Orbit Post Sitemap
$ETH fell harder than BTC today, and the answer isn't in the macro environment but within itself: the biggest buying pipeline over the past year is about to close.
BitMine added over twelve thousand more coins this week, pushing their holdings to 6.03 million, just one step away from 0.51% of the supply. But management themselves said they’ve basically finished buying. This statement is far more important than adding positions—a silent institutional buyer who accumulated nearly 5% of the supply will exit once their goal is reached, while the ETF side has been bleeding for almost two weeks without stopping. Between one entering and one exiting, what ETH lacks most isn’t a story, but people willing to buy at this price.
Tom Lee himself is warning about pullback risks. Founders can shout all they want, but positions don’t lie.
On the chart, 2400 to 2500 is the bottom formed by the forced liquidations in early October, and it’s the only thing bulls still hold onto; above that, 2600 is the resistance level that every rebound has hit, then 2800 further up. The price is stuck in the middle, with no convincing momentum either way.
Storage chains collapsed first in Asia, and the retreat of AI trading is spreading onto the chain. High beta assets like ETH are usually the last to be remembered and the first to be cut when funds withdraw.
Before that 50% figure lands, with one less hand to support the bottom, volatility will increase a bit more. Liquidation Map: The 82,000-83,500 Range Is Packed with "Face-to-Face Liquidations"
This is currently the most dangerous microstructure. Hyperliquid whale address 0x8eff holds $20.56 million BTC long positions with 40x leverage, average price 82,750, liquidation price 80,404, only about 4% away from the current price. This address was force-liquidated on August 19 for $61.11 million short positions, and now has reversed to 40x long, putting itself near the liquidation line.
At the same time, trader AguilaTrades opened $10.14 million BTC short positions with 40x leverage, liquidation price 83,432, only about 500 points away from the current price.
The 82,000 to 83,500 USD range is already packed with face-to-face liquidations—leveraged positions on both long and short sides are extremely fragile, and any directional volatility triggered by CPI data could set off a chain reaction of liquidations. $BTC $ETH $ZEC #阿斯麦财报前瞻：EUV订单成关键看点 $BTC buyback news hits low volume at a low point
There are reports that Metaplanet plans to repurchase 96.93 million shares through over-the-counter trading, spending about 25.3 billion yen, but this still needs verification. My first reaction is not "good news has arrived," but that such OTC buybacks usually do not directly impact the order book, so the short-term price transmission is questionable.
The price here is actually more worth noting: the just-closed 5-minute candlestick is at 83,530.1 USDT, close to the low point of the past few hours at 83,375 USDT, and the recent 15-minute movement is only -0.12%, with trading volume lighter than the previous hours. At this position, I'm not in a hurry to find direction; before the news is confirmed, the low volume at a low point looks more like waiting for confirmation rather than having already given an answer.The data hasn't been released yet, but the explanatory framework is already set up.
From Wednesday to Friday, three sets of US data will be released consecutively. The market has been sideways these days, just waiting for them.
Where does this money come from: The market's pricing for an October rate hike is only 19%, and a 25 basis point hike in December is already fully priced in. The difference between these two numbers is just time.
How is this number calculated: Institutions expect the overall CPI to be pushed to 0.6% by oil prices, with core CPI only between 0.16% and 0.2%. Back-calculating, the oil price portion accounts for the majority.
$BTC is fluctuating between 81,000 and 82,500, tried once above 84,000 but couldn't hold, and around 80,000 is the lower support line.
Before the data release, neither side dares to move first; whoever takes a heavy position first is essentially betting on the data.
$BTC Today's short-term position ETH viewpoint sharing
After a rapid drop impacted by the Korean stock market this morning, the 4H K-line has pulled back from the decline, preparing to build positions and ready to take profits at any time.
Currently, the short-term upward trend direction has been broken, but the larger-scale upward trend remains.
• After reaching a stage high of $2805 on September 21, it retreated; accelerated decline occurred on October 7-8, dropping directly from above 2700 to 2406
• Closed below the steep upward trend line for 7 consecutive trading days (this line is around $2760-2770 today, connecting the August 11 low of 1852 and the September low of 2358)
• Also broke below MA20 (around 2627) and MA50 (around 2556), closing below MA50 for 6 consecutive days
• The short-term structure changed from "higher highs and higher lows" to breaking below the early October low of 2651 — confirming the short-term upward trend is broken
Unbroken part — the mid-term uptrend since July
• The long-term trend line connecting the mid-July low (1803) and August low (1852) is only around 1980 today, with the current price 2511 still about 27% above it
• Price is running above MA100 (around 2245, still upward)
• The large structure of "continuously higher lows" since the July low remains intact: 1803 → 1821 → 2356 → 2406
Key observation levels
• Support below is first seen in the 2350-2410 range. #9月CPI公布在即，哈玛克称仍需进一步加息 South Korea's KOSPI plunged 6%, dragging down $800 million in crypto leverage.
In the past 24 hours, the entire market liquidated $803 million, with longs accounting for 86%—172,000 people were wiped out. The trigger point was not in the crypto circle: the KOSPI index plummeted more than 6% in a single day, Samsung and SK Hynix each fell 8%, and Bitcoin was directly smashed through $81,000.
1. What changed? Liquidation concentration was abnormal: $595 million in liquidations were squeezed into the last 12 hours, with Binance alone accounting for half of the total liquidation volume. The largest single liquidation was a $9.65 million BTC long position forcibly closed at $81,773. This was not a slow deleveraging but a concentrated explosion.
2. Why is this worth attention? On the liquidation heatmap of crypto perpetual contracts, SanDisk, SK Hynix, and SOXL appeared—stock contract liquidation volumes even exceeded those of XRP and HYPE. Panic in the traditional stock market is being directly transmitted into crypto leverage through the "stock perpetual contracts" pipeline on crypto exchanges.
3. Which observation is currently more supported? It leans more toward cross-market contagion rather than an endogenous crypto downturn. BTC's own narrative hasn't changed; what changed is the speed of contraction in external risk appetite.
Next, watch one thing: whether KOSPI will stop falling at tomorrow's open. If the Korean stock market continues to collapse, the crypto side's long leverage may have only cleared the first layer. $SNDK The data hasn't been released yet, but the explanatory framework is already set up.
From Wednesday to Friday, three sets of US data will be released consecutively.
The market has been sideways these past few days, waiting for them.
Where does this money come from:
The market's pricing for an October rate hike is down to just 19%.
A 25 basis point hike in December is already fully priced in.
The difference between these two numbers is time.
How is this number calculated:
Institutions expect the overall CPI to be pushed to 0.6% by oil prices.
Core CPI is only between 0.16% and 0.2%.
Working backward, the oil price portion accounts for the majority.
$BTC is fluctuating between 81,000 and 82,500.
It tried once above 84,000 but couldn't hold.
Around 80,000 is the lower support line.
Before the data release, neither side dares to move first.
Whoever takes a heavy position first is essentially betting on the data.
#9月CPI公布在即，哈玛克称仍需进一步加息
#BTC与ETH现货ETF同步流出，机构需求转弱了吗？ #贝莱德转向超配短债，下调美股评级 $BTC BTC negative funding rate and short liquidations occur simultaneously.
As of 15:25 today, the average BTC funding rate is -0.0109%, with shorts collecting fees; however, in the past 4 hours, short liquidations amounted to $34.8 million, while long liquidations were only $38,500. Shorts are collecting funding fees while being liquidated—this is the most contradictory signal in the past hour.
1. What has changed? Open interest shrank by 0.2% over 24 hours but increased counter-trend by 0.6% in 1 hour; long positions account for 58.5% of accounts, but active long trades are only 46.7%. Positions remain, but active selling pressure is suppressing the market.
2. Why is this worth attention? Negative funding rates should force shorts to exit, yet shorts are adding positions; open interest is not decreasing but rising, indicating both bulls and bears are betting simultaneously at the current level, with neither giving ground. Once this standoff breaks, the direction will move quickly.
