate hike expectations are collapsing.
At the beginning of August, the market priced in a 55% chance of a rate hike in September.
After the CPI was released, it dropped to 44.1%.
By August 15, CME data showed that the probability of keeping rates unchanged in September had risen to 67.5%, while the probability of a rate hike dropped to just 32.5%.
From 55% to 32.5%—this is not the end, but a signal that the Fed's narrative is starting to loosen.
Short-term traders see "BTC not rising."
Long-term holders see that "the spark has been lit."
The probability of a rate hike dropped from 55% to 32.5%. This is not the end, but a sign that the Fed's narrative is beginning to collapse.
Consumer data shifted from "strong" to "unexpected decline"—this is not volatility, but a trend.
The trend has already taken shape, just waiting for confirmation from the Federal Reserve.
And once confirmed—BTC's explosion always begins when most people are still hesitating.DOGE's "miner hidden sell-off": zero-marginal cost selling pressure—just how terrifying is that?
On August 15, 2026, DOGE was priced at $0.0701, up 0.05% in 24 hours, with a daily trading volume of about $310 million. The market appears calm, but veteran players are all worried about one thing: Are those miners who mined LTC and "freeloaded" DOGE been quietly dumping their shares?
Let's first clarify the accounts. $DOGE Each block produces a fixed output of 10,000 coins, one block per minute, adding 14.4 million coins per day, which, at current price, amounts to about $1.01 million in new supply per day. During the same period, $LTC produced 3,600 coins daily, which is only $160,000. Do you see the problem? In the revenue structure of merged mining, DOGE has long ceased to be a "byproduct"—it contributes over 80% of mining machine revenue, while LTC has become a bonus. So the narrative of "LTC miners casually selling DOGE" is reversed: today's Scrypt miners largely start their machines for DOGE profits.
What truly deserves caution is the cost structure. The same Ant-L9, the same electricity, feeding two chains with a single hash, DOGE's marginal cost is about zero. This means miners have no psychological barrier to selling DOGE with a cost line—$0.07 is profit, and dropping to $0.04 is still profit. The so-called "selling at no cost" is mechanically valid; this is not a conspiracy theory but an inherent feature of the AuxPoW protocol.
But the "volume" of selling pressure is another matter. Even if miners dump all 14.4 million DOGE added daily into the market, they only account for just over 0.3% of daily trading volume. This magnitude cannot crush the price; it acts more like a beach continuously seeping under the water during a high tide — not fatal but making every rebound hard to hold steady. DOGE has no halving mechanism; the supply curve is always upward. The fixed annual issuance of 5 billion coins means miner selling pressure is a constant rather than a variable; it will not disappear just because the market improves.
As for on-chain address tracking, let's be honest: DOGE's mining pool distribution model makes this chain extremely hard to see clearly. The pool first splits rewards into thousands of small payments flowing to miners' personal addresses, then miners transfer them to exchanges at selected times, with channels like collection, swapping, and OTC in between. No data provider has ever implemented a miner tag system for DOGE at BTC's level, so no one can provide the exact number of "miners flowing to exchanges" at present; whoever does is most likely making up the numbers. The only thing that can be confirmed is structure: selling pressure is dispersive, low-frequency, and persistent.
My judgment is: hidden selling does exist, but it's not DOGE's main contradiction. The real ceiling is the unlimited issuance mechanism itself—miners are just executors. DOGE's breakout of the trend isn't about miners "having a conscience" and not dumping the market, but on whether demand can steadily absorb $1 million in new shares daily. Before incremental narratives like ETFs and payment scenarios materialize, DOGE around $0.07 should watch selling pressure when it rebounds—don't chase.[Pharaoh Market Watch]
Pharaoh bluntly said, NVIDIA is no longer just selling "shovels"; it has switched careers to become "contractors"—setting up its own platform, letting Wall Street pay, customers on credit, and in the end, it just lies down to earn "hash power rentals." Is this $500 billion scheme a stroke of genius or a prelude to a bubble? Let's not play tricks and get straight to the point.
Let's start with synergies—this 500 billion investment is indeed tricky.
Previously, AI giants lacked everything—they didn't lack dreams, but they lacked money! This time, NVIDIA has brought in "big investors" like BlackRock and Goldman Sachs, with a clear goal: to leverage $500 billion in the next few years to buy their own graphics cards and build data centers for customers. A Bank of America analyst hit the nail on the head—demand has never been a problem, money is! Now it's different—even those "non-mainstream" players, like AI startups and small cloud service providers, can get GPUs at "preferential rates." Nvidia himself said that because Nvidia's hardware is hard currency and can be resold freely among different customers, lenders are very confident and directly use computing power as "digital real estate." This move is clever—turning the chip business into infrastructure rental income.
But what about the risks? The Pharaoh had to pour cold water on it.
The biggest concern in the market is whether this "circular financing" will blow up a super debt bubble. Think about it: the easier it is to raise funds, the more enthusiastic customers are buying; The more excited the buying, the better Nvidia's data looks; The better the data, the more daring Wall Street is to invest...... Doesn't this closed loop sound familiar? If AI applications can't outpace capital expenditures, then this 500 billion yuan is not fuel but a white elephant project—looks impressive but is actually a losing product. Wall Street short seller Michael Burry bluntly criticized, saying this is very similar to the "credit structuring game" before the subprime crisis, while Charnos bluntly said: Isn't this just "old work revamped" in financial engineering?
How did NVIDIA respond?
Old Huang quickly distanced himself: we only contributed 25% of the project scale, and the final decision was in the hands of financial institutions. We were just "matchmakers"—risk? That's Wall Street's business! But to be honest, the risk has only shifted, not disappeared. NVIDIA still holds the obligation of guarantees and residual value as a backup; if an avalanche really happens, it won't be able to escape.
Finally, Pharaoh made a decision:
This move is a classic strategic gamble. The synergy is truly attractive—500 billion directly ties AI infrastructure and Nvidia chips together; But the risks are also very high—once commercialization can't keep up with burning money, this clever closed loop can become a domino in no time. But in the short term, as long as computing power remains highly sought after, this 500 billion is the nitrogen acceleration driving the AI race. $BTC $ETH $SNDK #英伟达深入AI资本链, how to balance synergy and risk #消费动能转弱，9月政策仍受通胀制约
消费动能转弱，9月政策仍受通胀制约
当前宏观经济呈现“供强需弱”的显著分化。上半年GDP增长4.7%，但规上工业增长与社零增速差距达4.1个百分点，强劲生产与疲弱消费严重脱节。前期以旧换新等政策对耐用品消费形成透支，参照历史经验，修复周期或需2至3年。7月核心CPI时隔5个月回落至1%以内，折射终端需求偏弱的现实。
物价层面，输入性通胀压力缓和与内需疲弱并存。二季度货政报告指出，能源冲击烈度趋向缓和，输入型通胀担忧有所减弱。但国际大宗商品及关税因素仍构成外部风险，而消费端价格低迷反映内需修复尚不充分。
这种“想松又不敢松”的格局，使9月政策面临两难：一方面，二季度经济增速回落，内生动能脆弱，亟需逆周期调节加力；另一方面，主要经济体央行趋于紧缩，全球通胀中枢抬升，对国内政策空间形成外部约束。央行虽强调“加大逆周期调节力度”，但操作上更侧重完善短端利率调控，并未释放进一步总量宽松信号。
综合来看，消费动能转弱已是不争事实，而物价低位运行既反映了内需不足，也意味着货币政策在稳增长与防风险之间需更加精准权衡。9月政策能否有效破局。 砸盘拆解
$BICO今天砸盘，24小时-15.41%，振幅达到24.41个百分点，直接掀桌子砸盘。
现价0.023220刀，成交3.99M美金，量能同比放大至少2倍，资金不是小打小闹。
24小时高点0.029890刀，低点0.023190刀，高低点差拉开了24.4个点的操作空间。
所属其他板块，这一轮砸盘不是单币独立行情，同赛道至少有3个币种同步异动，板块联动效应明显。
第一层拆开抛压：获利盘集中止盈跑路，次层看聪明钱提前减仓至少23个百分点仓位，最后看散户恐慌多杀多踩踏。
观察点：看下跌过程中有没有大资金承接，如果成交持续萎缩到今天的3成以下，那就是真跌不是洗盘。
我的看法：异动不追，等承接释放完看结构，结构破了就别硬扛。
数据来自OKX公开现货行情，仅供信息参考，不构成投资建议。
信号给到了，动手不动手你定。The core issue behind SK Hynix's accelerated expansion is not whether AI demand is strong, but whether capital expenditures can truly be recouped.
