Orbit Post Sitemap
The entire market is moving sideways, but the three platform tokens have quietly outperformed BTC; this is what defensive assets look like.
The old script for platform tokens: they resist declines when the market falls, but lag behind when the market rises.
$BNB 740, +1.4%~+2.3%, market cap 97.7 billion, intraday range 737–745, trading volume surged 28.5% in one day, the most active among the three.
$CRO 0.0557, -0.69%, market cap 2.83 billion, over 1.5 billion tokens burned cumulatively. It fell less, but looking at a year-long perspective, it still dropped over 90%.
$LEO 9.05, +0.94%, market cap 8.33 billion. The most resilient today, relying on profit buybacks, not new stories.
Platform tokens can help you lose less, but rarely help you earn more; they are parking lots, not engines—don’t stay parked too long and forget to move.$ETH structure is bearish, with 1-hour and 4-hour charts both trending down synchronously. In the past hour, selling pressure clearly dominates, with active trades mainly on the sell side. Key levels are directly presented: the upper resistance at 2387.85 is the Point of Control (POC) in the dense trading area, followed by 2415.5 above; the lower support wall is at 2332.45, which is also the lower boundary of the value area. Orders are executed without hesitation, market price at 2373.4 directly enters short positions, and the current price is just below the POC, a sufficient position. Stop loss at 2390.5, placed above the POC and the midline; if the price reclaims the dense cost area, this downward structure will be invalidated and the short logic will be void. The target is 2332.4, precisely hitting the support wall and the lower boundary of the value area, which is a natural position for bears to take profits, with a risk-reward ratio of about 2.4; long positions still account for about 68%, and the long liquidation fuel below is much greater than above, indicating a tendency for the price to be pulled downward. Risk note: the price may first sweep liquidity around 2387-2390 before falling back, with a short-term possibility of stop-loss hunting. A sudden strength in BTC driving an overall rebound could also disrupt the bearish momentum. #中东原油供应风险再起，油价持续走强 #BTC现货ETF重回流入，ETH资金持续流出 #SK海力士Q3财报在即，HBM4能否提振盈利？
Actually, there's no need to worry too much about Hynix's current numbers; Nvidia is fully utilizing 12-layer HBM3E, keeping the profit base very stable. What really determines whether the market can continue to rise is still the management's stance on HBM4.
After all, in the HBM4 era, the foundation is directly customized by TSMC, which, if successful, will widen the moat and strengthen pricing power, but the R&D and capital expenditures are also real cash outflows. Moreover, Samsung and Micron are constantly pressing from behind; if the HBM4 mass production pace falls short of expectations even slightly, the valuation will have to be discounted again.
In the short term, first ride this wave of earnings catalysts; the key is to watch the HBM4 mass production schedule and customer qualification progress in the earnings call—no rabbit, no shooting the hawk.
For this earnings season, is everyone holding firm or cashing out early?IRYS peaked at 0.32 then fell back to 0.08, I won’t chase this short squeeze market for now.
Noticed: OKX spot around 0.080, up about 182% in 24 hours from around 0.028, but down about 75% from the high of 0.32.
The 4-hour K candle at 0:00 Beijing time on 10/24 hit a high of 0.32 and closed at 0.111, a typical wick followed by a pullback.
CryptoVank cited: roughly surged from 0.025 to 0.296 (+1,078%), about 90% of the $63.5 million liquidations were shorts; heavy trading volume on Korean exchange Upbit.
No single official announcement as catalyst, more like a short squeeze on a thinly traded circulating supply; OKX perpetual open interest nominally about $6.86 million, funding rate nearly flat.
I think this is a short squeeze on extremely thin liquidity, not a fundamental repricing, the pullback will be faster than the rise.
With the October FOMC this weekend, this kind of altcoin spike is even less suitable for holding through.
First watch if it can hold and reclaim 0.10; if it breaks below 0.063 (around today’s Beijing open), this rebound attempt is invalidated, better to stay out.
Do you think there will be a second squeeze, or will the spike fade before the weekend?
