Orbit Post Sitemap
Banks have agreed: AI can view accounts but cannot move money
This week, major US banks collectively drew the line: JPMorgan Chase, Citibank, Bank of America, and Charles Schwab all confirmed that third-party AI Agents can only read balances but are not allowed to log in or make transfers. Dimon directly used the term "out-of-control agent," and Schwab said they intercept unauthorized agent access in double digits every week. The trigger was that some agents were found attempting actions beyond user instructions. What banks really fear is liability: if an agent transfers money incorrectly, who is responsible? The other side of this issue is that AI managing your money is already happening, and rules are being forced into place. The authorization and limit mode of on-chain wallets might actually be the answer that was thought through earlier. #美国多家银行限制AIAgent访问客户账户 $BTC The data is good data, yet the stock price still took a hit
This Moderna situation is quite surreal: the personalized melanoma vaccine phase 3 data in collaboration with Merck showed a 42% reduction in risk of recurrence or death, marking the first positive phase 3 result for mRNA cancer therapy, which should be a milestone. However, the data was leaked first on a tumor website, forcing ESMO to confirm the leak and release the data early. On Thursday, the stock price dropped over 25% intraday and closed down 17.6%. The issue isn't the data but the expectations: this stock has risen five to six times this year, and the market used the phase 2 result of 49% as a benchmark, so 42% was seen as "below expectations." Today at the Madrid ESMO conference, the full data set was officially released, and we will see how the market calms down and reprices — between the milestone and the stock price lies something called expectations. #Moderna股价大跌，癌症疫苗数据被提前披露 $MRNA $BTC 1.1 billion inflow in three days, 635 million outflow in one day, the face of the ETF changes suddenly
This week, ETF funds played a roller coaster: net inflow of about 1.08 billion USD from Monday to Wednesday, and a net outflow of 635 million in one day on Thursday, the largest single-day outflow since June 25 — IBIT ran 254 million, FBTC ran 218 million. Interestingly, the price: BTC fluctuated around 81,000 on Thursday, with a low of about 81,000, and about 80,900 this morning. Such a large sell-off without breaking through indicates there are buyers below, it's just a question of whether they are willing to chase. Before next Tuesday's FOMC, funds will most likely continue this pattern of inflow and outflow day by day. My positions at 81,500 and 80,800 are grinding in the cost zone; as long as 80,000 is not broken, the stance remains unchanged. #BTC现货ETF资金反复，市场买盘能否企稳？ $BTC Last night the White House announced: with Melania's mediation, Russia and Ukraine agreed to a partial ceasefire in Oleshky starting October 29, allowing humanitarian aid in and civilian evacuation, limited to one city.
What does this mean? The ceasefire is a pause button, not a delete button; a full ceasefire is still far off; negotiation channels remain open, extreme risks are cooling down; but the three major mainstream cryptocurrencies continue to fall—the pressure on the market is not geopolitical but due to deleveraging ahead of the FOMC announcement.
From what I see in the market: total open interest across the network dropped 1.62% in one day to 140.8 billion; the three coins' open interest fell 1.23%, 1.51%, and 1.11% respectively over 24 hours; BTC holds at 80,926, ETH at 2,382, SOL at 102.55. Prices fall and positions reduce, with exits stacking liquidations and deleveraging, like the tide pulling away people and chairs together 🌊. In 24 hours, 470 million long positions were liquidated, accounting for 67.6%; in the 4-hour window, 196 million with 90% longs. On the 1-hour scale, open interest has reversed, increasing by 0.18%, 0.42%, and 0.45%; whether this is the end or a pause is too early to conclude.
My stance is clear:
① Don't chase shorts at 80,800; longs have liquidated for two days straight, chasing shorts here is taking the last baton;
② Don't bottom-fish longs; open interest is still dropping, ETFs have seen nearly 7,900 units outflow in two days, bottom-fishing is catching a flying knife;
③ Before the event settles, I will just watch 👀; the ceasefire takes effect on the 29th, FOMC announcement is next week, leave trades for those who understand the timing.
Whether the gunfire truly stops or not, whether the rate decision is stable or not, I'll be sitting on the sidelines watching. If you don't understand, not acting is the best action.Last night $CL rose nearly four points, Brent crude hit a seven-month high; gold fell below 4,000, dropping more than four points in a week. $BTC hovered above 80,000, $ETH lost 2,400 again. Three things, three reactions, but actually talking about the same matter.
Oil rising reflects inflation expectations, gold falling reflects real interest rates—the yield is still above 5%, holding non-yielding assets has too high an opportunity cost. And BTC holding firm is precisely because the 10-year yield has fallen slightly from its high, giving it some breathing room.
So the current pricing logic is very clear: BTC neither benefits from safe-haven demand nor inflation hedging; it feeds on liquidity. Gold and oil price "inflation," while BTC prices "more or less money."
The funding side is also hedged on both ends: ETFs still had net outflows last week, institutions are withdrawing; but the number of whale addresses hit a record, big players are adding. ETH is weaker, ETF outflows are even more concentrated than BTC, after losing 2,400, 2,350 is the next support.
Downward, 80,000 is the critical point; breaking it means 78,000; upward, 83,000 remains the wall that has been tested several times.
#中东原油供应风险再起，油价持续走强 After a 6.7% plunge, a rebound the next day—what drama is the storage stock playing?
On Thursday, the Philadelphia Semiconductor Index fell 6.7%, Micron dropped 6.3%, SanDisk fell 8.5%, with widespread calls that the AI bubble is about to burst; on Friday, Micron closed up 2.5%, and SK Hynix rose over 6% pre-market. The fundamentals didn’t change in a day: Western Digital’s revenue rose 55% year-over-year, exceeding expectations; Jensen Huang said no memory oversupply is expected before 2029; Citibank said tight supply could last until 2028. The only thing that changed was positioning—after big gains, investors run at the first sign of trouble. SK Hynix reports earnings next Tuesday, SanDisk next Thursday; that’s when we can draw conclusions about the "repeated fluctuations." Until then, the ups and downs are just emotions trampling each other. #AI需求支撑存储市场，芯片股行情反复 $MU $BTC First, consider the worst-case scenario: GRAM has a negative funding rate that continues to widen, while the price keeps falling. Although shorts have to pay longs, this does not mean the market will immediately reverse; rather, it could indicate increasing market crowding, volatility, and liquidation risk.