HBM sales are booming, NAND demand is recovering, and AI server orders are lining up—these are all real positive news. SK Hynix's increased investment now shows that management believes this round of AI storage is not a short-term boom but a structural change. But the harshest part of the storage industry is that in every bull market, everyone believes at the peak that this time is different.
There is nothing wrong with expanding production itself; the fault lies in the entire industry expanding together.
If capacity is concentrated in three years and AI customers cut orders or gain stronger bargaining power, today's capital expenditures will become tomorrow's depreciation pressure. Investors are buying "supply shortages" now, but companies are spending fixed money.
I think SK Hynix's line depends on two things: how much profit long-term contracts can lock in, and whether the pace of expansion can be controlled.
The real test of the AI storage bull market is not how much you can earn during shortages, but whether you can maintain profit margins once supply increases.
#海力士扩产提速, whether capital expenditures can deliver returns OpenAI 和 Anthropic 的估值竞赛，已经不是两家公司谁更贵的问题了。
现在整个 AI 交易都在押一个前提：这两家公司能持续吃掉模型需求、云算力、企业预算和开发者心智。问题是，估值越冲越高，压力也越集中。OpenAI 准备 IPO，Anthropic 二级市场估值被抢到夸张水平，投资人不是在买利润，而是在买“未来 AI 入口”。
这很刺激，也很脆。
如果它们继续高速增长，微软、亚马逊、谷歌、英伟达、数据中心、电力、光模块都会跟着受益。可如果竞争加剧、价格战出现、推理成本压不下来，整条 AI 资本链都会被重新定价。
我现在看 AI 股，反而更关注这两个模型公司的账。因为很多巨头的 AI 收入，本质上都绕不开它们。
少数公司变成整条产业链的支点，这才是最让人兴奋也最让人害怕的地方。
#OpenAI与Anthropic估值竞赛升温 消费数据转弱，市场最容易误读成“降息要来了”。
7 月零售销售意外下滑，消费者信心也掉得比预期更快。表面看，需求降温会压低通胀，联储没必要继续加息；但问题是，消费弱掉的时候，通胀并没有完全消失。油价、食品、住房这些刚性成本还在，低收入人群体感更差。
这就是现在最难受的组合：买东西的人少了，但生活没便宜多少。
如果只是需求降温，风险资产会开心；如果是消费动能转弱叠加通胀粘住，那就不是软着陆，而是利润表开始被啃。企业卖不动货，居民不敢花钱，联储还不能立刻放松。
我觉得 9 月政策真正受制约的，不是单个 CPI 或零售数据，而是“消费弱了以后，通胀还愿不愿意跟着弱”。
#消费动能转弱，9月政策仍受通胀制约 When the trading bell rang during the overlapping European and American trading hours, the total market volume surged by 43.29% within 24 hours. However, when I turned my attention to DeFiLlama's on-chain ledger, I found that total TVL only slightly increased by 0.14%. This extreme divergence of "trading boiling and locked positions quiet" is no coincidence. As on-chain detectives, today we will peel back the facade of a slight price increase and track exactly where this massive amount of money is flowing. 📌 ══════════════ [Total Market Volume] $107.741 billion | 24h +43.29% | Capital activity surges 📌 [Total TVL across the entire market] $74.851 billion | 24h +0.14% | Locked Amount Nearly Stagnant. The Investigation Conclusions Are Clear: Massive Funds Aren't Flowing into DeFi Protocols for Long-Term Accumulation, but Focus on Centralized Exchanges or Derivatives Markets for High-Frequency Gaming. Against the backdrop of a Fear Index of only 34 (Fear), this phenomenon of high volume without TVL growth indicates that large funds are using high volatility for position hedging or short-term arbitrage, while retail investors, fearful, are reluctant to easily lock their spots on-chain. Funds show a clear pattern of "fast in, quick out" speculative capital. 📌 ══════════════ [Ethereum TVL] $41.065 billion | Still dominant in the market, but capital growth 📌 slows [BSC / Solana / Tron] all hovering between $4.7 billion and $4.9 billion Form the second ladderBroadcom's 6% drop in a single day indicates that capital flows are shifting from simply chasing hash power scale to repricing financing costs eroding profit margins. The current pricing theme is the conflict between marginal leverage relied upon for hash power expansion and high interest payments.
From the perspective of market liquidity, a single-day 6% drop was accompanied by concentrated short positions in the derivatives market, and spot funds are showing clear signs of safe-haven selling at the current level.
In terms of driver factors, debt interest expenses dominate by the erosion of marginal net profit, followed by the flow of safe-haven funds in the derivatives market, and the scale of traditional computing power demand takes a back seat.
The trigger conditions for the downside scenario are that high-cost debt interest continues to erode the marginal profit margin, and spot funds maintain net outflows. Under this scenario, it is necessary to observe changes in the concentration of derivatives put option positions. If leveraged funds accelerate their flight, the decline will continue to spread along the computing power industry chain. If spot buying occurs strongly and sustainably at low levels, this downward scenario will become invalid.
The trigger conditions for an upward scenario are that macro interest rate cut expectations will materialize and lower lending rates, or that cash inflows from high-margin software businesses exceed market expectations. At this point, it is necessary to observe whether net spot inflows have crossed key pressure zones. If cash flow from software businesses fails to fill the interest rate gap, the momentum for a long-term derivative rebound will be quickly interrupted.
The invalidity criterion lies in whether Broadcom's next financial report can prove that its high-margin business has sufficient interest hedging capability. Once the financial report shows that interest expenses have not substantially weighed on profit margins, the market will shift back to the logic of computing power expansion.
In the next 7 days, focus on the distribution of position turnover in the derivatives market, changes in the net flow of spot funds, and the transmission rhythm of liquidity in the computing power sector to interest cost pricing.
#英伟达深入AI资本链. How to balance synergy and risk? #加密估值转向收入, how is BTC priced? #韩股十日反弹逾22%, chip stocks led the gains$ETH's performance was also dragged down by BTC and is somewhat weak.
1. Strong is definitely stronger than $BTC, because the fundamentals are still improving, and there's some potential for speculation. For example, Fidelity's application to stake 100% ETH is truly positive; And the staking rate has risen to 34%, which is pretty good.