3. Which observation is currently more supported? It leans more toward range compression rather than trend initiation. A large amount of leverage is also stacked near 81,100 on the upside.
From now on, just focus on one thing: the US CPI tomorrow night. Inflation data will determine which side in this bull-bear standoff will relent first. If CPI exceeds expectations, shorts under negative funding rates could instantly become the chasing side. $BTC TSMC's revenue hits a record high, so why is the market starting to get nervous?
TSMC's third-quarter revenue has set a record, growing about 50% year-over-year, but this hasn't made the entire chip sector complacent. Asian tech stocks are clearly under pressure today; the market's concern isn't whether AI demand exists, but rather the costs, returns, and duration of this expansion race. Stock prices still fluctuate after the revenue announcement, indicating that "good performance" and "exceeding market expectations" are two different things.
This is exactly what makes Thursday's earnings report truly worth watching: whether the gross margin can hold within the guidance range, whether the Q4 outlook will continue to be revised upward, and how management describes the capacity demand for advanced processes and advanced packaging. The market expects profits to continue growing significantly, but the higher the expectations, the more any caution in guidance can be magnified. Revenue reflects orders that have already occurred, while capital expenditures are bets on the future. If customer demand continues to accelerate, capacity expansion is a moat; if growth starts to slow, today's heavy investment could become tomorrow's depreciation and profit pressure.
Therefore, record revenue does not automatically mean the stock price will keep rising. What the market needs is not to hear "strong AI demand" again, but to see whether demand can sustainably translate into profits, and whether profit growth can cover the increasing investments.
Do you think the most critical aspect of this earnings report is the profit margin or the 2027 capital expenditure guidance? #台积电Q3营收创新高，盈利表现待揭晓 Coverage: Gold / Crude Oil / AI Storage Chips (SK Hynix, Micron, SanDisk, also Samsung, Changxin Storage) / AI Industry / Crypto Market (BTC, ETH) / Macro (US Treasury yields, US Dollar Index, rate hike probabilities) # Semiconductor sector concentrated sell-off, South Korea KOSPI down over 6% 1. Core viewpoints 1. Spot memory prices are still rising, contract prices no one dares to quote: Barclays data shows October DDR5 spot prices rose 8% month-on-month, up 20% in the last three months; DDR4 rose 4% month-on-month, up 24% in three months; Samsung first suspended October DDR5 contract quotes, SK Hynix and Micron followed, expected to resume quoting only by mid-November. This means — even sellers themselves believe prices will be higher later, so they hold back from quoting now. 2. The same industry chain gives two opposite answers in price and stock price: Micron closed down 3.88%, falling below $1000, SanDisk down nearly 6%, SK Hynix closed down 10.09% in Seoul, Samsung Electronics down 7.39%, South Korea KOSPI closed down 5.46%, triggering a 5-minute pause of programmatic sell orders during the session. Simply put — the market is not crashing due to prosperity but due to valuation and capital. 3. Broadcom and Intel CEOs said a key sentence on the same day: Broadcom CEO Hock Tan said in Singapore, "I think the biggest constraint may be capital," financing may become a bigger obstacle to AI expansion than semiconductor, storage, or power shortages;Many people focus only on K-line trading of BTC, but they overlook that its essence is a global high Beta asset, with price fluctuations jointly determined by three layers of forces. 【Macro Level】The core anchor is the 10-year US Treasury yield and Federal Reserve interest rate expectations. When interest rates rise, risk-free returns increase, causing funds to withdraw from risk assets; when interest rate decline expectations open up, BTC then has a basis for valuation appreciation. Under the Merrill Lynch clock framework, only when the economy + inflation combination shifts to an easing window will a true bull market be born; a rebound and a bull market are completely different things. Geopolitical wars cannot be simply regarded as positive: at the early stage of conflict, the market prioritizes risk-off, and highly leveraged crypto assets will first experience deleveraging and a sell-off; only when the war continuously pushes up fiscal deficits and weakens sovereign credit will Bitcoin's "digital hard asset" narrative slowly materialize. Do not blindly go long just because of war. 【Liquidity Level】Look at the total market cap of stablecoins, DeFi TVL, and ETF fund flows. Currently, the stablecoin pool expansion is weak, DeFi locked assets continue to be a stock game, and ETF funds repeatedly flow in and out, indicating that incremental funds have not entered the market; the market is just rotating existing funds, with strong but short-lived altcoin pulses. Without new ammunition, any rally is a rebound, not a trend reversal. 【Market Information and Sentiment】The market must distinguish between news-driven and fundamental-driven movements. Short-term positives can only create single-day surges; a true bull market requires resonance among macro, liquidity, and capital factors. The current market is a cautiously optimistic oscillation range, with huge long-short divergences, and can be disrupted at any time by geopolitical news or US Treasury data. The core of trading: do not predict the market, just wait for macroThe AI storage bubble is starting to burst.
South Korean storage giants are collectively plunging.
The market is re-evaluating the valuation premium of AI computing power storage.
$SNDK short positions realized huge profits, 30x leverage, floating profit +16555U, return rate 703%.
Whale data is clear at a glance: 87.58% of short whales are profitable, while long positions are largely trapped.
In the face of trends, don’t stubbornly hold on to the decline caused by the bubble bursting.
$BTC $ETH
#9月CPI公布在即，哈玛克称仍需进一步加息
#半导体板块集中抛售，韩国KOSPI跌超6%
#闪迪美光下跌，长鑫技术进展受关注 🌍 1️⃣ Global markets shift to risk aversion: U.S. Treasury yields fell across the board by about 5 basis points, with funds beginning to seek safe-haven assets again. Asian stock markets faced significant sell-offs; South Korea's KOSPI index intraday drop exceeded 6% at one point, and Japan's Nikkei index also declined sharply. Concerns about AI industry valuations, competitive landscape, and capital returns are intensifying. 💾 2️⃣ Storage chips face continuous selling pressure: $SNDK and $MU, after sharp declines in the previous trading day, still faced selling pressure pre-market, with both stocks hitting new phase lows. Trading plan overview: 📍 $MU: The stock price is approaching the previously monitored $950 area. 📍 $SNDK: The $1400–$1500 range is the previously planned observation zone for reassessment. 📍 $SOXL: A small position was previously established, but it is positioned as a short-term trade; if prices hit new local lows, risk will be reassessed. ⚠️ Note that reaching the preset price range does not mean the bottom has formed. The technological progress of Chinese storage chip manufacturer ChangXin Memory Technologies (CXMT) has intensified market concerns about future competition, supply growth, and pricing power in storage chips. ⚡ 3️⃣ Energy shows relative strength, tech sector under pressure: The current market shows clear sector divergence: the energy sector is relatively favored by capital, while stocks related to optical communications, storage, and semiconductors continue to face pressure. This round of adjustment is not only short-term profit-taking but also reflects investors beginning to reassess the costs and profit realization capabilities of AI infrastructure investments In blockchain security, a single addition error in legacy code is enough to cause concern.
The XRPL report released on October 9 disclosed an integer overflow vulnerability in the payment engine: specially crafted transactions could create XRP that should not exist. The official statement said a patch has been released, and no evidence of exploitation on the public network has been found. This makes me feel that when discussing AI and quantum threats, existing software vulnerabilities should not be pushed to the sidelines.
AI can participate in security testing, but after finding issues, someone still needs to reproduce them, assess the impact, and then have running nodes install the fix. Publishing the report is one step; actually running the patch on the network is another. For quantum risks, the discussion must also include implementation costs. The ML-DSA-44 signature size in the NIST standard is 2420 bytes; moving the new signature scheme on-chain requires re-evaluating the costs of transaction occupancy, propagation, and verification. Simply saying "switch to post-quantum algorithms" is not enough.
I am quite cautious about claims that a single upgrade guarantees complete security. Details on what exactly was fixed, which nodes have updated, and how to pause in case of anomalies should be transparent. Security budgets must also cover routine maintenance, not just be spent entirely on the scariest long-term threats. Users may not understand every algorithm, but project teams should at least clearly distinguish between resolved and ongoing issues.