$IRYS $ETH $BTC
#OctoberFOMCApproaching What variables remain with the Fed pausing rate hikes? #MiddleEastOilSupplyRisksResurface Oil prices continue to strengthen Xiaomi's stock price has recently been in a "sharp drop → rebound → decline again" volatile pattern, down about 30% since the beginning of the year, with a TTM P/E ratio compressed to 18 times. The core issue is not whether there is long-term value, but when the bottom of phone profits will be confirmed and whether car production capacity can accelerate.
The phone segment is heavily squeezed by storage costs: Q2 smartphone gross margin was only 8.5%, with institutions generally expecting profits to bottom out in Q3 and rebound in Q4. The automotive segment broke 40,000 deliveries for the first time in September, but the factory's full capacity is about 34,000 units/month. The annual target of 550,000 units was only 52% completed in the first nine months, with the actual final figure likely between 450,000 and 500,000.
There is significant institutional divergence: the average target price is HKD 34.7, but some see 38.7, while others only give 27.6. Fortunately, buybacks have exceeded HKD 8 billion this year, providing support at low levels.
In short: the valuation is not expensive, but before the financial report confirms the profit bottom, it is likely to remain in a wide range of volatility rather than a one-sided recovery.
$XIAOMI A strange phenomenon: this time, the liquidity drain triggered by the stock market decline seems to have hit gold the hardest, while Bitcoin barely dropped. According to the usual pricing logic, during a liquidity crisis, Bitcoin should be sold off more than gold.
The same story theme, but the market's pricing logic is completely opposite. This financial market is really unpredictable.
Looking at the sharp drop in gold these past two days, it all feels very strange and eerie.This is the situation with U.S. Treasury bonds. Usually, the narrative of a U.S. debt crisis is rampant, but when a real crisis occurs, comparing with other countries and other assets, U.S. Treasuries still prove to be the safest.
This round of declining U.S. Treasury yields clearly illustrates the point. If AI faces problems, buy U.S. Treasuries. If Europe faces problems, buy U.S. Treasuries.
In fact, this is all a result of peer comparison. Otherwise, if not buying U.S. Treasuries, would you buy European debt? Japanese debt? Or at best, Russian, Brazilian, or Indian debt? Are those reliable?
Even gold and Bitcoin, the so-called safe havens, don’t quite measure up. Their safe-haven quality shows as a price spike during crises, but in the long term, liquidity plays a much more decisive role.
The main problem with the renminbi is that you can’t earn renminbi easily, and secondly, there is limited supply of renminbi assets to the outside world. Domestic subscription for government bonds is several times oversubscribed, leaving no room for foreigners to buy.
Additionally, the financial market is not open enough. Holding renminbi and renminbi assets is definitely less convenient than holding U.S. dollars and dollar assets, with fewer options for converting between various assets. Our market still needs to strike a good balance between safety and openness.
Furthermore, even with more openness, the depth, breadth, and resilience of the stock market are still weaker than those of the U.S., which is a financial powerhouse. Ultimately, it comes down to the underlying assets: the MAG7 companies make money worldwide and are global monopolistic companies.
Of course, there is also the issue of interest rate differentials. No matter how many crisis narratives surround the dollar, a bit more interest can still convince investors.SanDisk’s $10.3B-$10.8B quarterly revenue guide makes the Oct. 29 report a useful read on whether AI data-center demand is broadening into storage spending. Rising NAND prices can lift revenue, but they also make the quality of growth harder to judge: investors will want evidence that profit gains can hold if pricing momentum cools.
#SanDiskQ1StorageGrowth In the futures market, a large short position with high floating profit usually indicates that the market has moved in line with the trader's judgment for a while, but this does not necessarily mean that the price will continue to fall afterward. Experienced traders, when seeing such news, first look at the positions and funding costs, then check if there are new supply and demand changes in the price. Focusing solely on how much others have earned can easily lead to mistaking the result for the cause.
According to Lookonchain monitoring this time, $HYPE has dropped about 22% from its previous high; the largest short address on-chain, 0xf02d, holds 715,334 HYPE short positions, valued at about $54.57 million, with a current floating profit of about $5.56 million, and has additionally earned about $620,000 through accumulated funding rates, totaling approximately $6.18 million.