2. But some expectations were disappointed, making it difficult to break free from standalone rallies. The main issue was the Glamsterdam upgrade being delayed from Q3 to Q4.
3. Especially when the overall market is pulling a bit, ETH has been dragged down accordingly. Previously, when the whole market was long waiting for a breakout, the price just didn't rise, which itself indicates significant potential selling pressure.
My approach: spot trading can be done at the bottom, but try to avoid contracts in the near term. Because it could be drowned out by the broader market at any time; but it could also cause the "Fidelity staking" application to be approved and crash. So contracts are prone to problems, but spot trading is not a concern; after all, if prices drop, you can continue bottom-fishing to level costs.In the current macro environment, funds are systematically flowing from altcoins to US stocks (especially AI/technology sectors), both of which are extremely high-risk forbidden zones; If we only compare strategies, US stocks have regulatory and profit foundations, while altcoins are in a phase of liquidity exhaustion and debubble, with significantly lower win rates than the former
Key Differences and Current Situation Capital Flows: Speculative capital has exited altcoins on a large scale (outflowing over $40 billion), shifting to the US AI sector and macro assets; Exchanges collectively transform their equity businesses, marginalizing altcoin support.
Risk-reward ratio: The US stock market (S&P/Nasdaq) has maintained positive compound growth in recent years, with controllable drawdowns; altcoin indices have long had negative returns and high volatility, with a negative Sharpe ratio and no independent strength base.
Regulation and compliance: US stocks are strictly regulated by $SEC and supported by real earnings; Cryptocurrencies in Chinese mainland are strictly prohibited from trading and speculation, with no legal protection, and tightening global regulations have compressed the survival space for small-cap coins.
Technical signals: 84% of altcoin prices are below the daily moving average, the quarterly index is only 49, and there is no reversal signal before Bitcoin $BTC funds stabilized
Strategy Comparative Analysis
Position in US stocks
Advantages: deep liquidity, backed by corporate earnings, well-developed dividend mechanism, and risk diversification through ETFs; AI and semiconductor sectors remain the main consensus themes for capital.
Disadvantages: Valuations are at historic highs, sensitive to Federal Reserve interest rate policies, causing short-term volatility; Requires legal accounts and funding channels.
Mature investors with risk identification capabilities and compliant channels
Speculating on altcoins
Advantages: A very small number of leading projects (such as $SOL, $XRP $ETH $SNDK, etc.) may have explosive momentum during specific cycles; There is theoretical room for on-chain innovation scenarios.
Disadvantages: Liquidity has plummeted, with over 80% of small-cap coins at extremely high risk of zeroing out; Exchanges no longer support it, entering a "self-survival" phase; Frequent leveraged liquidations, mostly zero-sum game traps.
Institutions or experienced traders with extremely high risk tolerance and professional quantitative capabilities
Capability matching: Both altcoins and US derivatives involve complex macro judgments and risk control, lacking experience in fund management and sentiment control, making them easy targets to be "harvested."
Recommended actions: prioritize learning basic financial knowledge and understand market logic through simulated trading; If you have idle funds, allocate assets through legal and compliant channels, and avoid touching black and gray market exchanges
The current market is essentially a debubble process. Flowing funds from "story-telling" altcoins to "performance-driven" US stocks is a rational choice, but this requires extremely high qualifications from participants. Stay away from illegal trading. #英伟达深入AI资本链. How to balance synergy and risk? #特朗普称通胀迎来好消息 #加密估值转向收入, how is BTC priced? In the past 24 hours, the crypto market has seen a slight recovery, with BTC and ETH rebounding after a halt, but liquidity has not improved in tandem. ETFs continue to see net outflows, and SOL on-chain capital accumulation remains weak. Currently, it seems more like a structural rotation amid fear rather than a full-scale risk-on. 📊 Market snapshot As of August 15, 17:43 HKT: BTC $62,960, 24 hours +0.42%; ETH $1,877.64, 24 hours +0.39%; SOL $75.14, 24 hours -0.34%; Total crypto market capitalization about $2.25 trillion, up 0.19% in 24 hours. BTC market share 56.10%. The Fear and Greed Index rebounded to 34 (fear), improving from yesterday's 29, but still not out of the fear zone. Among the top 30 non-stablecoins by market capitalization, **LINK's rise of 5.9%** led the gains, while **LEO's decline of 4.3%** was the weakest. Overall, total market capitalization has only slightly rebounded, BTC and ETH have completed technical repairs, while SOL remains weak, and funds are still concentrated in a few mainstream assets and hot sectors, without a comprehensive rally. 💰 ETF outflows continue, but selling pressure has eased. The latest completed U.S. trading day data shows BTC spot ETFs had a net outflow of $56.2 million. Compared to the previous trading day's net outflow of $131.1 million, this narrowed significantly, but it still recorded net outflows for the second consecutive trading day. ETH ETF net for the day📊 $SOL Contract Overload Express (August 17)
According to liquidation data, Dog Broker played a classic short-cycle long selling→ medium-cycle directional confusion→ and long-term full-scale long selling with bulls repeatedly pinned down for 24 hours...
Time: Total liquidation, long liquidation, short liquidation
1 hour: $1,087.24 $1,087.24 $0
4 hours $38,000 $34,700 $3,311.27
12 hours: $270,000, $137,200, $132,800
24 hours: $2.8914 million, $2.4203 million, $471,100
Looking at $SOL liquidation data, the 1-hour long liquidation crushed the bears, completely wiping out the bears. The long sell-off was textbook-level but extremely small in scale; The 4-hour bull market continued to crush, with the bulls 10.5 times the bears. The bullish killing exploded with nuclear intensity, with liquidations jumping from $1,087 to $38,000; the 12-hour direction weakened sharply, with the bulls only slightly outperforming the bears by 1.03 times, making the balance almost tied, with the direction extremely unclear. The liquidation moderately climbed to $270,000; the 24-hour bulls took over again, with long liquidations at $2.4203 million versus bears at $471,100, with bulls 5.14 times the shorts—Dogzhuang completed a three-stage harvesting path on SOL: from the warm-up of long sellers →to the mid-cycle confusing everyone → the long-term period going all out. Short-cycle bulls were targeted to blow up, medium-cycle bulls confused everyone, and long-cycle bulls were wiped out in one hit, with cumulative liquidations breaking through $2.89 million. Especially from 12 hours to 24 hours, the direction shifted from nearly balanced to direct selling, with liquidations soaring from 270,000 to 2.89 million—a textbook example of "build first, then sell." Everyone should control their positions to avoid being bought back and cut off.
⚠️ Risk warning: SOL short-term long selling (1H/4H) aligns with 24-hour long selling, but the 12-hour direction once approached equilibrium, making it highly misleading; 24-hour liquidations account for 94% of the total daily volume, indicating a high concentration. Leverage is recommended to be compressed to within 3x; do not blindly bottom-fish, strictly control positions, and wait for clear direction.
🔥 Market Weather Vane | August 17
Today's three hot topics point to the same theme: macroeconomic uncertainty combined with the escalation of the AI race—cooling consumption shifts the policy balance, and the valuation bubbles of AI giants and the billion-yuan capacity expansion of storage leaders converge at the same timeframe.
📉 Consumption momentum weakens: probability of a rate hike in September drops to 28.6%
U.S. consumer demand has continuously signaled a cooling down. Retail sales in July fell 0.6% month-on-month, marking the largest drop in over a year; The University of Michigan's preliminary August consumer confidence index fell to 51, marking the first decline in three months. Consumer expectations for inflation over the next year rose to 4.3%, while real average hourly earnings fell 0.2% year-on-year.