#AI与量子威胁下，区块链安全如何升级？ The yield on the US 10-year Treasury has risen above 5% for the first time since 2023. At the same time, BTC has only dropped 2.00% over the past week and is currently priced at $83,562.9. Yield is the most direct tether suppressing risk assets—so why hasn't BTC been pulled down this time? Let's start with the news. Over the past 48 hours, the market has repeatedly cited the same logic: rising yields mean higher risk-free returns, causing funds to withdraw from risk assets. But the actual market shows: BTC fell only 0.40% in 24 hours and remained between 80,400 and 87,283 over 20 days; the S&P 500 even rose 0.59% to 7,811.54 points. Prices haven't collapsed, indicating this logic hasn't been immediately priced in this time. Now looking at fund flows. The real money movement is reflected in two channels: first, the US spot ETF saw net outflows of about $221.8 million over the last 5 trading days, totaling $366.7 million over 10 days—money is indeed moving out; second, gold rose 1.43% to $4,216.3, and the US dollar index held steady at 102.21. These three figures together paint a picture of "risk budget contraction": funds are withdrawing from BTC but not moving into US stocks, instead going into gold and cash. The yield rise is not without casualties, but the wounds are not reflected in prices. What does this mean for BTC? The key lies in the cause of the yield increase. If driven by a strong economy, risk assets can usually absorb it because the numerator side (earnings and demand) is improving; if driven by inflation or fiscal$BTC
The ETF data still looks bad. Last week, the Bitcoin spot ETF saw an outflow of $681 million, with a cumulative net outflow of $386 million in October. BlackRock's IBIT had another outflow of $65.4 million on Monday. Institutions are still withdrawing; that hasn't changed.
Leverage data is a bit interesting. Bitcoin's own open interest contracts decreased by 6.13% over 7 days, and the funding rate is at the 2.2 percentile over 90 days. Long leverage is still very low, which means even if it falls, it won't be a waterfall-style chain liquidation. But it also means that once shorts start closing positions, a rebound could come quickly. $BTC
If BTC retests the 82,500-83,000 range without breaking it, I will consider closing previous short positions and even trying a small long position. If it breaks below 82,000 with volume, then I will reconsider shorting.
Tomorrow's CPI is still the biggest risk. The market expects a year-over-year increase of 3.7%, with gasoline prices being the main driver; the September average price of $4.33 per gallon is the highest for this period historically. If the data exceeds expectations, fears of rate hikes will return, and this BTC rebound might be directly pushed back. If it is below expectations, then 84,000 could be the starting point for a new round of gains.$BTC
In short: the market has changed, so the strategy must be flexible #9月CPI公布在即，哈玛克称仍需进一步加息 #BTC与ETH现货ETF同步流出，机构需求转弱了吗？ $BTC has held the 80,000 level twice, bouncing back from 80,985 and 81,000 both times.
Let's break down how this support held this time.
During the dip, all liquidations were longs, with 166 million long positions liquidated in 24 hours, including over 50 million shorts. Shorts only started getting liquidated when the price rebounded to 84,200. The support near 80,000 mostly came from forced buybacks by shorts covering after long stop-loss liquidations, not from new active buying.
Currently, funding rates are deeply negative, with the market average at -0.0109%, the lowest in nine months on some platforms. Shorts are paying.
Bitmine's $ETH holdings are close to 5% of total supply, 99% complete, with only 76,000 tokens left before stopping accumulation. The most steadfast marginal buying is nearly exhausted.
The daily chart is weakening: price has fallen below EMA5, EMA10, and EMA30; MACD histogram turned negative; RSI at 41. The 4-hour moving averages are acting as resistance, RSI at 33.95 has not bottomed yet. The weekly chart remains neutral.
It now looks like a fragile balance after leverage cleanup, with ETFs still seeing outflows.
Bitmine is about to stop; 30-day spot demand remains negative.
Before Wednesday's CPI, the price will likely continue to consolidate between 81,000 and 84,500.
$BTC Woke up from a sleep, and the market flipped completely.
BTC made a deep V, pulling back from the edge of 81,000 to 83,500, and Ethereum also simultaneously reclaimed the 2,500 mark. But let's look at the essence first: yesterday, BTC and ETH spot ETFs both saw net outflows, institutions are still retreating. The question arises, institutions retreating, price V-shaped, who's pulling?
The answer is clear: short covering triggered by whale liquidations near 80,000, combined with retail bottom-fishing, and a fresh start after leverage cleanup. These three forces share a common trait—they come fast but lack staying power. Without real institutional money on the table, the table is unstable. Seven out of ten times, this kind of V is short covering; don't shout reversal just because you see a V.
My measure hasn't changed: a one-day V is just a pause, two consecutive days means retreat. A true reversal requires sustained ETF inflows; money coming back is what counts. Tomorrow night at 8:30 PM, the CPI boot drops, and Harker just hawkishly said "faster and more frequent rate hikes"; if the data is again on the hot side, this V tonight is very likely a false V.
Discipline remains: don't chase deep V, wait for a pullback that doesn't break support before watching. A true reversal won't miss this day; chasing a fake rebound is just handing orders to the market makers. Defense at 80,000 is still there, low leverage before CPI, hold your chips and wait for the fog to clear $BTC $ETH $ZEC A halted tower crane is more honest than any war report—when Trump announced the Russia-Ukraine energy ceasefire on October 11, the substation in Kyiv was still smoking. This kind of "completion ceremony" without joint acceptance signatures is called unlicensed construction in the building industry.
I've been working on super high-rise projects for thirty years, and what I fear most is not the owner changing the plan, but two sets of construction drawings being excavated simultaneously in the same foundation pit. On October 12, US and Russian energy officials and companies began talks on diesel supply; Novak personally confirmed that Russia partially dismantled the export barricades—this is a quietly poured oil pipeline corridor. But the load-bearing walls of the geopolitical conflict were not reinforced in sync. The US warned Kyiv to stop striking Russian refineries or face a cutoff of weapons and intelligence, effectively posting a "No Blasting Allowed" sign at the construction site. The technical side was already discussing pipe diameter and flow, while the political side was still debating who would sign off on acceptance.
Diesel is the secondary structural masonry of the entire energy framework; it does not bear load, but without it, the whole building cannot be delivered for use. Russia's partial lifting of export restrictions is equivalent to opening a side door in the long-closed construction barricades, causing the global oil supply chain's flow to be rearranged. Fluctuations in crude oil prices are not a matter of surface color differences but a redistribution of stress in the foundational raft slab—once expectations change, the reinforcement at every node must be recalculated. The US using military aid and intelligence as leverage to demand a halt indicates that attacks on energy facilities have escalated from tactical actions to core variables in structural safety assessments.
Looking at the linkage with $xMSFT, a US stock token, it is not an independent building but an external cantilever corridor hanging on Wall Street's main structure—rising crude oil pushes up inflation expectations, raising interest rate anchors, and the high-rise valuation of tech stocks must face increased wind loads. The three clues of diesel negotiations, refinery attacks, and military aid pressure intertwine to form an unclosed seismic joint. Market sentiment, from the fear and greed index to funding rates, is all rebalancing within this gap.
From the design perspective, the biggest problem in this round of energy upheaval is not the lack of plans but the lack of unified reporting standards. A ceasefire without joint confirmation is equivalent to opening for business without passing fire safety inspection; ongoing negotiations indicate the foundation is still undergoing geological exploration, with unknown depth. Russia's easing of exports is a construction channel reserved for itself, while US pressure draws a height limit line for the opponent; both sides are waiting for the other's load test results.
My professional judgment is simple: any ceasefire announced based on verbal confirmation must be managed as temporary scaffolding, ready to be dismantled at any time. #russiausdieseltalks"BTC Leading the Rally Does Not Mean Risk Appetite Has Returned"
The market has already given the answer: $BTC pulled back from 81,033 to 83,700, just 0.3% shy of the opening price; $ETH rebounded from 2,409 to 2,532 but didn’t even reach 2,560, with 2,580 still acting as a ceiling; $SOL surged from 101.63 to 108.56, about 3% short of 112.19. As uncertainty rises, funds still prefer to hide in BTC.