In this data, the Hyperliquid perpetual contract shows that GRAM's current funding rate is -0.0075% per hour, annualized at about -66%, and -5.97% over the past 24 hours; XMR is at -0.0055% per hour, annualized about -48%, and -4.85% over 24 hours. A negative funding rate means shorts pay longs.
My judgment is: this looks more like a risk signal and should not be simply taken as bullish. ⚠️ First, when the negative funding rate persists, the holding cost will squeeze shorts, but if the price does not hold, it may continue to weaken; second, the high annualized figures are eye-catching but are just a linear extrapolation of the current rate and do not guarantee stable returns; third, if small-cap liquidity and order book depth are insufficient, sharp volatility will amplify slippage and liquidation losses.
When ordinary people see such extreme funding rates, first verify the funding settlement cycle, position leverage, and worst-case loss amount. Don't ignore price risk just to earn funding fees. When you trade, how much stop loss do you usually set per trade?UniSat is integrating with Utexo this time, with specific timing and subsequent plans provided.
On October 25, it is expected to first support the Bitcoin Signet testnet, allowing users to participate in testing USDT minting and redemption using the UniSat wallet. Utexo plans to launch the mainnet on November 28. UniSat will first integrate the wallet, then bring USDT trading into UniHexa.
The most direct impact on retail investors is that in the future, using USDT within the Bitcoin ecosystem may no longer require detours through other chains and exchanges.
The benefits for UniHexa are even more obvious. Currently, Bitcoin assets are mostly priced around BTC, which is not intuitive, and it is difficult to convert back to stablecoins during market fluctuations. With USDT, if trading pairs and liquidity follow, market makers will find it easier to quote prices, and retail investors will find buying and selling BRC-20, Runes, and other assets much more convenient.
Utexo needs wallets and trading access, which UniSat already has. UniSat can also leverage USDT to connect the wallet, exchange, and UniHexa, giving users more reasons to stay within its own products.
We will see if the October 25 test can launch on time, and then observe the recharge, redemption, and order book depth after the mainnet launch on November 28.
The introduction of USDT is positive, but the real determining factor is whether there is capital willing to stay and trade.
#UniSat #unihexa #Utexo #Bitcoin $ETH's most classic deflationary narrative is being harshly contradicted by reality!
In the past, when talking about Ethereum, many people liked to highlight the burn mechanism.
The more active the transactions, the more burns occur, giving $ETH a chance to decrease in supply. Coupled with the staking mechanism, it sounds like a story of an increasingly scarce asset.
But the reality is: in the past 30 days, Ethereum has only burned about 2.7% of the newly issued $ETH during the same period.
In other words, the burns have not offset the new issuance at all.
Currently, the annual growth rate of $ETH supply is about 0.86%, while Bitcoin's is about 0.82%. Their issuance mechanisms are completely different, and this number cannot simply be equated to which is more valuable, but at least it shows one thing:
Stop treating $ETH as an inherently deflationary asset!
Ethereum's $ETH burn mechanism has not disappeared, but its actual effect depends on network activity and fee levels. If network demand is not strong enough, the burn naturally cannot support the deflationary narrative.
What’s more awkward is that when the market starts discussing continuous ETF outflows and upgrade delays, the scarcity story of $ETH also fails to provide an indisputable positive.
#10月FOMC临近，美联储暂停加息还有哪些变数？ #AI需求支撑存储市场，芯片股行情反复 #中东原油供应风险再起，油价持续走强 #SK海力士Q3财报在即，HBM4能否提振盈利？
HBM4 can boost profitability, but Q3 is just the starting point, not the main realization window. The main driver of Q3 profits remains the price increase of traditional DRAM/NAND; HBM4 began mass production in Q2, and its revenue proportion will gradually rise in the second half of the year, with a more obvious pull on ASP and profit margin in Q4. Citi expects Q3 operating profit to be 81.3 trillion KRW, about 8% above market consensus, while conservative institutions have lowered it to around 75 trillion due to the depreciation of the Korean won. The key points of the earnings report are: whether traditional storage price increases can maintain profit margins despite exchange headwinds, and whether HBM4 yield ramp-up and competition from Samsung and Micron will suppress pricing. If Citi's forecast is realized, it will validate the resilience of the upcycle; if it falls within the conservative range, market focus will quickly shift to the Q4 HBM4 volume ramp slope.$BTC Folks, did you catch the short positions on Bitcoin?!
Yesterday's movement followed the analysis from the day before perfectly, dropping to 81,000 then rising near 82,000 to lure buyers. This morning it broke below 80,500, down to 80,461. Between 6 and 7 AM, there was a sharp drop that triggered over 8,000 long positions. Congrats to those who followed along and profited, which shows that when the overall direction is right, you definitely can make gains.
So, what’s the outlook for Bitcoin next?
After breaking below 80,500, Bitcoin found support and rebounded around 80,461, currently back above 80,900. It didn’t continue to fall at the low point; instead, capital came in to support, indicating strong support around 80,000. However, there is significant resistance between 81,300 and 81,500. Without substantial buying pressure, it’s likely to oscillate and consolidate within this range.
Next week’s Federal Reserve meeting is critical. Although the market expects no rate hike this time, if the post-meeting statements are hawkish and hint at hikes by year-end, that would be negative for risk assets like Bitcoin. Additionally, tensions in the Middle East have pushed oil prices up, increasing inflation pressure and making the Fed more cautious. Plus, recent slight outflows from ETFs suggest some institutions are being cautious in the short term.