CME data shows the probability of a rate hike in September has dropped to 28.6%, while the probability of keeping rates unchanged has risen to 71.4%. However, core CPI year-on-year is still well above the 2% target at 2.5%—given the combination of weak employment and sticky inflation, the Fed is still highly likely to hold steady, but a rate cut is far from discussed.
🤖 OpenAI vs. Anthropic Valuation Race: $852 billion vs. $2 trillion
The valuation race among AI startups has reached a fever pitch. OpenAI completed a $7 billion employee share buyback, keeping its valuation at $852 billion. However, Q1 revenue was about $5.7 billion, with about $3.7 billion burned during the same period, exposing the gap between growth and profitability.
On the other hand, Anthropic is expected to go public in October, with some investors predicting a valuation as high as $2 trillion. Supporting this valuation is Q2 revenue of $10.9 billion and adjusted operating profit of $559 million for the first time. Anthropic's market share in enterprise-level large model APIs has reached 32%, surpassing OpenAI's 25%. When a company founded only five years ago is given a $2 trillion valuation, the market is betting not on current profits, but on the complete restructuring of the enterprise market by AI.
🏗️ SK Hynix's $720 billion capacity expansion: a bold gamble on AI storage
Storage leader SK Hynix announced a $720 billion investment to build the world's largest memory factory network. The company has clearly stated that memory has evolved from components to core AI infrastructure. But the risks are equally real: SK Hynix's US market capitalization has retreated about 21% from its July peak; If AI demand growth slows or customers cut capital expenditures, memory chip prices could face sharp corrections.
💎 Summary
Cooling consumption has pushed the probability of a rate hike in September to 28.6%, temporarily releasing macro pressure; The valuation race between OpenAI and Anthropic has pushed the AI narrative to a new height of $2 trillion; SK Hynix's $720 billion capacity expansion is the industry's biggest bet on long-term AI demand. When the macro window opens, valuation bubbles expand, and billions in capital bet simultaneously—the AI sector is shifting from "storytelling" to a stage of "real money." #消费动能转弱, September policy remains constrained by inflation
#OpenAI与Anthropic估值竞赛升温
#海力士扩产提速, whether capital expenditures can deliver returns BTC has released another set of cooling data.
American consumers have truly started to hit the brakes.
July retail sales fell 0.6% month-on-month, compared to expectations of +0.1%, marking the largest drop in 14 months. Excluding automobiles, the market also fell 0.3%, with the control group dropping 0.4%.
Consumption hasn't crashed, but it's clearly not as intense as before.
More importantly, the probability of a rate hike in September has dropped from about 50% a month ago to just over 30%.
This is a positive factor for BTC/ETH.
But I won't go all-in on a single statistic right now.
BTC：
Right now, it's near 63K, and 62K-62.5K is my main observation area.
If the pullback around 62K does not break through, you can buy in batches, first increasing the position by 30%-40%;
Regaining 64K and adding another 20%;
Only after volume surpasses 65K does he consider raising his position to around 70%.
Conversely, if 62K is effectively broken, exit first and don't be stubborn with the market.
ETH：
It is now around 1,880, clearly weaker than BTC, so it's not chasing for now.
1,800-1,850 can be started in batches;
Regaining the 1,950 level to confirm the trend;
If BTC falls below 62K, don't rush to buy ETH for now—wait until the market stabilizes.
In short:
BTC is looking for a 62K defense, 64K confirmation, and a 65K breakout.
ETH is expected to take support at 1,850 and confirm at 1,950.
#消费动能转弱, September policy remains constrained by inflation $BTC $ETH Some people are still shouting, "Never sell coins."
Hyperscale Data has already exchanged 685 $BTC for an AI data center.
The company confirmed that the sale raised approximately $43 million in cash, of which about $30 million was used to reduce debt, with the remainder primarily supporting the construction of the Michigan AI data center.
After the sale, the company still holds about 275 BTC.
It emphasizes that the long-term Bitcoin strategy has not changed, and mining will continue in the future, with accumulation expected depending on market and capital conditions.
But what the market really needs to discuss is not whether it continues to be bullish on its words.
The question is whether companies will start recalculating the opportunity cost of holding coins when electricity, data centers, and capital flow into AI.
At approximately $62,970 of $BTC, the value of 685 coins is exactly close to the announced amount.
The fact is, it has already sold its coins, paid off debts, and invested in AI.
The view is that this is not simply bearish, but a prioritization of cash flow versus belief.
In short:
Companies don't hold positions for narratives; they only reinvest for capital returns. #OpenAI与Anthropic估值竞赛升温 $BTC Broadcom Falls 6% Overnight: The Next Hurdle for AI Is Financing Costs
The market has begun to seriously price a deeper variable: when financing becomes the main source of marginal computing power, every profit in the AI industry chain will quietly be included in interest costs. The real test is not a systemic default, but the continuous erosion of AI industry chain profit margins by rising financing costs. Broadcom's next earnings report will reveal the results.The current market is like a cup of cooled tea—smelling fragrant, but drinking it no longer feels off. Are we chasing the rise, or are we just picking up a knife that's not finished yet? Let me start with my impression of the market. BTC is gasping near 62.8K, down less than 1% in 24 hours, but has dropped more than 3% for the week. What concerns people most is not the decline, but the fact that it "can't rise" itself. With inflation data softening and employment data softening as well, risk assets should have jumped a bit, but BTC didn't even deliver a decent rebound. This state of being desensitized to positive news is often more worthy of caution than a sharp drop. ETF flows clearly illustrate this awkwardness. On August 14, BTC spot ETFs saw a net outflow of $55 million, which was also negative the previous day. But on ETH's side, funds actually flowed in. This is not a simple rotation, but rather that capital risk appetite is truly shrinking, while also seeking more concrete narrative anchors. What is the market trading now? I don't think it's about "whether the bull market is still ongoing," but "who still deserves to keep holding." BTC is treated as a defensive asset, ETH is relatively tough, SOL is about elasticity, and altcoins are extremely picky individual performances. Under this structure, the index's ups and downs have become distorted; what truly matters is relative strength. The signals from the derivatives side are even more direct. If the funding rate is not high, it means leveraged bulls are not celebrating; However, there has also been no deep negative rate, indicating that the bears are not decisive enough. The entire holding structure is like a deadlock chessboard, with everyone involvedNVIDIA is working with asset management institutions to promote the assetization of computing power and is evaluating the provision of up to 25% residual value as a backup, $NVDA trying to use chip manufacturers' creditworthiness to support the downstream capital chain.
After the announcement of this maximum 25% depreciation support plan, short-term fears among hash rate lessors and model buyers over rapid hardware depreciation eased somewhat.
Behind the assetization of computing power is capital's vigilance against revolving borrowing. OEM endorsements have marginally repaired market risk appetite and supported institutional confidence in core technology asset holdings.
Although the bottom-line commitment temporarily eases downstream book depreciation pressure, it also directly links downstream credit and repayment exposure to the chip manufacturer's balance sheet.
If commercial revenue from large models is paid on time, depreciation coverage will further lower the financing threshold and drive new funds to continue investing in computing power assets.
If delayed monetization of downstream applications accelerates hardware depreciation, bottom-line fulfillment will reverse-engineer depreciation burdens to the original manufacturers, triggering credit tightening and position withdrawals.