The macro environment hasn’t warmed up. Harker remains hawkish, stating inflation has spread from energy to AI infrastructure and electricity; September’s rate hike is done, and the probability of another hike in December is nearly 88%, with Wednesday’s CPI report looming. ETFs are even more straightforward: BTC saw a net outflow of $679 million in the first week, ETH has been bleeding for 10 consecutive days, and ETHA lost $477 million in a single week. BTC has buyers, but ETH is truly bleeding.
This round looks more like a BTC-led recovery, not a full return of risk appetite. Going forward, focus on three points: whether BTC can hold above 84,360—if it spikes without volume, it may still retest 82,800–83,000; ETH must reclaim 2,580, and if it loses 2,536, the correction will deepen; SOL needs to break through 113. Only if all three move together can it be considered a broadening; if only BTC runs, it’s still a safe haven. If BTC falls below 83,000, the recovery is discounted.
SOL’s sharp rise is due to its inherent volatility, not real strength. 💥 The real skill is whether it can reclaim the previously lost ground.
#9月CPI公布在即，哈玛克称仍需进一步加息 #9月CPI公布在即，哈玛克称仍需进一步加息
These rotten old men are really bad; they must have opened short positions before today's data release, which is why they are speaking recklessly in public.🤨
Hedge funds are cautious about stocks mainly because the dual volatility in interest rates and energy markets is repricing risk assets:
The Federal Reserve raised rates by 25 basis points in September 2026, the first hike since 2023, and hinted at possible further hikes by year-end. The backdrop is persistently high inflation, with new Chair Kevin Warsh emphasizing price stability.
The result is that the U.S. 10-year Treasury yield rose to about the 5.2%-5.3% range, reaching a roughly 20-year high. High interest rates directly suppress stock valuation multiples, especially in rate-sensitive sectors. Hedge fund leverage costs also increased, further dampening risk appetite.
The Middle East situation (involving the Iran conflict, Strait of Hormuz shipping risks, tanker attacks, Houthi armed actions, etc.) is pushing and maintaining oil prices at high levels. Oil prices not only fluctuate themselves but also support rising rates by pushing up inflation expectations, forming a transmission chain of "oil price → inflation → interest rates → stock valuations." Europe is particularly sensitive to energy costs.
Goldman Sachs data shows that fundamental long-short equity funds have significantly reduced net long exposure, with global long-short strategies' net exposure only about 0%-9%. This is a direct response to interest rate, energy volatility, and other uncertainties. Fundamental long-short strategies averaged losses in September, while systematic/trend-following strategies performed better. Changxin Technology made an announcement over the weekend, causing the entire global storage sector to lose a market value equivalent to that of one Changxin. Today, the Korean stock market experienced another circuit breaker, which is very likely just a turning point in this AI wave. Coupled with the four major giants crazily expanding their production capacity, the storage sector is generally heading for a deep correction, waiting for the narrative to reemerge before recovering.🌪️ On the radar echo map, a warm and humid airflow from the high-pressure zone of traditional finance is fiercely converging with a cold front over the crypto continent—on October 12, a major international banking institution announced institutional-grade custody for a certain USD stablecoin and plans to collaborate with market makers and brokers to expand institutional trading and collateral management, pending testing and approval.
📡 I'm monitoring this data from the duty room: this is not retail weather; it's a systemic circulation adjustment. Stablecoins have been stuck in "localized showers" for the past three years, but now they are beginning to connect to the main vapor channels of traditional finance. Once the three chains of custody, trading, and collateral management are connected, it’s like installing a stable updraft for this cloud system—cumulonimbus clouds grow into typhoons not by momentary convection but by continuous energy supply.
🌦️ Even more noteworthy is the accounting-level wind vane. The US Financial Accounting Standards Board is soliciting opinions on whether qualifying stablecoins can be classified as cash equivalents. In our terms, this is called "recalibrating the pressure baseline." Once the baseline changes, the institutional capital’s direction will shift as a whole: previously, they could only observe this as "peripheral cloud systems," but in the future, it may be directly included in the core pressure field of the balance sheet. Custody is the ground observation station, trading is the upper-air sounding, and accounting standards are the numerical model—only with all three can it be forecasted.
📡 Next, look at the linkage of US stock tokenized assets like $xCOIN. Its current performance resembles a spiral rainband on the typhoon’s periphery: the main body hasn’t arrived yet, but wind shear is already perceptible. Every handshake between traditional financial institutions and stablecoin issuers raises the "sea surface temperature" of the entire chain, and sea surface temperature is the fuel for typhoon formation. Institutional adoption is not a thunderstorm; it’s a climate state shift—it changes the season, not just the weather of a single day.
🌪️ From my duty experience, the real danger is not the heavy rain but the unusually quiet and sultry period before it. The current market pricing of stablecoins merging with traditional finance is still in the early stage of "low-level convergence and high-level divergence": the ground is quietly storing energy, but the upper atmosphere has not yet formed a complete outflow. When testing and approval are finalized and the collateral framework is operational, that will be the moment when the jet stream strengthens and vertical wind shear weakens—the cloud wall will rapidly close then.
📡 I will continue to increase crypto observation frequency here: closely monitoring the pressure readings of custody approval, the wind direction changes of trading counterparties, and the cloud map evolution of accounting standards consultation. The integration of stablecoins into traditional finance is superficially a channel issue but essentially a major atmospheric circulation rearrangement. #standardcharteredrlusdYesterday, October 12, 2026, marked the end of the 4-year vesting cycle for early investors and the Aptos team.
Key takeaways:
• Scheduled unlocks have dropped from 11.31 million to 4.54 million APT per month.
- Allocations for the community and the Foundation will continue until 2032.
• Staking rewards have decreased from 5.19% to 2.6% annually. The total supply cap is set at 2.1 billion APT.
• The Foundation has committed to permanently locking and staking 210 million APT.
- The token buybaI stared at SanDisk's candlestick chart for a long time. The time in the lower left corner of the screen was 17:46, pre-market. The price was 1434, down 4%.
I admit, for a moment I thought about closing my long position.
Just for a moment.
Then I opened a document and listed all the negative news from today one by one: Changxin Storage's 4F² DRAM breakthrough, Samsung Hynix's $40 billion buyback exit, foreign capital net selling for five consecutive days, Samsung plummeting 6.7%, Hynix dropping over 8%, and Hammock hawkishly saying rates will still rise in the early morning... After listing them, I stared at this list for half a minute,
and suddenly realized a problem: all these are external negatives, none directly hitting SanDisk's own fundamentals.
Data center revenue growth of 437% remains unchanged, Citi's target price of 2100 remains unchanged, and the judgment that NAND supply will be tight until 2028 also remains unchanged. The market is killing sentiment, not logic.
I returned to the market. The 4-hour chart dropped from 1806 straight down to 1408, a 22% decline, with all moving averages in a bearish alignment, and RSI entering oversold territory. This kind of accelerated bottom-chasing move often means the last drop is just ahead. My own long position is still there, with a very small position size, margin only 0.38U, so the forced liquidation price is unreachable. I use the lightest position to try and err; if wrong, I lose only a little money; if right, it's emotional repair.
I don't know if it will still fall tomorrow. But I know one thing—after listing all the negatives, you realize there is actually nothing truly fatal.
I'm just recording today's thoughts. Not advising anyone to follow, nor shouting any slogans. Just feel that this position is worth a look.
$BTC $ETH $SNDK
#半导体板块集中抛售，韩国KOSPI跌超6% The storage sector has taken quite a harsh hit this time, and even I am asking myself if I got in too early?
SanDisk has been pulling back along with the storage sector these past couple of days. The trigger was the breakthrough in ChangXin Technology's 4F² architecture, combined with the post-holiday correction in the Korean market and the early end of the largest buyback in history by Samsung and SK Hynix. In the short term, these are indeed bearish factors, and the market is worried that changes on the supply side will alter the supply-demand balance.