My judgment is that the short-term bearish sentiment has indeed been amplified, but this news itself is more like a reminder of the position structure and is insufficient alone to prove that the price will continue to fall. In traditional futures, large short profits often attract profit-taking, and closing positions may bring reverse volatility; the difference in the crypto market is that on-chain addresses and funding rates are all transparent, making it easy for ordinary users to be emotionally influenced by "whale floating profits." If short concentration continues to rise, once the price reverses, the squeeze could come quickly.
What I care more about is whether shorts will continue to add positions afterward, and whether funding rates and position changes align. Are there friends who have also transitioned from traditional markets? How do you view this kind of large on-chain public positions in the crypto market compared to position reports in commodity futures? Which is more likely to influence sentiment?If we treat this 24-hour period as a small-scale repricing, then BTC's weakness is not just a "drop," but a re-voting of funds. Guess why ETH was caught counter-trend at the same time? I stared at this set of numbers for a while: BTC futures net outflow of $200 million, spot net outflow of $131 million, totaling $331 million, the weakest among the three assets; ETH futures outflow about $34 million, but spot net inflow about $43.43 million, totaling a slight net inflow of $9.42 million; SOL futures inflow of $9.76 million, spot outflow of $9.24 million, nearly balanced overall. On the surface, it's a matter of price rise and fall, but at the core, it looks more like derivative positions deleveraging, while on the spot side, someone is quietly changing seats. First, looking at the bullish path: ETH shows a "futures position reduction + spot support" combination, which usually means part of the leverage bubble has been squeezed out, and the selling pressure is not from spot panic but from the contract side actively reducing risk. If this structure continues, ETH's relative strength will manifest before price, especially when BTC's contract outflow slows, risk appetite may shift from "defending BTC" to "picking up ETH." SOL's near-zero net amount also indicates it is temporarily not a cash cow; altcoin sentiment hasn't completely collapsed, just no one is willing to light the torch first. But the risk signals are also very clear: BTC spot and futures both have net outflows, indicating this is not just a single contract squeeze, but spot holders are also reducing exposure. In this case, any rebound is likely to become short covering rather than new money entering, so the sustainability of chasing highs is questionable. ETH's spot inflow absolute value is still not large,$ZEC
The ZEC NU7 upgrade is scheduled around November 5, at block height 3543405. I've marked this number on my calendar for several days. According to the usual pattern, there should be a rush before the upgrade, but instead, it has been steadily declining.
Actual data: OKX current price 1165, 24-hour high 1258, low 1158; Binance price 1164, down 4.5%. From 1258 to now, seven percent is gone.
Perpetual positions dropped just over 12% in 24 hours (according to coinotag), on OKX I counted about 108,000 coins, worth $126 million. Leverage is retreating—not a panic sell-off, but a gradual withdrawal.
A couple of days ago, ETFs were still seeing net inflows; money was coming in during the drop, I acknowledge that. But the money coming in is from monthly allocation funds, which can't push short-term moves.
On the actual day NU7 goes live, it will likely be another case of good news already priced in. I said last time that if 1100 doesn't hold, I accept it; now it's only about sixty dollars away from that, so that still stands.
#10月FOMC临近，美联储暂停加息还有哪些变数？ Approach to identifying the end of a market trend.
$BTC is oscillating at a high level, $ETH is unable to rise, and $SHIB is still pulsing repeatedly.
The main targets are weak, memes are exploding one after another, liquidity is gradually tightening, position control is necessary. 💰 Lost 20,000U down to 5,000U, he wants to flip IRYS 30x: The truth is this coin is down 74.63% from today's highest point
Orbit hot post 25970 views: The author lost over 20,000U from the peak, with only over 6,300 left in the account, finally taking 5,000U to announce "starting over, target 150,000U."
The same coin, another post with 17,365 views only says: There were 7 explosions tonight in total.
Market: IRYS current price 0.08118, surged from 0.0326 to 0.32 in 24h, up 149%, but has retraced 74.63% from the high — those chasing the top have lost on average three-quarters.