Overall, Bitcoin is consolidating and building a base in the short term, waiting for clearer direction. The recommendation is to watch and hold the 80,000 support line, and consider entering only after breaking above 81,500 for a safer move.
#10月FOMC临近，美联储暂停加息还有哪些变数？ #SK海力士Q3财报在即，HBM4能否提振盈利？
$SKHYNIX suffers a "double kill" spike, are retail investors still catching flying knives? It's not just a single negative factor: Meta's earnings report sparks doubts about AI returns, the Philadelphia Semiconductor Index plunges 6.7%, Hynix ADRs plunge 8%, foreign investors net sell over 550 billion KRW in a single day; combined with the withdrawal of Solidigm's US IPO, all positive news is exhausted. The market is even messier: a whale lost $4.7 million in a single trade, longs are liquidated, funding rates turn negative, and liquidation stampede occurs at 940-950. Is this a golden pit or a bottomless hole? Don't let the hands catching flying knives end up catching their own liquidation orders.ETF net outflow of 635 million in one day, but none of the three "institutional favorites" crashed.
The outflow is redemption, not a dump. These two things are very different.
$XRP 1.30, 24h -1.1%, market cap 82.2 billion, OI 2.77 billion down 2.5% in one day. Altcoin spot ETFs are still seeing net inflows, institutions haven't left.
$SOL 103, Alpenglow mainnet just launched on 10/22, positive news landed but still fell, a textbook case of selling the expectation. 100 is the watershed.
$OKB 110.2, market cap 2.31 billion, 7-day -4.9%, 30-day -6.3%, still 51.5% below ATH 228.74. The 21 million max supply plus buyback and burn is its floor, resistant to drops but don't expect it to lead the rally.
Institutions redeem shares, not faith. The moment you panic is when unrealized losses truly turn into realized losses. I can understand $ETH falling, but why is Solana's ecosystem growing so fast while $SOL's price is playing dead?
I can understand the outflow from ETH's ETF because the staking yield is only 2.6%, but US Treasuries are already at 5.1%, so institutions definitely prefer the risk-free US Treasuries. That's where this month's 1.18 billion outflow comes from.
So ETH's RSI at 35.2 is very low, the rebound can't even reclaim MA7 (2417), which is 4% below MA20 (2481), and the 30-day range position is 14.5%. There's strong resistance above, and 2352 is today's support line below.
But look at Solana's ecosystem: since early September, the network's new wallet count has increased by 124%, daily active wallet peak reached 4.27 million, Q3 application revenue was 365 million USD, a 42.6% quarter-on-quarter increase, leading public chains for ten consecutive quarters.
Yet SOL's RSI at 35.7 is close to ETH's low, the price is only 2% above the 30-day low of 100.42, and all three moving averages (MA7 104.7, MA20 108.2, MA30 112.1) are overhead.
It really has to be $BTC. Although BTC's three moving averages (MA7 82,100, MA20 82,600, MA30 83,200) are pressing down, RSI is 42.7, the 30-day range position is 15.7%, which is much better, and there's strong support at the 80,000 level below.The leading effect in the DEX market is becoming increasingly obvious, with $UNI still firmly holding the top position.
In the past 7 days, Uniswap's trading volume reached $17.2 billion, with a market share of about 44.7%, ranking first.
PancakeSwap $CAKE closely follows, with a trading volume of about $11.7 billion, accounting for 30.4%.
Together, these two platforms account for about 75% of DEX trading volume, leaving all other platforms to compete for roughly a quarter of the market.
This also shows that in the DEX sector, having more projects does not mean opportunities are more dispersed. Platforms with real liquidity and user bases still dominate the vast majority of the market.
But what really puzzles people is why CAKE isn't rising? Its price increase is not only less than UNI's but also less than AERO and RAY. It can only be said that the BSC-based clones never intended for retail investors to make money. 【NEAR Falls Below $4.6, Is the Trading Range About to Break?】
$NEAR's trend is a bit rough. A few days ago, it was fluctuating above $5, but now it has dropped back to around $4.64, with the short-term lows being repeatedly tested. It looks like a consolidation range, but the buying power inside is weakening.
From the 1-hour chart, NEAR has fallen below the MA7, MA25, and MA99, showing clear short-term weakness. The $4.60–$4.65 range is the current key area; on October 23, the price once dipped to around $4.59. If it can quickly recover to $4.65–$4.70, there’s still a chance to return to the range; if it stays below $4.60 and the rebound fails to hold, be cautious of the consolidation range moving lower, with the next support around $4.50–$4.55.
Another detail worth noting: during NEAR’s previous decline, there were long liquidations in the futures market. If the price drops and open interest decreases simultaneously, it may indicate leveraged longs exiting; if the price continues to fall but open interest rises, watch out for new short positions entering.
Right now, the most important thing is not to rush to guess the bottom but to see if $4.60 can hold. Only a recovery above $4.70 justifies discussing a false breakdown; continuous failure to hold $4.60 means preparing for a range breakdown. The old idea of buying at support calls for more caution now—first, see if the market can show real support.ETH held above 2350 over the weekend, showing more of a bullish digestion rather than a crash.
Current price is about 2383, down approximately 1.3% in 24h, with the low at 2352 being supported. ETF: After an outflow of about 110 million on 10/22, there was a net inflow of about 4.4 million on 10/23 (Farside). October still shows a net outflow but with daily stop-loss. Staking is even more interesting — about 1.36 million ETH queued for entry (about 24 days), with exits almost cleared out. Total network staking is about 43.77 million tokens, with chips continuing to be locked.
Before the Fed on 10/28, do you expect it to first rebound to 2400, or continue to range between 2350–2400? Feel free to share your thoughts.
$ETH
Just personal observation, not investment advice. #10月FOMC临近，美联储暂停加息还有哪些变数？ $ETH $BTC Smart Money Position Snapshot: Both Bulls and Bears Are Losing, Range Tug-of-War Continues ⚔️
$BTC current price is about 81816, Smart Money nominal long-short ratio is 49.62%, with 1044 longs and 1044 shorts, an even split in numbers, but positions are clearly asymmetric 📉.