But looking at the longer term, Citigroup just reaffirmed a buy rating on SanDisk in early October with a target price of $2100. The core logic is that NAND supply tightness may continue until 2028. In Micron's latest earnings report, NAND revenue grew 42% quarter-over-quarter, and average selling prices rose about 30%. This is not data typical of an industry at its peak. AI data centers continue to drive demand for SSDs, and pricing power in enterprise storage is shifting from consumer to data center segments.
From a technical perspective, the 1-hour chart shows a long lower shadow around 1408, indicating that there is buying interest trying to support this level. However, a long lower shadow does not mean an immediate reversal; subsequent candlesticks need to confirm it. Short-term resistance lies between 1520 and 1550. Only if volume expands and the price breaks above this range will the rebound's sustainability be worth expecting.
I opened my long position at 1588 and am currently at a floating loss. It’s not easy to accept, but trading is a game of probabilities. The initial logic for buying was the long-term bullish cycle of storage, and that logic has not been disproven yet. As for the short term, I leave it to the market and to discipline—set your stop loss properly, and the rest is just waiting. $SNDK Yesterday, some were still shouting that ZEC would surge to 1400, but what about today?
It didn't even hold 1300; a big bearish candle smashed it straight back to 1208.
Those who chased longs at 1300 probably can't laugh now.
To be honest, I've taken quite a bit of criticism over the past three days. Since $ZEC was pulled up to 1300, the comment section hasn't stopped—some say I'm stubborn, some say I'm going against the trend, some say I'll liquidate sooner or later.
I didn't respond because I knew the market would speak for me.
It rose 200 points in three days, then dropped half of that in one day. With this kind of movement, tell me—is it strong or weak?
No need for many reasons, just two suffice.
First, look at the volume. This rebound from 1111 to 1300 saw trading volume shrink by nearly half compared to the decline. Pumping the price relies on real money, not just shouting. Rising on shrinking volume is just playing dirty.
Second, look at the technical structure. On the K-line, the 1300 level is exactly where the 30-day moving average is pressing down. A long upper shadow candle pushed back—classic false breakout.
The RSI is still hovering below 50; the rebound hasn't even touched the overbought zone, indicating the bulls have no strength.
My short position at 1218 is currently up 41%, not much, but I'm in no rush.
With this kind of movement, the rebound is just giving you a chance to short.
If it dares to surge to 1250 again, I'll add to my position.
The previous low at 1111 is just below; if it breaks, 1000 won't hold at all.
Don't ask why—just ask the market, it told me.
$BTC
$ETH
#台积电Q3营收创新高，盈利表现待揭晓 BTC重新站上8.3万美元，韩国股市下跌、纳指期货走弱，却没能阻止这波反弹。短线走势确实有点硬，但币价撑住了，不代表行业里的其他故事也撑得住。 曾经的以太坊NFT龙头Blur就是个例子。随着交易量较巅峰时期暴跌超过99%，平台最终走向关闭。牛市里，流量和补贴能把市场炒得热火朝天；等热度散去，才知道有多少需求是真实的。价格可以反弹，失去的用户和流动性却没那么容易回来。 而比项目衰退更让人警惕的，是连资产安全都可能出问题。Ledger针对CryptoBilis设备中发现的可疑植入芯片发出警告，涉及钱包密钥通过移动网络外传的风险。从市场流动性枯竭，到硬件供应链出现安全隐患，问题虽然不同，最终影响的都是用户对加密行业的信心。 BTC反弹值得关注，但别把一根阳线当成行业全面复苏。行情决定能赚多少，项目决定能不能活，安全决定最后还能剩多少。ETH at $2510, are you still holding on stubbornly?
ETF withdrawals have continued for 10 days totaling $780 million, the most steadfast bulls stopped buying after reaching 5%, and 2400 was supported twice—but the indecisive position at 2510 might be the last escape window the main players are giving you.
Let's look at the surface first: a dip to 2409 at midnight was forcefully bought back to 2510, with a 24-hour range of 2410-2560. The daily chart shows support near 2400 twice, RSI at 45 neutral, price below the short-term moving average but above the long-term moving average. Everyone is shouting: "Double bottom success, bull market returning fast!"
But I advise you not to get excited yet.
This is not a trend reversal to bullish; it's a recovery after liquidation.
In plain terms: after a wave of bull liquidation, the market makers pulled it back to see how many retail investors are willing to take the chips.
First thing: ETFs are running, institutions are voting with their feet.
The US stock spot ETH ETF has had net outflows for about 10 consecutive trading days. On October 12 alone, $85.4 million flowed out, with BlackRock's ETHA alone withdrawing $75.1 million. Since September 29, the cumulative outflow is about $780 million; last week alone saw $542 million outflow, the worst since January.
Are you still waiting for institutions to support the bottom?
Institutions are dumping their chips onto you.
Don't tell me "long-term bullish." Long-term bullish and short-term selling never contradict. BlackRock talks about tokenization while selling ETH—that's professionalism.
Second thing: the most stable buyer is about to stop.
BitMine (Tom Lee) holds about 6.029 million ETH, close to 4.9% of total supply, just shy of their 5% target. They added 12,500 ETH last week but clearly stated: stop at 5%.
In recent months, BitMine has been the most stable and mechanical spot buyer in the market. Now this "buy buy buy" machine is shutting down.
Guess what, when the most steadfast bulls stop buying, who will take over?
Retail investors?
You can't even support yourself.
Third thing: the unlocking queue hangs overhead, the deflation narrative is long broken.
The staking exit queue recently rose to this year's high, about 770,000 ETH. The potential selling pressure after unstaking is like a knife hanging over your head.
What's more painful: after the Dencun upgrade, mainnet fees dropped sharply, and burning can't keep up with issuance. ETH is now mildly inflationary, not a deflationary asset at all.
Are you still brainwashing yourself with "deflation"?
That's an old script from 2021.
Bull vs. bear, you decide:
On one side:
2400 supported twice, mid-term structure intact
RSI 45 neutral, not oversold yet
ETH/BTC at low levels, theoretically room for recovery
Long-term narratives (tokenization, L2, stablecoin settlement) still intact
On the other side:
ETF bleeding for 10 consecutive days, cumulative $780 million outflow
BitMine stops buying at 5%, the most stable buyer margin disappears
Staking exit queue at 770,000 ETH, selling pressure looming
Fed raised rates in September, dot plot still hints at one more hike
This week's CPI data is hot, ETH gets hit first
BTC at 83,000, ETH/BTC weak, altcoins have no independent rally
Key levels: 2400 is the lifeline, 2575 is the ticket, 2510 is the meat grinder.
Resistance above: 2550-2575 (short MA + recovery line, failure to break means a pullback) → 2600 → 2660-2740 (previous high concentration, volume breakout needed for recovery)
Support below: 2490-2500 (trendline) → 2430-2480 (key defense) → 2400-2410 (lifeline) → 2220-2280 → 2140-2240 (100/200-day MA)
Daily close below 2400 leads to abyss.
4-hour close above 2575 is needed to talk about reversal.
2510? Indecisive, perfect for a wick.
Trading strategy (range-biased, no chasing):
Perpetual longs:
Don't chase at 2510. Consider light positions on pullbacks to 2430-2460 if 1-hour chart doesn't make new lows. Stop loss below 2385. First target 2550-2575, reduce position there. Only if 4-hour closes above 2575, extend target to 2660-2740.
Perpetual shorts:
Don't short at 2510 top. Wait for 2550-2575 to fail with upper wick or volume shrinkage, then short. Stop loss above 2600. Targets first at 2460, then 2410. Don't hold shorts if daily doesn't close below 2400.
Spot:
Mid-term positions can be held but no adding at this price. To add, wait for one of two conditions:
Daily close back above 2575 and hold;
Drop to 2300-2400 with ETF outflows starting to narrow.
Before BitMine stops buying, don't expect mechanical support.