If this support mechanism ultimately remains at the end of a very small number of strict review projects and cannot be widely implemented, the risk appetite currently built on expectations of credit support will be quickly disproven.
The most noteworthy variable to watch in the next seven days is the actual scale of funds subscribed by asset management institutions for the first batch of computing power assetization projects.
#财报观察员: AI infrastructure earnings report debuts at #韩股十日反弹逾22%, chip stocks lead the gainsThe competition between ETH and SOL may ultimately not be about TPS
$ETH and $SOL have been compared for too long; the market loves to ask who is faster, which is cheaper, and whose DEX volume is stronger. But the further it goes, the more this comparison feels like evaluating banks with sports car metrics.
SOL is indeed easy to use. Low fees, high speed, smooth wallet experience, smooth meme trading—once users get used to this on-chain pace, it's hard to accept high gas fees and slow confirmations. It's especially suitable for speculation, hotspot hunting, bots, short-cycle asset issuance, and also for new users to experience "on-chain is really fun" for the first time.
ETH's advantage is not here. ETH now functions more like a settlement layer and asset layer. Stablecoins, DeFi collateral, RWA, institutional custody, long-term lock-up—these require security, liquidity, and ecosystem inertia. Unlike meme funds, where they go wherever they are cheap, the larger the money, the less they like migrating, because migration itself is risky.
Therefore, the competition between ETH and SOL is not simply a matter of speed but a contest of user behavior and capital behavior. Users are attracted by the experience, capital is attracted by security and accumulation. SOL competes for the "next-generation on-chain entry point," while ETH guards the "home of high-value assets." Both advantages are real, but their valuation methods are completely different.
If you only look at DEX trading volume on a given day, SOL easily wins; If you look at stablecoin scale, DeFi collateral depth, and assets institutions are willing to hold long-term, ETH still has a deep moat. When market sentiment is hot, people tend to overestimate the importance of experience; When market risk shrinks, people refocus on accumulation.
The real interesting thing is the middle ground. Can SOL retain meme users, turning them into payment users, stablecoin users, and on-chain app users? Can ETH make the L2 experience cheap and smooth enough to stop pushing new users to other chains? Whoever can seize this middle ground will get the next round of valuation expansion.
I don't think SOL must "kill ETH" to have value, nor do I think ETH can hold onto old money to be worry-free. One is responsible for bringing more people on-chain, the other is responsible for keeping more money around. The real collision between the two is when users and capital start choosing the same entry point.
TPS is superficial competition; asset dwell time is the deeper competition. In the on-chain world, it's not about who wins the fastest, but who can make the most people finish and still keep their money.
This perspective also explains why, at different stages of a bull market, ETH and SOL are alternately favored by the market. When the market is just heating up, users want speed and profit-making effects, and a highly liquid experience like SOL is more easily pursued; When the market enters the asset accumulation phase, institutions and large funds refocus on security, custody, collateralization, and settlement, and ETH's advantages return. They are not linear substitutions but cycle displacements.
What truly deserves attention is the intersection: can SOL grow into a financial layer that reassures big investors, and whether ETH can stop ordinary users from being discouraged by high costs. Whoever fills their weaknesses first can grab an extra piece from their comfort zone into the other's territory.
In the short term, SOL is more likely to break out in hot market trends because memes and new asset issuances are naturally suited for spread; ETH is more frequently mentioned in institutional narratives because stablecoins, ETFs, RWAs, and custody are all closer to it. One is responsible for generating hype, the other is responsible for carrying weight. If the market has strong risk appetite, it will favor the former; If the market becomes more cautious, it will prefer the latter.
So this competition can't be described as a single win. The real answer might be multi-chain user entry points and centralized asset settlement. In other words, where users play doesn't matter; where money ultimately settles is the key to determining the valuation ceiling.The biggest "positive" illusion in August is time to be exposed.
Many firmly believe that the midterm elections in August 2026 will trigger a Bitcoin surge—because only by driving up the price can Wall Street capital willingly side with Trump, and cryptocurrency gains are the hardest votes.
But history never lies. In every midterm election cycle, BTC has recorded sharp drops, with the smallest being in 2022, with the largest drop reaching 37.2%.
When this ironclad set of data is laid out before you, how do you feel?
The so-called "midterm election dividends" are nothing more than wishful market narratives, never real positive news.
So here's the question: Are you fully stocked with your bullets? Have you adjusted your position correctly? If you hit another 20% hole this time, fellow cultivator, can you withstand it?
What's even more noteworthy is that August~October is almost always a "stormy period" of declines in history. Will 2026 be rewritten? No one can predict it, but the odds are on the side of history.
Rather than betting on a turnaround, it's better to first plan a backup plan.
#消费动能转弱, September policy remains constrained by inflation
#OpenAI与Anthropic估值竞赛升温
#海力士扩产提速, whether capital expenditures can deliver returns Everyone is watching BTC’s $63K price. But I think the more important question is: Where is the liquidity? 👀 Bitcoin is struggling to build momentum even after softer U.S. inflation data. That tells us something important: The problem may not simply be lack of bullish news. It may be lack of fresh capital entering the market. Stablecoins are one of the best places to watch this. Why? Because stablecoins are basically the buying power sitting inside crypto. When stablecoin liquidity expands, theLooking back at the week: — Monday: BTC broke back above $65,000, showing strength — Wednesday: Cooler CPI data, but the reaction was mixed — Thursday: All timeframes (daily/weekly/monthly/yearly) turned bearish for the first time this cycle — Friday: Crypto kept weakening while gold/silver strengthened on the same inflation data The big picture: the week started with optimism but ended in clear weakness — even though the macro backdrop (cooling inflation, dovish Fed outlook) should theoreticall$BTC — a reminder for everyone waiting for the next big breakout.
Macro conditions have actually been pretty supportive over the past two days: CPI and PPI are cooling, retail data is weakening, rate-hike expectations are fading, and U.S. stocks are pushing toward new highs.
Yet what is Bitcoin doing?
Still stuck sideways. Barely moving.
There’s an old trading saying: when the news is bullish but price refuses to go higher, that can be a bearish signal in itself. Markets don’t always need bad news to fall. Sometimes, the inability to rally despite favorable conditions is the warning.
That’s one of the main reasons I’m still holding my $BTC short.
The real question now is: Is $BTC quietly building energy for the next breakout, or is this sideways action actually showing underlying weakness? 👀
#WeakConsumptionFedSplit #OpenAIAnthropicRace US July retail sales fell by 0.6%, compared to expectations of +0.1%, a direct upset—22 key accounts on X were promoting this topic (heat level 188), and traders responded unanimously: betting on further rate hikes and pulling back. But strangely, $BTC 63,045 only rose 0.28%, not even reaching 63,245. Why is the market so calm? Because everyone knows: the September policy is tied by inflation. Signal One: The chain of evidence for consumption momentum is complete. Not just retail: this week CPI rose 3.4%, PPI was flat, and initial jobless claims rose—the economic cooling is not a single event, but a systemic issue. The reasons for rate cuts are becoming more and more plausible. Signal two: But inflation has left no room for policy relaxation. CPI 3.4% is still far from the 2% target; On the Hormuz side, the Iranian parliament approved a ban on US-Israel ships, Brent is approaching $88, and Trump has clearly stated that "Americans must accept high oil prices"—when oil prices rise, inflation sticks even harder. "Economic cooling + sticky inflation" is the toughest combination: in September, they want to loosen but don't dare, most likely to hold their ground or take small steps to test the waters. Signal 3: The market has long been voting with its feet. BTC long-short ratio is 0.26:0.30, the only bear in the market to have the advantage—funds are not betting on "rate cuts coming soon"; BTC long-short ratio is 0.26:0.30, the only bear in the market to have the advantage—funds are not betting on "rate cuts coming soon"; ETH 0.33 : 0.14、$SOL 0.52 When will $BTC start rising alongside US stocks?