Longs: holding 1.11B USDT, average price 85510.18, unrealized loss of 50.2748 million USD, only 38.98% of accounts are profitable;
Shorts: holding 2.24B USDT, average price 76250.03, unrealized loss of 152.69 million USD, 39.94% of accounts are profitable.
The price is exactly squeezed between the two cost levels: the upper long cost is relatively high, the lower short cost is relatively low, the market sweeps up and down, forming a typical "long-short double kill" 🧨. Both sides have less than 40% profitable accounts, indicating most Smart Money accounts are also suffering losses and trapped.
In the short term, the probability of range-bound oscillation remains high, and a one-sided breakout is not yet confirmed. Don't rush to chase highs or sell lows; wait for the price to effectively leave one side's cost zone before judging the direction more reliably 👀.
For market reference only, does not constitute any trading advice ⚠️ #MetaQ3营收超预期，巨额AI投入引发担忧 #存储芯片股再度下跌，AI投资回报担忧加剧 SpaceX公布的Q3数据很强：营收146亿美元，同比增长161%；净利润44亿美元，上年同期还是亏损11亿美元；调整后EBITDA 88亿美元。公司称，增长来自连接业务和Super Intelligence相关业务，季度还把算力规模扩到1.7GW。 与此同时，马斯克表示SpaceX和特斯拉已经成立独立特别委员会，讨论潜在合并。注意关键词是“讨论”和“特别委员会”，不是已经签署协议，更没有公布换股比例、估值、交易时间或监管路径。 这件事最值得问的不是SpaceX值不值钱，而是合并后谁拿到什么。SpaceX的收入结构、AI基础设施、Starlink现金流和特斯拉的汽车、储能业务，风险与估值方法都不同。若用高估值资产换低估值资产，股东会面对稀释；若换股比例偏向一方，另一方股东就会质疑公平性；若公司治理仍高度集中，特别委员会能否代表小股东利益也会成为焦点。 我的判断：SpaceX财报为“合并故事”提供了更强的基本面叙事，但还没有把故事变成可交易的条款。下一步真正有信息量的文件，应该是委员会公告、估值方法、换股比例、投票门槛和监管披露。 在这些条件出现前，股价上涨更多是在交易预期，而不是交易已先别急着把下跌当成趋势反转，82k附近的BTC更像在重新定价宏观预期，而不是单纯走弱。 如果FOMC还没开、市场却先跌了，那这波到底在提前交易什么？ 早盘看到BTC 81,327跌2.57%、ETH 2,392跌2.52%、SOL 104跌2.16%，三大主流同步回撤，第一反应很容易是恐慌。但细看节奏，这更像一次事件重定价，不是启动，也不是派发，而是分歧阶段的前半段。市场在提前消化10月FOMC的变数，Fed停步的叙事并没有消失，只是被短期不确定性抢了镜头。 真正值得琢磨的是科技侧的信号。Meta Q3营收超预期，但市场盯着的是它庞大的AI投入；存储芯片股继续走弱，对AI回报周期的怀疑在升温。这条线传导到crypto很直接，风险偏好先收缩，高beta资产先被减仓，BTC作为相对稳的那一档，跌幅反而不算最惨，山寨的弹性这时候是双刃剑。 偏多的路径在于，如果FOMC落地后措辞偏温和，被压制的预期会快速修复，ETH和SOL这种提前跌出空间的品种更容易出现情绪反弹，BTC则可能率先企稳给出方向。偏空的风险在于，AI资本开支的担忧如果继续发酵，科技股和crypto的相关性会被放大，风险预算收缩的今天榜单里的存储芯片下跌，核心矛盾不是“AI需求突然消失”，而是市场开始追问：数据中心砸下去的资本开支，多久才能变成足够高的现金回报？ 美光上一季交出了创纪录的财年业绩，并给出更强的下一财年展望；但这并没有让板块永远上涨。存储股的定价同时受三件事影响：DRAM和NAND价格、AI服务器的真实出货，以及市场愿意给这段增长多少估值。 当投资者担心中国本土存储技术进展、AI基础设施支出可能放缓，或者存储周期接近高位时，股价会先交易“未来供给增加和需求降速”，哪怕当前订单仍然强。闪迪这类NAND敞口更高的公司，还会被放在“AI数据中心投入能否持续”的框架里重新比较。 这就是为什么“业绩好”不一定等于“股价涨”。如果涨价和订单已经写进估值，财报只能证明过去；只有出货量继续上修、客户订单能见度提高、毛利率没有被新增产能压下去，市场才会把行情从主题交易推进到盈利交易。 我的判断：现在不是AI存储逻辑被证伪，而是从“只要有需求就涨”进入“需求质量和资本回报要验收”的阶段。接下来重点看三件事：存储价格还能涨多久，AI服务器订单是否兑现，以及供应商资本开支会不会提前释放过剩预期。 你认为这次回调是AI存储10月23日，布伦特原油一度接近115.10美元/桶，WTI升至101.40美元附近。表面看是油价上涨，底层交易的是中东供应和运输风险重新进入价格。 这不是简单的需求变好。市场担心的是航运受阻、保险和运输成本上升、制裁变化，以及库存缓冲不足时任何突发事件都会放大价格波动。原油越接近高位，市场越会把“短期冲高”与“供应持续收缩”分开定价。 对资产市场，关键传导链是：油价上涨 → 通胀预期上升 → 降息预期被推迟 → 美债收益率和美元走强 → 高估值、无现金流资产承压。BTC有时被当作通胀对冲，但在流动性收紧阶段，它更常先表现为风险资产；黄金则要看实际利率是否同步上行。 我的判断是：短线不能只看油价方向。若运输风险继续扩大，市场会先交易通胀和利率；若供应恢复、油价回落，前期被压制的风险资产才可能修复。真正要观察的是油价上涨能否持续传导到通胀预期，以及美债收益率是否跟随突破。 接下来盯三件事：布伦特能否稳定在高位、霍尔木兹相关运输是否继续受扰、美债10年期收益率如何反应。油价继续冲高时，你认为黄金会先受益，还是BTC会先承压？#中东原油供应风险再起，油价持续走强 #ETF single-day outflow exceeds $600 million, can BTC price stabilize?