This is not a tutorial. Turning 5,000U into 150,000U is 30x, IRYS needs to rise to 2.44, which is 7.6 times today's high. Betting 7.6x on a coin that has plunged 74% is not a strategy, it's buying liquidation.
Those who made money won't say "starting over" in posts.
Key levels: BTC 80765 breaking 80513 looks to 78K; ETH 2375 is weaker, reduce positions on rebound.
📊 Investment advice
• BTC — Bearish
Take Profit (TP): $74,300
Stop Loss (SL): $87,200
• ETH — Bearish
Take Profit (TP): $2,090
Stop Loss (SL): $2,660
⚠️ Disclaimer: The above is personal opinion only and does not constitute investment advice. The market is volatile, please DYOR, risk at your own discretion.[Holding $2.96 billion in $HYPE, PURR is buying back $100 million worth of shares, let's rise quickly!]
This time, PURR has directly increased its stock buyback limit from $30 million to $100 million, with about $72 million remaining, and extended the deadline to October 2027. It still holds 38.8 million HYPE tokens, valued at approximately $2.96 billion according to disclosures.
PURR is buying back its own shares, not HYPE tokens. The company is betting on the Hyperliquid ecosystem and aims to reduce circulating shares through buybacks, potentially increasing the HYPE asset exposure per share. However, this limit is only an authorization; management can choose when to buy or not buy, so it cannot be considered that $100 million has already been invested.
The market's attitude is also worth pondering: PURR's stock price once dropped over 30% from the October high, and the company increasing buybacks indicates it wants to support the stock price under pressure. Holding a large amount of HYPE does not necessarily mean the stock is cheap; the stock price is also affected by dilution, operating costs, and market discounts on assets.
The key is to watch the actual buyback volume and whether the discount of PURR's stock relative to its net HYPE asset holdings narrows. If the buyback truly materializes and the asset exposure per share continues to increase, this move will be effective; just increasing the limit without spending money, no matter how loudly confidence is proclaimed, is only a paper gain.$XAU Trump raises the possibility of military action against Iran again! But what might truly impact BTC is not war, but inflation.
On October 23, Reuters reported that Trump stated the US might take military action against Iran again.
Core market conflict: With the midterm elections approaching, Trump needs low oil prices, but military escalation could push energy costs higher.
① Crude oil: Supply risk premium rises $BZ
Focus on the Strait of Hormuz. If shipping is obstructed, the short-term geopolitical premium could turn into an actual supply shock.
② Gold: A battle between safe haven demand and real interest rates
Escalation of conflict stimulates safe haven demand, but rising oil prices may increase inflation expectations and delay rate cuts, while rising real interest rates would suppress gold.
③ BTC: Beware of tightening liquidity
Rising oil prices → inflation pressure → delayed rate cut expectations → possible strengthening of the dollar and US Treasury yields.
Once this transmission chain forms, $BTC may face dual pressures of declining risk appetite and deleveraging.
④ Trump's political game
With the November 3 midterm elections approaching, sustained high oil prices could increase voters' cost of living pressure. There is a clear contradiction between military deterrence and controlling energy prices.
Key point: War news affects sentiment, energy prices affect inflation, inflation expectations affect liquidity, and liquidity is the key variable BTC needs to watch.
Do you think Trump's move this time is a prelude to military action or a bargaining chip?
#10月FOMC临近，美联储暂停加息还有哪些变数？ Calculating, BTC single-day inflow to exchanges is about 2.6 billion USD, which is the largest scale since April. This figure alone is enough to indicate that the chips available for trading, transferring, or cashing out in the market have suddenly increased significantly; combined with the last similar peak after which BTC prices topped within weeks, I would regard this as a bearish risk signal, but it cannot be directly equated to an immediate drop.
① 2.6 billion USD: The single-day inflow amount is very concentrated, indicating this is not ordinary transfers from scattered addresses, but at least reflects a group of holders moving chips to exchanges.