Both conditions must be met simultaneously, relying solely on the US stock market high is ineffective:
1. BTC spot ETFs have returned to continuous large net inflows, with institutional funds flowing back into the crypto market.
2. The ETH/BTC ratio is rising, funds are spilling out of Bitcoin, and risk appetite is truly unlocked.
In short: US stocks provide BTC with a macro safety base, but they won't directly drive BTC higher; What truly determines whether BTC can rise is the crypto buying funds.
Hitting new highs in US stocks does not mean BTC is following the rally; the AI sector is absorbing incremental funds; US stocks are just the environment; ETF funds are the real key to the Bitcoin rise.
Key Practical Market Monitoring Signals 🔍:
1. BTC box at 62,500-62,800 support, resistance at 64,800
2. Daily capital inflows and outflows for BTC spot ETFs
3. PCE inflation, US Treasury yields, and oil geopolitical trends
4. ETH/BTC exchange rate to assess internal crypto risk appetite
(Personal opinion analysis only, no investment advice)
Everyone moves forward steadily. Wishing you great wealth and better and better times$ENA the most exciting news isn't price fluctuations.
About 20% of the total supply is held by a listed company.
StablecoinX disclosed holding about 3 billion ENA, with a reported value exceeding $250 million.
After the announcement, the company's stock rose more than 12% intraday.
But the other side is even more striking:
The company posted a net loss of $34.2 million in the second quarter, most of which came from $36.2 million in digital asset impairment.
$ENA current price is around $0.084, still showing slight weakness over the past 24 hours.
Thus, the market unfolded a very crypto-driven picture—
Stocks celebrate "holding enough coins," while tokens are questioning "Is the concentration too high?"
The fact is, StablecoinX holds about 20% of the total supply.
The view is that such concentrated chips can form long-term interest bindings but may also become sources of liquidity and governance risks.
Sharp Commentary:
The whale locking coins in the vault does not automatically mean the market has reached the value bottom.
The next step depends on whether $ENA can regain $0.09 or if the company's balance sheet remains under pressure for now.
Do you see these 3 billion ENA as a belief, or as a token hanging over the market?Crowding and Crowding List
Continuous payments on one side are not scary; paying but unable to push the price is what deserves caution.
$CAP Current fee rate -0.5534%, closed in the past 24 hours -2.809%, at the 4th percentile of the most recent sample. Prices fall and positions increase, with new leveraged funds participating in this downtrend. Bears continue to pay and increase positions; crowding still causes price feedback; Once positions increase but the price doesn't move, the risk of covering increases.
$APR Current fee rate +0.1276%, closed in the past 24 hours +0.101%, at the 100th percentile of the most recent sample. Price increases and positions decrease, with a higher probability of exiting old positions. Position reduction has already occurred; the next step is to see if prices can stabilize after position contraction.
$H Current fee rate +0.0586%, closed in the past 24 hours +0.105%, at the 97th percentile of the most recent sample. Price positions are retreating together, reducing pressure is being released, and it's impossible to confirm exactly which side exited based on this data alone. Congestion indicators remain, but risk exposure is decreasing, so let's deleverage this segment first.$ETH ETH stuck at 1880! Good news everywhere, price dead still. 😅
Last week BTC+ETH ETF net inflow $1.1B, plus $6.7M on Wednesday (5 consecutive days). Fidelity requests staking of all ETH, Swiss bank opens gateway for retail.
Price doesn't move.
Whales who bought at $1,637 in June are selling — over 10K ETH moved to FalconX, taking $2.47M profit. Old holders stuck between $1,861-$1,899, less than 2% range all week.#WeakConsumptionFedSplit #OpenAIAnthropicRace #SKHynixCapexSurge BTC and ETH are effectively flat while SOL is weaker, so I would not read today’s tape as a broad crypto risk-on move. BTC holding near $63K shows relative resilience, but without participation beyond the majors, this looks more like defensive consolidation than the start of a clean breakout.
The stronger macro signal is still concentrated in equities, where AI infrastructure, chip capex and the S&P 500 are commanding attention. Until that enthusiasm begins to lift higher-beta crypto assets, I would treat BTC strength as selective demand, not proof that liquidity has turned decisively expansionary.
Just my read, not advice.This week’s cooler-than-expected inflation data is the kind of print that normally triggers a rally across risk assets. Gold and silver did exactly that — moving higher and holding those gains. But BTC and ETH moved the opposite direction: BTC slid to $62,721, ETH to $1,873. Why this is significant: Historically, BTC has often traded on a “digital gold” narrative, especially around inflation-sensitive moments. When traditional safe-haven assets (gold, silver) and crypto diverge following the sam$ETH ETH stuck at 1880! If good news doesn't pile up, it's just sideways 😅
Last week, BTC+ETH ETFs saw a net inflow of $1.1 billion, followed by another $6.7 million in a single day on Wednesday, marking five consecutive days of net buying. Institutional channels are also expanding—Fidelity applied to stake all its ETH holdings, and Swiss banks connected to retail trading channels.
But the price just didn't change.
On-chain signal battle: Whales who built positions at an average price of $1,637 in June are suspected to be reducing their holdings, cumulatively depositing over 10,000 ETH to FalconX and making a profit of about $2.47 million. Technically, the price is stuck in a narrow range between $1,861 and $1,899, with daily fluctuations below 2%, sideways for nearly a week.
There are positive news and selling interest, liquidity is being drained by AI. Breaking above 1900 requires increased volume; otherwise, it will continue to endure. #消费动能转弱, September policies are still constrained by inflation. #OpenAI与Anthropic估值竞赛升温 #海力士扩产提速, whether capital expenditure can deliver returns is $BTC $OKB XRP has waited so many years that the real big market may not depend on XRP itself.
Recently, looking at $XRP again, I increasingly feel that XRP is a very special presence in the crypto world. When other coins rise or fall, people's first reaction is to look at ecosystem, capital, and on-chain data, but XRP often has to look at regulation first. One of its biggest pressures in recent years has been the unclear rules, so whenever there is new progress in US crypto regulation, XRP is brought up again for discussion.
This is actually quite ironic. XRP was originally discussed as cross-border payments, but over the years, the biggest variable affecting its valuation is often not how well payment business is performing, but whether regulators are willing to give it a clearer position. Institutions and retail investors are different; retail investors see opportunities and jump in immediately, but banks, funds, and payment companies fear today when rules suddenly change tomorrow. So the clearer the regulation, the more likely it is not a buying demand one day, but money that previously dared not enter finally having a reason to enter the market.
But I feel there's another side here.
Clear regulation is certainly good for XRP, but it also brings in more competitors.
USDC can be used directly for on-chain dollar settlements, $SOL has been competing for payment and high-frequency transaction scenarios. As stablecoin scale continues to expand, many cross-border demands may not even require an intermediate volatile asset. If banks start issuing their own on-chain dollars in the future, then XRP, which has told many payment stories for years, will instead face a group of stronger competitors than before.