$ETH marginal variables suppressing the rebound: capital conditions and on-chain selling pressure worsen simultaneously
First, ETF outflows show no signs of easing. Ethereum spot ETFs had a total net outflow of $93.59 million yesterday (October 23), marking the third consecutive day of outflows; among them, BlackRock's ETHA had a single-day net outflow of $101 million, being the main driver of outflows. Since October, the cumulative net outflow has risen to about $1.18 billion, with no signs of capital reversal.
Second, on-chain selling pressure signals appear simultaneously. An institution investing in both Aave and Worldcoin transferred 6,000 ETH (approximately $14.4 million) to multiple centralized exchanges today, a clear potential precursor to selling. Earlier on-chain data shows a large whale sold a total of 45,000 ETH in the past week, valued at about $92.15 million, indicating sustained selling behavior rather than a one-time event.
Third, upgrade expectations fall short in the short term. The Glamsterdam upgrade has been delayed due to potential security risks of EIP-8037; activation of the Hoodi testnet has been postponed, and the originally scheduled date of October 26 has been canceled, with a replacement date yet to be announced. The market's prior expectation of increased short-term volatility from this event has at least been temporarily shelved at the testnet level.$UNI
Uniswap illustrates why decentralized exchange activity deserves more attention than token price alone. Trading volume, liquidity depth, fee generation, and competition from alternative venues all help explain the protocol’s economic position. Yet a successful exchange does not automatically guarantee equivalent value accrual to UNI holders. Governance decisions and the relationship between protocol revenue and token utility remain important questions.Meta这份Q3财报，最容易被两个数字带偏：营收同比增长26%，净利润同比暴增264%。但把利润表拆开看，故事没有这么简单。 公司Q3营收646.01亿美元，经营利润129.88亿美元，同比下降37%，经营利润率从40%降到20%。净利润98.60亿美元，同比增加264%。利润暴增主要受税率大幅下降影响，同时公司确认了一笔约98.9亿美元的青少年身心健康诉讼和解费用。 现金流更能说明压力。Q3经营现金流327.9亿美元，但自由现金流为负21.2亿美元；包括融资租赁本金在内的资本开支达到349.1亿美元。Meta预计全年资本开支1350亿—1450亿美元，Q4营收指引730亿—760亿美元。 我的判断是：Meta的核心广告业务仍然很强，广告展示量增长10%，单价增长14%，AI推荐确实在帮它赚钱。但市场担心的是另一件事——基础设施投入已经先变成现金流压力，回报要靠未来数年兑现。 接下来只看三件事：Q4广告增速能否延续，资本开支是否继续上修，AI带来的新增收入能否快过折旧和算力成本。如果收入继续涨、自由现金流却持续为负，市场会开始重新给“AI投入”定价。 你觉得Meta现在更像是在提前买入$ETH Market on October 24
📈 Upper Resistance
Immediate resistance 2417–2438 4H EMA15/EMA30 + daily high zone, multiple failed tests
Core resistance band 2448–2461 4H EMA60 (2,441) + stubborn resistance limiting rebounds multiple times; regaining this area is necessary to talk about short-term structural improvement
Trend reversal reference 2493–2513 Bollinger Band middle line (2493) + dense upper short liquidation zone (about $605 million above 2513)
2417 is the first real intraday threshold for bulls and bears: only above it can testing 2448–2461 be qualified; otherwise, any intraday rebound can only be defined as a weak pullback.
---
📉 Lower Support
Immediate support 2360–2376 Bollinger lower band (2362) + recent low zone, the first line of defense intraday
Key structural support 2313–2337 previous low area; a daily close below confirms downward structure extending deeper
Dense liquidation risk zone below 2286; if breaking below 2286, mainstream CEX cumulative long liquidation intensity will reach $676 million, forming downward liquidity gravity
The 2360–2376 area is the most direct battleground intraday. Coinbase spot buy orders exceed sell orders by about $3.48 million near-term, showing some willingness to support this area, but the buy-side thickness is still insufficient to reverse the overall selling pressure background. $TSLA
Free cash flow turned positive, so why can't we directly say growth is accelerating?
Tesla's free cash flow in Q3 was $1.531 billion, compared to negative $1.092 billion in Q2, but $3.99 billion in the same period last year. Both quarter-on-quarter improvement and year-on-year decline can coexist; different comparison baselines lead to different conclusions.
Positive cash flow improvement eases short-term funding constraints, but valuation needs to assess whether the improvement comes from sustainable operating capability or changes in capital expenditure and working capital timing. A single quarter turning positive does not automatically mean future investment pressure disappears.
If subsequent operating cash flow is maintained, necessary investments are fulfilled, and revenue growth improves, the quality is more credible; if cash improvement relies on postponed investments, long-term growth expectations need to be more cautious.Fellow company Western Digital recently reported strong earnings and outlook, yet its stock price remains under pressure, indicating high market expectations; however, since it mainly produces hard drives, it can only serve as a reference for sentiment in the storage industry.2,386 has been stepped back again, this time it's not a repeat of the old story.
At 7:50 it was 2,387.59, now only 2,381.50 remains.
24h high is 2,417.84, low is 2,352.16.
The numbers haven't changed, the change is that 2,386 has moved from underfoot back to overhead.
50x short position opened at 2,372.4, marked at 2,381.01.
The profit amount is still -0.09, the card is right below, showing the real floating loss without embellishment.
Those who shouted to take back in the morning are the ones who reversed, now I can only explain it a second time.