② Largest since April: This is not an isolated small fluctuation, but an extreme value covering a period of time. From another quantitative perspective, the potential selling pressure the market needs to digest is at a high level during this period.
③ Price topped weeks after the last similar scale: The historical correlation provides a warning, but it describes the result in the following weeks and does not mean this time will necessarily repeat. The data is more suitable to remind me to lower optimistic interpretations of short-term sentiment.
My judgment is that this news has a bearish bias on sentiment, especially making short-term funds pay more attention to whether the inflow continues to expand after it occurs. If subsequent inflows continue to increase, the pressure signal will be strengthened; if it is just a one-time peak, the impact may be overestimated. When looking at this kind of information, do you usually focus more on the exchange inflow amount, duration, or price reaction?$BTC
About $2.6 billion worth of BTC flowed into exchanges within a day, the largest amount since April. Some in the group are already calling it a sign of a dump, but I'll hold off on that conclusion for now.
The price I pulled: OKX current price 80765, 24-hour high 82060, low 80514; Binance low at 80461, both align, daily drop just over 1%. If it were really a dump, this drop is not enough to notice.
Perpetual fee rate is less than 0.05%, with 30,600 BTC positions, about $2.47 billion, no reduction seen. Leverage funds haven't fled, nor are they rushing to add.
At the same time, Bitdeer sold another 283 BTC this week, clearing their positions—miners are selling, ETFs are flowing out, yet exchanges still have $2.6 billion inflow. Sellers have plenty of reasons, but the price is stuck above 80500.
In previous rounds, I kept saying this is consolidation, not a crash, to the point of annoying myself. But it just won't crash, and I have no solution.
#10月FOMC临近，美联储暂停加息还有哪些变数？ Brothers, Goldman Sachs just dropped a very painful piece of news: last week institutions set a historic record with massive sell-offs of tech stocks.
Not long ago, the market strategy was simple: blindly go long on AI and ignore all other sectors. That strategy has now completely collapsed. Investors no longer buy into AI hype stories; they want real money and need to see whether the huge investments from these big companies can actually turn into real revenue.
Everyone says AI will pour $1.1 trillion in next year, but just burning money isn’t enough. Cloud providers must prove that their orders can be monetized and cash flow can improve. Plus, there are two practical problems ahead: a big power shortage next year that will choke computing expansion; and many cloud companies have high debt with borrowing costs rising.
An even bigger issue is the market’s current dilemma: on one hand, worries about insufficient computing power; on the other, fears that after frantic capacity expansion, there will be direct overcapacity. Coupled with huge AI investments and rising borrowing costs driven by fiscal deficits worldwide, the market is becoming more cautious and unwilling to blindly chase highs.
In short: the AI market has moved from a phase of speculation on expectations to a test period where performance returns are mandatory. Going forward, AI sector volatility will increase, the risk of hitting pitfalls is clearly rising, so don’t blindly rush in anymore. $SNDK $SKHYNIX $MU
#AI需求支撑存储市场，芯片股行情反复 #SK海力士Q3财报在即，HBM4能否提振盈利？
Q3 profit surge mainly relies on long-term contracts for HBM3E + price increases for regular DRAM, HBM4 is still in the early ramp-up phase, contributing limitedly to this quarter's profit, the real push will be seen later.
SK Hynix started mass shipments of HBM4 in Q2, but Q3 is still the expansion and ramp-up stage, shipment volume is not large yet. The main driver of this quarter's performance remains the high-margin HBM3E, combined with continuous price increases in server DRAM contracts and spot prices. Institutions generally expect Q3 revenue to approach 98–101 trillion KRW, with operating profit likely to reach 77–81 trillion KRW, a clear quarter-on-quarter increase.
$SKHYNIX
The value of HBM4 is not in this quarter but in the medium to long term. It corresponds to Nvidia's next-generation Rubin platform, with stronger performance, and most of the capacity has already been locked in long-term contracts by top customers, making it a high premium product. When Q4 and next year come, as HBM4's share increases, it will become the new profit driver. $SNDK
Risk points to watch: Samsung is also aggressively pushing HBM4, and after supply increases next year, the overall HBM premium may be compressed; additionally, if AI capital expenditure slows down, the pace of long-term contract orders will also be under pressure.