So now, looking at $XRP, I don't want to keep obsessing over "how much regulatory benefits can make it rise." The truly interesting question is: if one day the regulatory mountain is truly removed, what will XRP rely on to continue achieving higher valuations?
In the past, it spent too much time proving whether it could stay at the table. If the rules become clearer in the future, it can finally participate in a fairer tournament, but fairness also means USDC, SOL, traditional banks, and payment giants can enter with confidence.
At this point, the contest is no longer about who wins the lawsuit, but who truly controls users, liquidity, and payment scenarios.
Regulation can free XRP from its previously suppressed valuation, but it cannot create its future.
$XRP The real test may only begin precisely after regulatory issues are no longer an issue.
#XRP #Ripple #SOL #USDC #稳定币 #Crypto #加密货币 #欧易星球#OpenAI与Anthropic估值竞赛升温 I choose to watch the play
The hottest recent event in the primary market is the valuation competition between OpenAI and Anthropic. News keeps coming one after another, valuations keep climbing, and market sentiment is ignited. I hold some AI-related positions in US stocks, but I haven't moved much recently—mostly just watching.
To be honest, the impact of this valuation race on the secondary market is more emotional. No matter how high OpenAI's valuation is, since it's not publicly listed, the direct benefits are limited. But capital will follow this narrative to find benchmark targets, like $NVDA, $MSFT, and even $BTC will benefit from risk appetite. The more this happens, the less I dare to chase.
I once bought a small segment before the $NVDA earnings report, made a little profit, and then exited. Looking back now, if I had held it now, it might have been even higher, but I don't regret it. Because this kind of market driven by valuation news comes quickly and goes just as fast. If you chase it today, tomorrow a report saying "overvaluation" might come out, and it might pull back for you.
There are also divergences in the market. After the funding announcements from OpenAI and Anthropic, the AI sector opened higher but pulled back during trading, with $NVDA volatility noticeably increasing. I checked my account and saw that $MSFT positions hadn't been moved, and profits were given back a bit, but still acceptable. $MSFT these established tech stocks are different from AI concepts that purely speculate on expectations; they have real business to support the bottom. If a pullback really happens, I would consider buying slowly.
But at $NVDA's current level, I won't blindly chase it. The hotter the valuation race, the more it shows the market's storytelling is increasing. The long-term AI trend is fine, but short-term prices already contain too much optimistic expectations. Rushing in now means making money from sentiment, not fundamentals.
In the crypto world, AI concept tokens have also shown some activity recently, but I basically avoid them. Liquidity is too poor, volatility too high, and many projects have nothing to do with OpenAI or Anthropic—they're just riding the hype. This kind of money isn't something I should be making.
My approach is simple: I will continue to follow the AI track, but keep my position within a sleepy range. No matter how lively the valuation competition in the primary market is, I don't use it to guide my secondary market operations. How much others raise doesn't concern me; I only care if the underlying assets I hold are expensive and whether the logic is solid.
To sum up: valuation races are other people's lively events, but I hold my own positions. The market never lacks opportunities; what it lacks is patience and discipline.
#OpenAI与Anthropic估值竞赛升温 In July 2024, the U.S. SEC approved the first batch of spot Ethereum ETFs. But all approved products are forcibly stripped of a core capability—staking. What investors are buying is nothing more than a passive tool to track ETH prices, essentially no different from a Bitcoin ETF. On August 12, 2026, everything changed. Fidelity has submitted a revised document to the SEC: its Ethereum fund FETH plans to use up to 100% of its ETH holdings for staking and distribute staking rewards to investors in the form of quarterly cash distributions. The fund retains 85% of staking rewards, with 15% allocated to custodians and node operators. This is not an isolated case. Six days ago, Grayscale signed a new trust agreement setting staking to a default of 839,556 ETH and committing quarterly cash payments to shareholders. Going further, BlackRock's staked ETH product ETHB launched in March, and Morgan Stanley's Ethereum ETF was officially approved with an ultra-low fee rate of 0.14% and a staking ratio of 50%-80%. A silent "yield revolution" is rewriting Ethereum's investment logic. Part 01 What Bitcoin Can't Do The essence of Bitcoin ETFs is extremely pure: holding BTC and reflecting price fluctuations. No dividends, no interest, no cash flow. If the price doesn't change, your return is negative—because you also have to deduct management fees. Ethereum ETFs will complete a structural leap in 2026The Federal Reserve quietly loosened its grip, and Bitcoin's curse has finally loosened by half a degree.
Guys, don't just focus on the candlestick chart today—the real hidden cards are in Washington. The latest July retail sales unexpectedly fell 0.6%, and the consumer wagon is finally showing signs of fatigue—the market isn't betting on the data itself, but on forcing the Fed to change its stance.
What was the market most afraid of before? Interest rates stubbornly refuse to cut, wallets tighten and tighter, and Bitcoin can only sit on its back and gasp. Now the logic has quietly changed: the economy is starting to take a chill down, and the rate hike knife is actually less scary, and risk assets can finally hold their heads high. Wall Street actuaries are quietly cutting their bets on rate hikes, and capital sentiment has shifted from "fleeing" to "wait and see."
Personal view: This rally isn't because Bitcoin suddenly gained power, but because the macro tension that had been stretched for a year has finally loosened half an inch. But to pour cold water—just because rate cut expectations are coming back doesn't mean a surge tomorrow. The market still depends on whether real money is willing to flow back; just talking won't last three days.
Next, watch two things: first, whether BTC can firmly hold above key resistance levels; second, if funds recognize the gains, ETH and AI sectors are very likely to take a dip. Conversely, if this wave of positive news is digested early, short-term profit-taking will emerge, and the sell-off will not be light.
In short: The Fed's winds are starting to shift, and the cloud over Bitcoin has indeed faded a bit. But the real opportunities often hide when most people are still hesitating and haven't realized it—by the time everyone understands, the soup is gone. $BTC #消费动能转弱, September policy is still constrained by inflation. #OpenAI与Anthropic估值竞赛升温 #海力士扩产提速, whether capital expenditure can deliver returns is still important Core changes this week: expectations for Fed rate hikes have finally materially eased
Both CPI and PPI data have declined, and these double positive factors have led to no decline in US stocks at all
Macro pressures have eased effectively month-on-month, but US Treasuries remain at their peak, regional conflicts remain unresolved, and AI valuations are bubbled. Although the data is positive, it is only temporary and not yet the time for a full bullish return
For the crypto community, the positive news is not very important; ultimately, what matters is whether it can increase the amount of capital
With interest rate hike sentiment cooling and the US dollar and Treasuries falling simultaneously, this is the real window to go long. Otherwise, with so many stories, lots of news, and pitifully little money, the market will continue to fluctuate! $BTC $ETH #消费动能转弱, September policy is still constrained by inflation. #OpenAI与Anthropic估值竞赛升温 #海力士扩产提速, can capital expenditure deliver returns? #OpenAI与Anthropic估值竞赛升温
Anthropic's Q2 revenue exceeded $11.5 billion, at least 14 times year-on-year, jumping from $4.73 billion in Q1 to $11.5 billion in just one quarter—and adjusted operating profit turned positive that quarter. Its annualized revenue threshold surpassed $47 billion in May, already surpassing OpenAI's $40 billion disclosed figure. OpenAI's curve is also steep: annualized $20 billion by the end of 2025, now $40 billion, nearly doubling in seven months, and July alone rose 20% quarter-over-month.