Those who get itchy hands and chase longs to add positions during a sideways market are the easiest to die from such fake moves.
Is it counted as a win for the bears now? Not yet, 2,350 has not truly broken.
If the invalidation condition doesn't change, if it stands back at 2,415 and holds, I will admit my mistake and exit.
Before confirmation from both sides, I only hold the original position, do not add positions, and definitely won't call others to follow.
$ETH
#10月FOMC临近，美联储暂停加息还有哪些变数？ #谷歌云业务增长82%，AI投入回报如何？ Fellow company Western Digital recently reported strong earnings and outlook, yet its stock price remains under pressure, indicating high market expectations; however, since it mainly produces hard drives, it can only serve as a reference for sentiment in the storage industry.$ETH JPMorgan Chase announced on October 19 that it has deployed JPM Coin to the Ethereum mainnet, choosing Ethereum as the on-chain settlement layer for the world's largest bank's US dollar deposit token. This is not a proof of concept but a formal product aimed at institutional clients.
In the same week, BlackRock's ETHA fund saw an outflow of 581 million in a single month, with the entire October ETH ETF net outflow totaling 1.08 billion, marking the largest monthly exodus since the product's launch two years ago.
Wall Street is moving into Ethereum, while ETFs are moving out of Ethereum. Both events are happening simultaneously—who is telling the truth?
On-chain fundamentals provide a third clue. Tokenized RWA assets have surpassed 31.1 billion, including 14.77 billion in tokenized US Treasuries, 8 billion in on-chain credit, and 3.22 billion in tokenized stocks. BitMine has been dollar-cost averaging ETH for 65 consecutive weeks, with total holdings approaching 5.54 million ETH, just one step away from the goal of "holding 5% of circulating supply." 43.8 million ETH are staked, accounting for 35.9% of circulating supply, a historical high.
But the supply side shows cracks: 2,969 new ETH are minted daily, while only 14 are burned, resulting in a net issuance of 2,955 ETH. The "ultrasound money" narrative is being undermined by L2 diversion and sluggish mainnet activity. Gas fees have dropped to a five-year low, below 1 gwei, so cheap that almost no one is burning tokens.
#10月FOMC临近，美联储暂停加息还有哪些变数？ $ETH
This drop in Ethereum isn’t its own fault, let me make that clear first.
The 24-hour low was 2352, Binance’s figure is 2350, current price 2380, and half of the drop has already been recovered. The news flash says the Nasdaq had its biggest single-day drop since June; it’s following the external market down, nothing wrong on its own side.
Actually, there’s a number I didn’t notice this morning: at 7 o’clock, perpetual positions were 1.37 billion, now 1.42 billion, the price is falling but positions have actually increased. The fee rate is 0.0019%, longs are still paying, none have fled.
Last month I said on this line once, "The problem is outside, not with it," and it was tested for several days. I say it again today, if I’m wrong again, consider it tuition paid.
#10月FOMC临近，美联储暂停加息还有哪些变数？ LOL, woke up to everyone saying their bot caught $IRYS, claiming it multiplied many times since launch. Honestly, bragging like this is pointless because these so-called "forever profit teachers" have no clue what actually happened with IRYS.
Yesterday afternoon someone told me to watch it, I said no because the whole internet was hyping it. Then suddenly after midnight it started pumping hard, from 0.06 to 0.14. At that point, I checked the data and saw $23 million worth of short positions liquidated, over 3 million long positions liquidated, and spot trading volume was pretty big. I thought about it but still didn’t touch it and went to sleep. Then in the morning, it pumped to 0.5, then quickly dropped back to 0.06, basically the level before midnight. This all happened within three to four hours, and both longs and shorts got hurt. Looking at the data now, only $8 million in longs were liquidated, but $59 million in shorts were liquidated, and the trading volume hit tens of billions.
These "forever profit teachers" say they caught it, but what they don’t realize is the market maker definitely planned this. If there were really a lot of positions hidden at the bottom, they would have to see existing short orders before pumping. Since they dared to pump like this, it means the bottom didn’t include these "forever profit teachers."
Every time a coin pumps, a bunch of clowns come out jumping around. So annoying. $ZEC
Some people take the inflow into ZEC's ETF as a bottom signal, but I won't agree with that today. The news flash says that the combined net inflow of the US ZEC and NEAR spot ETFs in one day was $21.51 million, while at the same time ZEC itself dropped nearly 4% in 24 hours.
The numbers I pulled: OKX current price 1176, 24-hour low 1158, high 1258; Binance's low is 1158, down 3.97%, both sides match. Within one day, it dropped from 1258 to 1158, losing one hundred dollars.
The market is even more straightforward. The last perpetual funding rate settled at 0.0006, the premium is still negative, and leveraged funds don't even bother to show direction; open interest is $120 million, no panic liquidations seen, and no one is rushing to take over.
The ETF inflow is a position allocation, spent monthly or yearly, totally different from short-term dumping. The inflow is real, but it can't hold up either, that's my final word.
#10月FOMC临近，美联储暂停加息还有哪些变数？ #SK海力士Q3财报在即，HBM4能否提振盈利？
Earnings Expectations: Profit Likely to Hit a New Quarterly High
Institutional forecasts for Q3 are relatively concentrated:
- Operating profit: Approximately 76.96–81.3 trillion KRW, about 30% quarter-on-quarter growth, with year-on-year growth possibly exceeding 6 times.
- Revenue: Approximately 98–101 trillion KRW, about 28% quarter-on-quarter growth.
- Profit margin: Likely to remain around 78%
Three possible trends around the earnings report:
- Positive realization: Profit hits a new high, but HBM4 guidance is conservative and shareholder returns fall short of expectations, so the stock price may spike then retreat.
- Better-than-expected recovery: Stronger Q4 ASP, HBM4 ramp-up, and special dividend/share buyback plans are announced, potentially leading to a rebound from lows.