Therefore, the focus of this earnings report is not on how much money HBM4 made, but on the yield and Q4 capacity guidance given by management, which is the key to determining future valuation.🏖️ This afternoon, three things that are currently lukewarm
$ASTER 0.709, the biggest problem with contract DEX now 🚀
Technically, perpetual contracts can be done, but users are all on established major exchanges, and DEX lacks depth and has large slippage. 0.709 has position but no volume; it needs regulation or a major event to push users onto the chain, otherwise it's hard to take market share from CEX.
$XAUT 4100, why gold is stagnant over the weekend 🥇
XAUT is pegged to spot gold. Before next Tuesday's interest rate decision, rate hike expectations are suppressing it, while rate cut expectations are supporting it, with no conclusion on either side. The weekend is just sideways; any real movement will be on the day of the Tuesday announcement. It's meaningless to bet on direction early.
$BICO 0.0164, why the infrastructure coin isn't rising now 📈
The direction is correct; abstract accounts are indeed the necessary path for mass adoption on-chain, but the market now only pays for "what money can be made immediately," not for "what might be useful in the future." 0.0164 is just moving with the market. It will have a chance when on-chain applications truly attract new users.
This article is for market analysis only and does not constitute investment advice.
#10月FOMC临近，美联储暂停加息还有哪些变数？ #October FOMC Approaches, What Are the Remaining Variables for the Fed's Pause on Rate Hikes? $BTC $ETH $IRYS had a pretty surreal night. Before midnight, it was stuck at 0.06, then a few hours later someone saw it hit 0.5, and when you opened your eyes again it had dropped back to the starting point—all within just three or four hours. Short positions were liquidated for $59 million, while long positions only saw $8 million, with trading volume reaching tens of billions. A bunch of people claim they had already positioned themselves early, but the order book was really full of retail traders—why would the big players dare to pump it like that? If they dared, it means they were sure there was no one below. Every time it pumps, a crowd claiming to have caught the big bull appears; it’s really exhausting to watch.#SK海力士Q3财报在即，HBM4能否提振盈利？ In the short term, Q3 HBM4 contribution is limited; customer certification and chip schedule are delayed, shipments fall short of earlier expectations, and the main profit driver this quarter remains HBM3E. In the long term, HBM4 is the core growth point with higher premiums, but risks include Samsung catching up and squeezing market share, as well as large upfront investments.
During earnings season, I will focus on HBM4 yield rates and shipment guidance, not betting early, waiting for data to materialize before judging opportunities, and ensuring risk control.HSBC has issued its own coin, and it's a Hong Kong dollar stablecoin 👀
On October 23, HSBC's HSBC RedCoin (code RCHKD) launched on PayMe. HSBC's first stablecoin, holding a Hong Kong Monetary Authority issuer license, pegged 1:1 to the Hong Kong dollar.
A few numbers I find quite key:
PayMe has over 3.3 million users, and in the next few days, wallets can be opened directly in the app; users can transfer directly to friends and also pay at over 100,000 merchants; later, the HSBC HK App will also integrate it for subscribing to tokenized wealth management products.
Why I think this is bigger than it looks:
Previously, stablecoins were mostly played within the crypto circle, but now a central bank is directly embedding it into a payment app used daily by millions, bringing on-chain money this close to ordinary people's phones for the first time.
As for the market, don't expect it to pump prices in the short term; Bitcoin is still hovering just above 80,000 today, and ETFs are still seeing outflows. But in the long run, the more compliant stablecoins there are, the smoother the path for off-chain money to enter the market, and RWA and payment sectors will continue to benefit.
My view: Traditional banks entering the scene is not about stealing jobs but about "stamping" this industry. Will you use stablecoins issued by banks?