Position rule:
Single loss controlled within 1% of total capital. Lower leverage around CPI. If funding rate suddenly turns very positive, it means bulls are crowded in the rebound; prioritize reducing positions, not adding.
ETH now is like August 2025—
Everyone thinks "it dropped so much, it should rebound," but once ETFs stop and vaults stop supporting, price slowly falls from 2700 to 2400.
Don't mistake a rebound for a reversal, don't mistake recovery for a bull return.
ETFs are running, vaults are stopping, what are you still waiting for?
2400 is the lifeline, 2575 is the ticket.
2510 is in between, suitable for waiting, not heavy positions.
$BTC $ETH $SOL $OKB ICE filing landed, but OKB dropped 11% in 7 days. Where is the problem?
1. OKB's role on X Layer is just Gas; tokenized US stock trading does not consume OKB.
2. The joint venture company does not hold OKB; settlement layer profits go to ICE and OKX's parent company.
3. The trapped positions are in the dense trading zone of $130-$135; whenever it rises, people sell.
In my OKB valuation, the traffic part is already sufficient, but the token value part is still lacking. Short-term follows the market trend; long-term depends on whether the ICE joint venture can find new scenarios for OKB.
However, the story has been quite lively recently. The OKX ICE filing by the 10 groups is real; the joint venture between OKX and ICE, the parent company of the NYSE, submitted a 24/7 trading notice for 63 tokenized US stocks to the SEC, running on the X Layer's Uniswap v4 permissioned pool, with counterparties USDC/USDG/USDT.
Andrew Cuomo is the co-chairman of this joint venture. If this path succeeds, OKX will upgrade from a crypto exchange to a global stock settlement infrastructure. In the long run, I think OKB is a good asset. Sisters, the guillotine for ZEC has completely fallen last night.
My biggest regret now is not decisively adding to my position and rolling it over when it was at 1300. Looking at this candlestick now, I’m filled with regret.
Don’t rush to bottom-fish and go long; first, take a look at the nine-figure USD amount Grayscale withdrew overnight.
$ZEC dropped from 1315 to 1122 in 24 hours, now barely holding around 1213.
The moving averages are in a bearish alignment pushing down, and the MACD green bars can’t recover at all.
This isn’t a pullback; it’s clearly the dog whales preparing a guillotine meal for the bulls.
Looking at Grayscale’s ZCSH ETF, there was a weekly net outflow of $93.56 million, with assets under management falling from nearly $1 billion peak to $751 million.
On October 8th, another $18.66 million flowed out.
Institutions are running, but retail investors are greedily taking the chips.
Going deeper, the most fatal factor now is human nature.
There are too many users going long at this level; everyone thinks it must rebound after such a drop, right?
But the market never rises just because you think it’s dropped enough.
The more it’s like this, the more the dog whales want to kill it further, burying all the bottom-fishing chips.
This human nature game is more real than any candlestick.
My short position at 1656 is now steadily earning 1336% profit.
In terms of operation, you can slightly add shorts when it rebounds to around 1230-1250, with a target firmly holding at 1000 or even 800.
Don’t ask me when to enter anymore; my position is already set here.
$BTC $ETH
#9月CPI公布在即，哈玛克称仍需进一步加息
#半导体板块集中抛售，韩国KOSPI跌超6% Recommendation on the $NEAR Grayscale fund: a drop of 13.6%, bulls have not yet surrendered, and high-beta public chains are also being bloodied.
Following the risk-off trend of the major market, the public chain sector has become a hard-hit area. $NEAR experienced a single-day pullback of 13.6%, ranking among the top decliners alongside BOME (-30%), BLUR (-29%), and STRK (-28%).
1. Reason for the decline: NEAR's previous rally was entirely driven by leveraged funds, relying on three main narratives: ETF expectations, token issuance volume, and the Intents ecosystem. When the price surged to the 5.9-6.0 range, the market was already severely overbought.
When BTC fell below 82,000, triggering a market-wide deleveraging, approximately $19.8 million long positions in NEAR were forcibly liquidated, causing a stampede.
2. Market signals: Current open interest is $349 million, with an annualized funding rate maintained at +11%, indicating that on-exchange bulls have not completely given up, and there remains potential downward selling pressure.
3. Going forward, NEAR will likely oscillate between 4.8 and 5.6 to digest floating supply; if the major market breaks down, the price may drop to 4.5; if the 4.5 support line fails, this round of upward structure will be invalidated, with a downside target of 4.0. Intraday trading range is 4.85-5.35, with an intraday stop loss reference at 4.75. 今早 BTC 还在 80,961 挣扎，下午直接拉到 84,000 以上，日内涨了 1.55%，反弹超过 3,000 点。明天（10.14）晚上美国 9 月 CPI 数据就要公布了，今天这个时间点拉升，耐人寻味。 市场在赌什么？赌核心 CPI 环比 0.2% 符合预期。巴克莱和摩根士丹利都预测核心环比 0.24%，比 8 月的 0.29% 放缓。如果核心真的降到 0.2%，市场会解读为 "通胀见顶，10 月不加息"，风险资产会涨。目前 10 月加息概率已经跌破 20%，市场基本定价了 "不加息"。 但我要提醒几个被忽略的变量。 第一，整体 CPI 可能很难看。 预期同比 3.6%-3.7%，比 8 月的 3.4% 大幅回升，主要是油价驱动。如果整体 CPI 冲到 3.8% 以上，哪怕核心符合预期，市场也会恐慌 —— 因为美联储官员最近反复强调 "通胀还没到目标"，哈玛克今早还说 "倾向小幅高频加息"。 第二，12 月加息概率已经 70% 了。 就算 10 月不加息，市场也在定价 12 月加。这意味着 "通胀见顶" 的利好已经被提前消化了一部分，明天数据出来可能是 "利好出尽"。 第三，中国长鑫储存技术进展导致存储板块集体下跌，这可能是今天最大的热点 其实早在Q2财报季的时候就可以发现，中国在人工智能方面的突破与发展是必然的，并且中国产能效率高，对海外市场释放产能快，所以现有人工智能所有产业链被冲击是必然的 不过，这种冲击短期是对高性价比市场以及中低端市场冲击最为明显，而中国在人工智能的产业突破我认为是起到了“鲶鱼”效应，对全球AI产业链短期是有威胁力，但是中长期却可以促进AI产业走向高效+低价的路径，这其实是好事，“卷”往往在长期来看带来的反而是促进作用 #闪迪美光下跌，长鑫技术进展受关注 不过，虽然短期存储板块被冲击，但是这并不代表基本面被破坏，国际上高端存储以及半导体芯片，目前头部大厂依旧拥有技术优势，起码短期几个季度不用担心被中国产能替代 而今年，韩国半导体以及头部存储企业的跌幅明显，就是我昨天前文所说的宏观压制效果，利好被缩小作用，利空反而被放大，归根结底就是高长债利率+加息预期下，市场流动性信心不足导致 接下里要等周三CPI靴子落地+美国银行股、阿斯麦、台积电财报提振市场，对冲掉目前的利空情绪与信心不足 趋势上，基本面没被破坏，产业叙事依旧存在，宏观压The weakness across memory stocks may be driven by a growing concern: China’s progress in DRAM technology could reshape the global semiconductor supply landscape. Reports surrounding ChangXin Memory Technologies (CXMT) and its development of domestic 4F² DRAM have raised expectations of potential mass production by year-end. If manufacturing scales up successfully, investors may begin pricing in stronger competition, future supply growth, and pressure on memory pricing and profit margins. Elon MIt would be such a pity if Bitcoin and Ethereum keep dropping like this.
I placed a short on Bitcoin at 84350, and it just peaked at 84134.
I placed a short on Ethereum at 2560, and it peaked at 2537.77.
There should be this kind of close miss feeling.
But if it keeps falling, it confirms that this V-shaped reversal is a short covering plus buying support rally.