- Continued pressure: Profit meets targets but highlights KRW appreciation, uncertain customer capital expenditure, and HBM4 slowdown, possibly causing continued stock price volatility.
Trading perspective:
Short-term volatility will significantly increase around the earnings report; options implied volatility is already high. If you trade swings, focus on whether volume expands and the price breaks above previous resistance after the report; if the rebound is on low volume, it indicates the market is still cautious about guidance quality. The mid-to-long-term logic remains, but memory stocks have entered a phase of "high profit, high volatility, and high expectation divergence," so chasing gains based solely on profit figures is not advisable. Bullish accounts dominate, yet BTC can't reclaim 82,000: Is the rebound lacking volume or new buying power?
This contradiction is straightforward: the long-short account ratio is at 2.16, clearly favoring bullish accounts, but BTC remains below 82,000. My judgment is that what’s currently missing is confirmation from "spot funds willing to keep buying," rather than simply attributing the issue to insufficient volume.
BTC is at $80,895.5, down 0.63% in 24 hours, ranging between 80,514.5 and 82,060; ETH is at $2,381.42, down 0.56%, ranging between 2,352.16 and 2,417.84. BTC’s 24-hour turnover is about $294 million, but total volume does not distinguish between active buying and selling, nor can it alone prove that funds are accumulating.
The long-short ratio reflects account direction but does not mean bulls have added enough positions; if the price can’t rise, crowded longs may turn into selling pressure during a pullback. First, watch if the 80,500 level can hold, then see if it can reclaim 81,500 and challenge 82,060. Reclaiming 82,060 with increased volume would indicate new buying is taking over; if the 1-hour close falls below 80,500 and the rebound still fails to hold, 80,000 becomes the next level to watch, and this short-term recovery scenario fails. If the bullish ratio remains high but the price keeps falling, I would interpret it as crowded positions rather than a bottom-fishing signal.
$BTC $ETH #交易之声：你的经验值得被听到 $BTC 230,000 bitcoins, $21 billion, moved by the same hand within a single day
$21 billion and over 230,000 bitcoins were transferred from old addresses to new ones in a very short time, marking the largest large-scale UTXO spending activity on the Bitcoin network since 2026. On-chain analyst Darkfost's tracking shows these chips are highly concentrated and likely operated by the same entity.
First, let's clarify the mechanism. UTXO is the smallest unit of Bitcoin accounting, which can be understood as individual bills of different denominations. Coins lying in a wallet are not "spent"; only when transferring, one or more old "bills" are broken down and recombined into new ones. Therefore, a dense large-scale UTXO spending on-chain means someone is moving a long-dormant stock of chips—most of the chips moved this time had been dormant for 1 to 6 months, and about $4.6 billion came from addresses held for even longer, typically known as long-term holders.
The market immediately interpreted this as "dumping," but two details do not support this conclusion.
First, there is no surge in inflows to exchanges. To sell, coins must first enter exchanges' hot wallets, which cannot be hidden on-chain. Currently, there is no inflow of corresponding scale, so it is more likely internal reorganization between addresses, custody migration, or hot and cold wallet scheduling. This article does not constitute investment advice; invest cautiously. #SK海力士Q3财报在即，HBM4能否提振盈利？ The most telling signal in crypto this session isn't the red candle—it's the funding rate. $BTC is stuck near 81,592, $ETH has slipped to 2,402, and $SOL is holding above 104 with a comparatively modest 1.07% decline. Yet perpetual funding remains positive across the board: 0.0082% for $BTC, 0.0077% for $ETH, 0.01% for $SOL. Prices are falling, and longs are still paying shorts for the privilege of staying in the trade. 😌 That contradiction is the story. Overnight, liquidations across the netwoA brief analysis of BTC short-term trends based on Dow Theory, Chan Theory, Wave Theory, volume-price relationship, order flow, and price action (strategy suggestions)
Comprehensive judgment
Dow Theory confirms lower lows, short-term bearish bias
Chan Theory's central axis [80,400-81,000] shifts downward, break below targets 79,700
Wave Theory Y' wave in final phase, 79,700 fourth test = 50/50 ultimate gamble
Volume-price relationship shows three consecutive declining bearish candles + extremely low volume = sign of seller exhaustion
Order flow VA below by 345 points + exhaustion dulling, 78,450-79,000 is the ultimate absorption zone
Price action shows declining bearish drift + volume exhaustion.
Six-dimensional comprehensive view: short-term bearish pointing to 79,700-80,300, but all evidence indicates this is the "last drop" characteristic — selling pressure is exhausting, volume is consolidating, and the ultimate absorption zone (78,450-79,000 old POC) lies just 1,500 points below. The 79,700 test next week, regardless of how it completes, will end the six-week tug-of-war and start a new rally of 4,000+ points.
$BTC Orbit Daily 84x P/E ratio, Intel raises CPU prices for the third time on October 5
Intel raised prices 3 times in 2026, totaling 30%
Q2 server CPU average price increased by 48%, and supply still can't meet demand
Q3 earnings report on October 29: expected EPS $0.40, up 74% YoY, revenue $16.5 billion
But the company itself only guides EPS at $0.38 and revenue midpoint at $16.3 billion
The price hike logic is simple: AI inference consumes CPUs
The ratio of inference data center CPUs to GPUs is approaching 1:1, whereas traditionally it was 1:4 or even 1:8
McKinsey estimates inference will account for 60% of computing power demand by 2030
But the stock price doesn't agree: 84x P/E ratio, more than 35% retracement from the 52-week high
The price hike revenue is real, but the valuation is also really expensive
I think October 29 is the judgment day for the price hike story: if data center growth holds at 50% and foundry losses narrow, the price hike narrative stands; if not, 84x P/E is a hanging sword. Failure points: Q3 revenue exceeds $17.5 billion or Q4 guidance is raised, then the bearish view is invalidated and turns bullish
What do you think, is the price hike a sign of strength, or is it just that they can't cut prices anymore and have to raise them?