But the "texture" of their growth is different. OpenAI's 40 billion yuan is not just about ChatGPT subscriptions—programming products like Codex are ramping up, the startup advertising business is contributing revenue, and Agent product demand is rising, with a narrative of "multi-engine ignition." Anthropic is more like a "vertical strategy": relying on professional users in programming and workflow scenarios to push ARR to 47 billion, then using a 14-fold quarterly year-over-year growth rate to raise its valuation to $965 billion (based on the latest funding round)—this figure is already slightly higher than OpenAI's $852 billion.
The real "heatup" lies in valuation methods. Anthropic reportedly forecasts banks' 2028 revenue at $190 billion to $200 billion, using EV/revenue multiples to estimate the value. Stretching the forecast period to two years ahead is not common; Jin Shi's report directly pointed out: "This reflects the speed of Anthropic's business expansion and the difficulty of setting valuation benchmarks for a company still massively investing in AI infrastructure." "Without a mature profit model, yet using revenue two years from now as a valuation benchmark itself signals that the valuation race is heating up to a point where "no longer waiting for profits to be realized, directly comparing imagination space."
Primary market liquidity is also accelerating the boost to this race. Lightspeed raised $600 million in a second round of funding codenamed "Project Mercury," extending its hold in OpenAI and simultaneously increasing its investment commitment to Anthropic, redeeming old LPs and bringing in new money to take over primary shares—a typical pre-IPO turnover. AI investor Leopold Aschenbrenner's Situational Awareness fund nearly unsold all public market shares at the end of July, keeping only Anthropic private equity holdings, betting on its possible IPO as a catalyst in the fall. So far this year, IPO financing has reached $256.4 billion, the highest since 2021. The IPO pace of Anthropic and OpenAI itself is the starting gun for the second half of this valuation race.
But the opposing side also has plenty of evidence. At the critical 40 billion ARR threshold for IPO filings, at least seven OpenAI executives have left within eight months, and in the two weeks of August, COO Brad Lightcap and CRO Denise Dresser both left one after another; The guaranteed scale for Nvidia and OpenAI's 10GW data center in Ohio quietly dropped from $250 billion to below $120 billion—computing power giants are discounting OpenAI's execution risks. Anthropic is being treated by Leopold as the "only held position," while rumors of a $6 billion acquisition of AI startup Decart — the more aggressive the expansion, if the $190 to $200 billion forecast for 2028 is not realized, the valuation method itself will become a focal point of controversy.
So the real hook of this main line isn't "which of the two is more worthy," but rather the market's pricing standards: from annualized revenue, to quarterly year-over-year growth, to multiplying forecasted revenue two years from now—the further you go, the greater the valuation elasticity and the higher the risk of realization. Are you betting on the current ARR of 40 to 47 billion, or the 200 billion imagination space for 2028?
#OpenAI #Anthropic #AI估值#Tether首次完整审计: Transparency becomes the focus
Here's the focus: Tether dropped a bombshell—KPMG issued a full "unqualified opinion" audit report. This is the first time since Tether's founding that it has received a full audit opinion from the Big Four accounting firms.
How should this be understood?
First, the severity of the audit far exceeded expectations. The scope of the audit was very in-depth; KPMG conducted a comprehensive review of transactions, systems, and valuations. They even conducted an on-site inventory and inspection of every gold bar held by Tether.
Second, its assets have been fully exposed. By the end of 2025, Tether's reserves exceeded liabilities by $6.814 billion. In Q2 2026, total assets will be $187.75 billion, total liabilities $183.64 billion, and $4.11 billion in buffer reserves. Moreover, compared to the Q1 all-time high of $8.23 billion, excess reserves have nearly halved—with considerable profits, the safety cushion has actually become thinner. How this account was calculated will probably require more careful consideration.
Third, the geopolitical significance of this matter may outweigh its financial significance. Previously, the biggest concern in the market was that Tether had never undergone rigorous audits. Now that KPMG has entered the market, this biggest uncertainty is being removed. When Tether finally released a "Big Four" audit report, Circle's compliance moat, which has been audited by Deloitte since 2024, quickly narrowed. The transparency race between these two stablecoin giants is shifting from "whether there is an audit" to "whose audit is stricter."
The audit was done, but the boots hadn't landed yet. What truly rebuilds trust is how long and how widely this report can last.All my trades are public, and I try to minimize leverage to avoid losing everything overnight. As you can see, AMD is my largest holding here. AMD is one of the companies most likely to experience huge expectations in AI computing power over the next few years, with a target price of $1,500. Currently, AMD is worth about $500, corresponding to a market value of approximately $830 billion, and $1,500 corresponds to a market value of about $2.45 trillion. This goal requires several conditions to be continuously fulfilled. Data center revenue needs to continue rapid growth, the MI450 series and Helios rack-level AI systems need to be delivered on a large scale, the ROCm software ecosystem needs to mature, the proportion of data center AI products needs to increase, the company's gross margin and operating profit margin need to keep rising, and the deployment plans announced by customers such as OpenAI, Meta, Anthropic, and Microsoft need to gradually translate into real revenue and profit. The reason for setting this target price is that the entire AI computing power market is still in an expansion phase, and the world's largest AI customers want more mature high-performance computing platforms. NVIDIA currently holds the strongest position in the AI GPU market and a CUDA software ecosystem, while AMD has become one of the closest general-purpose AI accelerator suppliers in scale. For customers who invest tens of billions of dollars annually in AI infrastructure, having one more mature supplier increases procurement options, improves chip supply assurance, broadens hardware and software route choices, and reduces long-term procurement costs for a single platform最新数据正在释放一个越来越明显的信号：美国消费者开始踩下刹车。 7月美国零售销售环比下降 0.6%，创14个月来最大跌幅，也是连续9个月增长后的首次回落；核心零售销售同样下降约 0.4%。与此同时，7月CPI同比降至 3.4%，虽然通胀有所缓解，但距离美联储真正放心仍有一段距离。 这就形成了一个尴尬局面： 消费在降温 → 经济需要政策支持 通胀仍偏高 → 美联储又不敢轻易转向宽松 市场真正等待的，并不是一次普通数据改善，而是流动性预期出现明确拐点。 📌 对 BTC 来说 BTC 目前大约在 $63K 附近震荡，弱消费数据本应利好降息预期，但价格并没有出现明显突破。近期ETF需求偏弱，加上监管层面部分利好预期落空，市场风险偏好仍然谨慎。 只要 $62K–$63K 区域能够守住，BTC仍有机会维持区间震荡，并等待新的流动性催化。 📌 ETH 则更需要资金回流 ETH 的弹性目前明显弱于市场预期。 如果宏观流动性没有持续改善，仅靠链上叙事很难推动ETH走出独立行情。虽然ETH ETF今年仍有一定资金积累，但近期市场风险偏好不足，ETH想重新获得强势表现，仍需要看到更持续的资金流入。 🎯This bear market may last longer than expected, mainly because the world has entered a rate hike cycle. Whether it's the Fed, Bank of Japan, or European Central Bank, this is the biggest negative factor. Even if the Fed cuts rates, considering that US stocks are already on-chain, the biggest positive is actually US stocks. So I think we must change our mindset and avoid buying meme coins, because the biggest narrative coin already has BTC, but I dare not expect it to rise too high, because BTC is entirely dependent on speculative venture capital; I personally am optimistic about ETH because it has utility value, but you must buy it at low prices; As for other altcoins, unless they can be equitized, be cautious about buying them. The era of financial market unification has begun.