During the decline, selling volume increases, while during the rise, volume actually shrinks and meets resistance, making it easy to fall back for a second drop.#Storage sector fund divergence intensifies, foreign capital selling Korean manufacturers drives sector down
The market shows some funds exiting, while other funds are accumulating chips at low levels. Foreign holdings data shows Samsung's foreign shareholding ratio has dropped to a nearly 17-year low, with continuous selling of Samsung and Hynix by funds, spreading panic sentiment outward. The storage sector is under overall pressure, and SNDK is impacted along with the sector.
From institutional research reports, Citibank maintained a buy rating last week with a target price of $2100, judging that NAND supply tightness will continue until 2028; Morgan Stanley proposed that AI is triggering a storage super cycle. The company’s data center revenue grew 437% year-over-year, and the Sentinel series new products are positioned in security and automotive storage tracks. 25 analysts unanimously rate it as a buy, with an average target price of $2173, nearly 40% upside from the current price. The short-term selling by funds has not changed the original fundamental logic of the target.
Does the short-term stampede of funds mask the true flow of long-term chips? Like the tide temporarily covering the reef, the reef itself has not shifted.
$SNDK $BTC $ETH #闪迪美光下跌，长鑫技术进展受关注
Storage stocks have dropped sharply these past two days.
On October 12, Micron fell 3.88%, SanDisk dropped 5.97%, and Samsung and SK Hynix also came under pressure. Today at the Korean market open, Samsung and Hynix directly fell over 7%, with the entire storage sector adjusting.
Why the drop? Two reasons. One is that ChangXin Memory disclosed progress on its new generation 4F² DRAM architecture, planning to launch server memory products by year-end. The market reacted to the news of a new competitor entering, raising expectations for future capacity expansion. Although actual mass production capability is not yet verified, sentiment moved ahead of fundamentals. The other reason is that Samsung and Hynix's large-scale stock buybacks are nearing completion, foreign investors are reducing holdings, and after a big run-up, a short-term correction is normal.
What’s the impact?
First, on miner hardware costs: if storage chip prices truly fall due to competition expectations, it’s good for miners as payback periods could shorten. But ChangXin’s technology is still in the validation phase, so short-term impact is limited; don’t expect hardware prices to drop immediately.
Second, the storage stock decline drags down tech stock sentiment, indirectly affecting risk appetite in the crypto market. Bitcoin is already pressured by macro factors, and when tech stocks weaken, short-term sentiment worsens.
Here’s my view.
Don’t panic; this is a normal correction after a strong run. AI demand remains, the storage cycle isn’t over, and short-term profit-taking needs to be digested.
What’s your take?
$BTC $ETH 最近看盘，我越来越觉得，币圈最难的事情，不是看懂已经发生的行情，而是在行情真正启动之前，判断现在到底值不值得出手。 以前我也会盯着K线猜方向：这里是不是底部？突破了是不是要起飞？跌下去是不是该做空？ 后来我开始用历史K线做量化研究，才发现一个很重要的问题： 判断行情会不会变大，和判断行情往哪个方向走，是两件完全不同的事情。 我做过一轮历史数据实验，专门研究大幅波动事件发生之前，哪些特征具有区分能力。 在测试过的多个特征中，有一个现象尤其值得关注：成交量异常扩张。 在我设定的样本和事件定义下，15分钟级别的相关特征在样本外测试中取得了约0.73的AUC。 这里先说明，AUC不是73%的胜率，也不意味着未来73%的行情都能预测正确。它只能说明，在这套历史数据和标签条件下，这个特征对区分目标事件与普通窗口有一定的区分能力。 但真正让我重视的，是后面这个发现。 成交量异常扩张，可能告诉我们市场正在进入值得关注的大波动环境，却不能单凭它判断接下来一定上涨还是下跌。 放量不等于做多信号，缩量也不意味着马上下跌。 很多人看到成交量突然放大，就急着追涨或者抄底。但如果价格已经走出一大段，等你注意到放量Today's market, $BTC and $STRK are like heaven and earth. Bitcoin stubbornly rebounded to over $83,000, seemingly about to reclaim lost ground, but STRK is like a hopeless case, dropping 18% in 24 hours and still falling today. Its recovery ability is really disappointing.
A big player couldn't stand STRK's sluggishness anymore and decisively liquidated over 26 million STRK long positions this morning at an average price of 0.0848. Although not sold at the peak, he still pocketed a solid net profit of $184,000.
Thought he would take a break? Not at all! This guy immediately turned around and used over $11 million at an average price of 81,819 to enter a 40x leveraged BTC long position! This position has already gained $245,000 in unrealized profit, but he also set a full-position stop loss at 79,000, clearly preparing for both outcomes.
Honestly, this rotation logic is very clear: abandon the weak, embrace the strong. But 40x leverage combined with such a tight stop loss really has a bit of a gambling element to it.Unichain has been "sent away," what remains of the $UNI story?#闪迪美光下跌，长鑫技术进展受关注
$SNDK has recently weakened rapidly from the 1600 level, dropping more than 9% over two days, with a low of 1444.
Many people are puzzled:
Changxin's breakthrough this time is in *DRAM technology, while SanDisk mainly focuses on **NAND**, so there is no direct competition. Why is it also being heavily hit?
The core logic is simple:
The market is speculating **not on current supply and demand, but on long-term expectations**.
The breakthrough in domestic storage technology directly breaks the belief in overseas technology barriers, causing capital to directly slash valuations across the entire storage sector, regardless of the track, collectively avoiding risk and fleeing.
Additionally, there are two implicit negative factors:
1. Nanya Technology's performance is explosive, with gross margins off the charts, but the market has priced in **2027–2028 capacity oversupply** in advance, and the high-profit cycle is about to peak.
2. The Korean manufacturers' trillion-level buyback is nearing completion, and the market's largest bottom-supporting funds are about to be cut off, causing sector sentiment to weaken completely.
Personal position management:
Short-term remains bearish, but after continuous overselling, a technical oversold rebound may occur at any time. **Absolutely do not catch a falling knife**.
My trading approach:
No bottom-fishing, no left-side speculation.
Wait for the rebound to face resistance before looking for a second short opportunity; if no position, continue to observe.
The focus going forward is only on two key points:
ASML and TSMC earnings sentiment, with **the real direction set by Samsung's earnings on October 29**.
If Samsung still expects long-term tight supply, this panic is just an emotional trap;
If supply and demand expectations weaken, this storage cycle is basically declared over. $CORE https://scan.coredao.org/address/0x307328448396c925bbc671944285547619e21838#internalTx The project team's contract just unlocked over fifty million, watch out for selling pressure. Additionally, it's worth noting that the circulating supply actually decreased after unlocking, which is suspicious. Judge for yourself.TRX is relatively resistant to decline today, with small intraday fluctuations, showing certain defensive characteristics when the overall market is weak. The underlying logic of TRON remains stablecoin transfers, on-chain payments, and continuously active network usage, and this real demand provides relatively stable support for TRX. However, short-term funds prefer high-volatility hotspots, and TRX usually does not surge the fastest in sentiment-driven markets. Next, pay attention to on-chain stablecoin liquidity, fee burning, and ecosystem activity. If the market stabilizes, TRX's steady recovery is more worth observing. $TRXGold is also a relatively good liquidity channel for capital "hedging," with room for growth in the medium to long term.
However, Bank of America predicts that gold $XAU could reach around 4800 by 2027.
In the short term, Bank of America believes gold won't hold above 4000 and may dip to around 3750.
In fact, we can see that gold's price has some support around 3950.
But rising Federal Reserve interest rates or increases in long-term U.S. Treasury yields could suppress gold's performance. Central banks like China and India have somewhat reduced their gold purchases, making gold relatively weak.