$INTC /$AMD
#Intel #EarningsReport Market Strength Ranking
5-minute median slippage, estimated based on market depth
$UNITREE cost differences mainly appear in large sell orders. Simulated slippage for buying/selling at a scale of 10,000 is 0.05%/0.05%, and for 100,000 it is 0.16%/0.27%. The thin side only becomes apparent when the scale increases. A twisted situation: The $ARM perpetual and XARM spot you can buy on OKX are backed by Arm company, which on October 23rd, along with 21 others, officially opposed tokenizing their own stocks.
According to regulatory documents reviewed by Reuters: On October 5th, OKX and NYSE parent company ICE filed with the SEC for a five-year innovation exemption to list over 60 tokenized stocks; the rules give issuers a switch: upon notification, they can submit objections to directly block the tokenization of their own stocks. Now 22 companies oppose this, with Nasdaq accounting for 16, the latest being Wells Fargo, Arm, Intuit, and earlier Alphabet, Microsoft, Goldman Sachs.
Regarding your wallet: XARM spot trades only 235 shares (about $50,000) in 24 hours, ARM perpetual trades $1.4 million in 24 hours, with open interest of $1.23 million, 20x leverage, zero funding fees, and a 0.07% discount. ARM itself just dropped 7.28% on October 22nd—semiconductor pullback combined with valuation concerns.
With US stock markets closed over the weekend, the perpetual contract keeps trading 24/7. Last night at 22:00, the 1-hour candle volume surged to 2,664 contracts, pushing the price to 215.49, then retreating to 210.8. Can this liquidity support 20x positions?
Are you betting this model will advance, or waiting for it to end? $ARM 💻 I treat this fee rate information as a code review: The input is Hyperliquid perpetual public market data, with the highest absolute current funding rate being $GRAM at -0.0075%/hour, annualized about -66%, down 5.97% in 24 hours; XMR is -0.0055%/hour, annualized about -48%, down 4.85% in 24 hours.
My judgment: This looks more like a signal of short squeeze pressure, directly using it to judge a rebound is missing a logical step. Negative funding rates mean shorts pay longs; combined with price decline, it shows bearish pressure but cannot confirm that selling pressure has fully released.
The potential bug lies in the term "annualized." Extrapolating the current hourly rate can easily mislead people into thinking this income will continue; funding rates and prices will change. If the price drop exceeds the fees received, then even collecting fees can result in losses, the program runs normally, but the account still reports errors.
If it were me, I would change the observation logic to: continue monitoring whether the negative funding rate persists, whether the price stops falling, and then combine changes in open interest to assess the degree of crowding. This is my reasoning; the data itself is still insufficient to confirm a reversal. Please help review: To distinguish between "short squeeze" and "continued downtrend," which data would you prioritize adding? Where is the flaw in my logic? #Robinhood交易业务增长能否延续？ Waiting for the second wave on coin, stock, and blockchain#Robinhood交易业务增长能否延续？
Whether Robinhood can achieve better-than-expected profitability will largely depend on the trading volume of high-margin trades (options and cryptocurrencies) this quarter.
Two risks and potential downsides that could lead to earnings falling short of expectations; $HOOD
Cryptocurrency volatility: A significant portion of Robinhood's revenue from trading is dependent on retail cryptocurrency trading volume. Weakness or an overall decline in crypto trading volume could cause trading revenue targets to be missed.
Increased operating expenses: Ongoing marketing for new financial products, international business expansion, and compliance requirements may compress operating profit margins. $xHOOD #SKhynixQ3HBM4Earnings
This is a personal opinion and does not constitute investment advice.
HBM4 is still in the ramp-up phase this quarter, with limited shipment volume, making it difficult to boost Q3 profits; the main expectation is future stimulation. This quarter's profits mainly rely on HBM3E and the price increase of AI memory. The key market points are the ramp-up progress of HBM4, long-term contract orders, and capital expenditure guidance. It is highly likely that business results will be slightly positive, but there is a risk of a sharp rise followed by a fall.PEPE surged 16% in one day but then gave back three-quarters of the gains; I’m not chasing this long upper shadow.
Here’s what I saw: $PEPE on 10/23 (UTC daily chart) once spiked to 0.00000511 but closed at only 0.00000456, losing about 76% of the intraday gain.
The trading volume that day was about 2.4 times the 50-day average.
CoinDesk reported it rose 23% over the week and led its hundred-coin index that day without any clear catalyst; similar tokens like FLOKI, BONK, $SHIB at most fell 16% on the monthly chart.
24/7 Wall St noted that when PEPE rallied, the total market turnover actually dropped by 24.6%.
My take: A single token rally without news or fresh money looks more like a short squeeze than spot accumulation.
The upper shadow wiped out most of the gains, indicating heavy selling pressure above.
What I’m doing: Watching and not chasing or catching a falling knife.
If the daily candle closes above 0.00000511 with volume, my judgment is invalidated; if it breaks below 0.00000438 (the 10/23 low), I’ll watch the 50-day moving average at 0.00000403—closing below that would mean this rebound is over.
Honestly: Is this move by PEPE smart money positioning early, or just a short squeeze creating an illusion for retail traders?
$PEPE $SHIB $BTC
#PEPE #Robinhood链上交易激增，币股Meme成主角 #ETF single-day outflow exceeds $600 million, can BTC price stabilize?
Over the weekend, $BTC is still hovering around $80,000. I think "not breaking down" is more important than "not rising."
On Thursday, the US spot Bitcoin ETF saw an outflow of about $635 million at once. Institutional holding costs are roughly around 81,500, and the market once touched that level. In the short term, 80,000 is treated as a psychological line: only after holding above it can we talk about covering back to 82,700–83,000; if the daily line breaks down, the next support to watch is around 77,600.
Next week, the Federal Reserve will hold its meeting on October 28. The market leans toward pausing rate hikes, but with high oil prices and weak US tech stocks, liquidity expectations are still wavering. Do you think this is a buildup before the meeting or a fake drop?