Physical gold demand is relatively weak, but gold ETFs still see capital inflows. In the short term, there is indeed a need to watch for the risk of gold price declines, but in the medium to long term, gold still holds certain appeal 🤔
@OKX星球 #BTC与ETH现货ETF同步流出，机构需求转弱了吗？
刚看了下资金数据，ETF这波流出有点狠。
10月5日到9日，美国比特币现货ETF净流出约6.79亿美元，结束了连续三周的净流入。以太坊更惨，同期净流出5.42亿美元，创1月以来最大单周流出，而且已经连续九个交易日净流出。两类产品单周合计流出约12.21亿美元。
钱为什么跑？主要是宏观不确定性。美联储9月纪要偏鹰，油价又在高位，CPI马上要出，机构选择先撤出来观望。这不是对加密资产本身的否定，是阶段性的避险调仓。
有意思的是，BTC价格一度反弹，跟ETF资金流向出现了背离。ETH就更明显，资金在走，价格也在跌。ETH的机构需求短期确实在减弱，连续九天的流出不是小信号。
接下来要看CPI。如果通胀数据温和，加息预期降温，资金可能重新回流，BTC有机会反弹测试84500。如果数据偏热，加息预期重新升温，资金还会继续撤，BTC下方81500到82000是短期支撑，破了看80000。ETH跌破2400的话，下一档看2350。
操作上别急着抄底。等CPI落地，看资金流向和价格反应再动手。$BTC $ETH A $481 million buyback quota, not a single share bought, directly canceled.
Metaplanet's move, frankly, is to take all the money prepared for supporting the price and use it to accumulate $BTC.
They also made two acquisitions casually. The reason is straightforward: holding significant undisclosed information, making buybacks inconvenient.
As someone who has held for a long time, I actually feel quite tired seeing this kind of announcement.
It's not bad news, but it's not a booster either.
It clearly states a priority: Bitcoin comes first, shareholder returns come later.
A new quota of 385 million is set, but setting it doesn't mean it will be used.
The previous one was lost just like that.
What should be watched most now is not how big the quota is, but whether the next holding announcement shows continued accumulation of coins.
#BTC与ETH现货ETF同步流出，机构需求转弱了吗？
#Strategy增持放缓，企业财库策略生变？ #代币化资产加速扩张，规模化落地还有多远？ $BTC "ETF is bleeding, but Meme is returning to Solana"
Yesterday, Bitcoin spot ETFs saw a net outflow of $65.4 million, and Ethereum ETFs experienced a nearly $577 million outflow over the past 5 days, marking the largest scale since January. However, Solana's Meme coin spot DEX trading volume share quietly surged to 91%, the highest in nearly 14 weeks.
What happened? Mainstream asset funds are withdrawing, but Meme trading volume is concentrating on Solana. It's not that Meme overall is heating up, but the answer to "where Meme plays" has changed.
Why is this worth attention? ETF outflows reflect institutional behavior, while Meme share reflects retail behavior. The opposite directions indicate retail investors haven't exited, they're just changing battlegrounds. Solana has absorbed almost all Meme liquidity, while Meme ecosystems on other chains are rapidly marginalizing.
Which observation is more supported currently? More support for "structural concentration under stock game" rather than a full retreat. Money is flowing from large-cap assets to high-volatility targets, and Solana is currently the only receiving end.
What to watch next? Whether this 91% is a short-term pulse or a trend turning point depends on whether Solana Meme DEX volume can be maintained. If the share stays above 90% for more than a week, Meme on other chains may face liquidity drain. Do you think this round of concentration is risk aversion or the last spree of speculative funds? $ETH $SOL Why are semiconductor chips and memory stocks weakening one after another $SNDK $SKHYNIX $MU
Yesterday shorting US stocks was really satisfying, but why shorting is not well understood by everyone. Today, Xuan will break it down and analyze. There are two main reasons:
First, domestic memory manufacturer Changxin Memory has made progress in the new generation DRAM cell structure and plans to launch related server memory products by the end of the year. The competition for market share is increasing, which will significantly impact the status and profit margins of various chips.
Second, buying in memory chips is starting to decline. Without new capital inflows and only buybacks remaining, it proves that institutions and major players are no longer buying. AI memory is beginning to decouple from the market trend. Therefore, placing short orders at this time has great cost-performance potential!!
Personal trading advice:
Short SanDisk on a rebound near 1485, target 1410
Short Hynix on a rebound between 1120-1130, target 1070Numb from the drop, is there really no hope for CORE? I'll lay out the most painful data and the strongest trump cards
Opened the square today, and all I hear is wailing.
CORE has dropped to around 0.019, down another 2.7% in 24 hours. The haters are on schedule—"zeroing out," "delisting," "running away." A brother messaged me privately: "I can't hold on anymore, is there really no hope?"
I’m not giving trading advice or sugarcoating. Today, I’ll put all the harsh realities and strongest trump cards of CORE on the table for you to see.
1. Why is it numb from the drop? Three reasons, none avoided
First, liquidity is continuously shrinking. KuCoin delisted CORE in July, CoinEx followed in September, and multiple exchanges have suspended deposits and withdrawals for risk control. The number of supporting platforms for CORE has sharply dropped from the initial 34 to only 13 still maintaining trading. The smaller the market, the easier it is for a single sell order to create a deep pit.
Second, 69 million ghost tokens hang overhead. On August 31, a validator reward loophole was verified; before the hard fork, 69 million excess CORE tokens were transferred to external wallets. The hard fork can only upgrade forward and cannot roll back confirmed transactions. These tokens cost almost nothing, and the official side has yet to provide any on-chain verifiable locking or destruction plan. Every rebound has a knife hanging over it.
Third, the flagship product SatPay has been delayed. This highly anticipated new Bitcoin bank was originally planned to launch within 2026 but has been postponed due to licensing and regulatory issues. Over 20,000 users are in the queue, and the launch date is undecided. The entire "revenue era" logic for CORE in 2026—buying back CORE with ecosystem income—heavily depends on SatPay generating real cash flow. Without the product launch, buybacks are just empty talk.
2. But next, some things the haters won’t tell you
Trump card one: CORE is still the largest Bitcoin sidechain by TVL in the BTCFi sector. According to Spark Money’s BTCFi research report, CORE has over $314 million locked on-chain, with more than 5,500 BTC staked, accounting for about 26% of all Bitcoin sidechain TVL. Why would a "zeroed out" project have so much real money locked inside?
Trump card two: The hard fork is indeed working. After the v1.0.26 hard fork activation on September 3, over 150 million excess CORE tokens were destroyed, permanently removed from total supply. Staking rewards returned to normal within 48 hours. The chain didn’t break; blocks are still being produced. This is the baseline.
Trump card three: 76% of Bitcoin’s hash power protects this chain. CORE’s underlying security directly inherits the hash power barrier of the Bitcoin mainnet. Why would a project repeatedly declared dead still have so much hash power working for it?
3. So, is there hope or not?
I won’t lie to you. CORE won’t take off tomorrow, nor will it zero out tomorrow. It’s stuck in a very uncomfortable middle ground—the underlying chain is alive, but the upper-level business loop hasn’t been completed.
Three key data points truly decide CORE’s fate; just keep an eye on these:
When SatPay truly launches and generates sustainable income. This is the only fuel for the buyback narrative.
Whether there is a disposal plan for the 69 million ghost tokens. If the official side proposes a community vote to destroy these tokens someday, that would be the first step toward trust repair.
When ecosystem fee income can cover inflation. Currently, real on-chain income is far from supporting buyback promises, which is the fundamental reason institutional funds are still watching.
Before these three questions have answers, treat all price ups and downs as emotional fluctuations, not trend reversals to chase.
Finally, a heartfelt word
From $6.47 down to $0.019, a 99.7% drop. Who can hold on until now without a steel head? If you didn’t run after a 99% drop, are you really dumb to sell now just because it dropped to 0.019?
I’m not telling you to hold to death. I want to say—if your original logic for buying CORE was "BTCFi is the future," has that logic changed today? It’s still the top Bitcoin sidechain by TVL, 76% of hash power still protects it, the chain is still running, and the code is still updating.
The logic hasn’t changed; what changed is the price. And price has never been the only standard to judge whether a project is dead or alive.
When you can’t hold on, look less at the K-line and more at the chain. The chain isn’t dead, so why panic?
Are you still here? During this drop, did you add to your position or sell?
$CORE