Sources: HSBC official announcement, South China Morning Post, The Block
Personal opinion only, not investment advice, DYOR
#stablecoin #HongKong #RWAOne of the easiest mistakes to make in the privacy sector is mistaking a rebound for genuine support. The truth is that derivatives heat up first, while spot markets may not necessarily follow. The strength you see— is it new money entering, or leverage propping up the scene? I've been watching NEAR, WLD, and ZEC, three privacy narrative tokens, and the more I look, the more I feel there's a gap between surface-level hype and real support. WLD bounced about 29% a few days ago, roughly between 0.455 and 0.59, but its recent decline has been slow, with today's low only down to 0.5138. This kind of "resistance to falling" can easily be read as strength, but if I enter now, my stop loss would have to be below 0.45, which is a risk I can't bear, so I skip it. ZEC is more subtle. A stop loss at 1187 seems reasonable, and I've repeatedly questioned my trading logic for flaws, concluding there aren't major issues. However, at this level, I don't have confidence in capturing a larger profit margin, so I choose not to open a position. It's not that the token is bad, but the risk-reward ratio doesn't give me peace of mind. What really made me act is NEAR. Its range is 4.54 to 5.5, with a recent low near 4.7. I quickly sold in batches on the 15-minute chart to build a position, not a large one, as the pattern has emerged, so I added a bit more. My current cost is 4.8, and based on the range, as long as the cost is below 4.82, it's still acceptable, with a relatively comfortable risk-reward ratio. The second layer of transmission here is: if the privacy sector's rebound is mainly driven by derivatives, sentiment will warm up before the spot market, and the risk appetite for altcoins will be temporarily ignited.Same downward sound, diverging mindsets
In the past 24 hours, the market fell together, but sentiment took two different paths. BTC bullish sentiment dropped to only 0.82, falling into extreme fear territory; ETH rose to 2.05, heating up. One feels like waiting in an emergency room line, the other like grabbing the mic at a celebration party.
Liquidation data is even more straightforward: total network 498 million, longs 340 million; BTC longs 81.33 million, ETH longs 50.55 million. Bulls were all shaken out, but holders reacted differently: BTC holders are silent, ETH holders are still waiting for 3000, continuing to shout "hold on" in the group.
The rift comes from pricing logic. BTC is treated as a macro asset: no Fed rate cuts, ETF outflows, expectations cooling. ETH is treated as a narrative asset: staking, ETFs, Layer2, always making people believe "this time is different."
But the market doesn’t follow sentiment. BTC’s fear and ETH’s greed, at least one side will be wrong. Which side are you on? 1: ETH catches down, BTC true bottom; 2: BTC catches up, ETH leads.
On-chain data compiled, not trading advice.
$BTC $ETH
#10月FOMC临近，美联储暂停加息还有哪些变数？
#BTC现货ETF资金反复，市场买盘能否企稳？
#中东原油供应风险再起，油价持续走强 #AI需求支撑存储市场，芯片股行情反复
Storage chip stocks previously experienced a significant pullback, mainly influenced by the following factors:
• Overall tech/AI sector valuation digestion and concerns about the interest rate environment.
• Doubts about the sustainability of AI capital expenditures.
• Supply tightness supports prices but also raises concerns about a cycle reversal after future capacity releases.
The rebound on October 23 was supported by multiple factors:
• The halt of Solidigm's IPO eased shareholder dilution concerns.
• A broader tech stock rally.
• Fundamentals still point to strong AI storage demand.
The current core market focus, as shown in the screenshot, is:
• Whether demand growth can sustainably translate into profit improvement (rather than relying solely on price increases).
• The pace of capital expenditures and future supply expectations.
• Repricing of valuations in a highly volatile environmentThe fundamentals of the storage industry remain strong in 2026. Demand for HBM, enterprise-grade SSDs, and high-capacity HDDs continues to be driven by AI training, inference, RAG, and model checkpoints.
Leading manufacturers like Micron, SK Hynix, and Samsung have their production capacity basically sold out through 2027, with long-term contracts locking in volume and price, and breach costs being extremely high. Hard drive manufacturers such as Seagate and Western Digital have also clearly stated that nearline storage supply is insufficient to meet demand, with some products already allocated through 2027. Data from institutions like TrendForce shows that although contract prices for traditional DRAM and NAND have retreated from the ultra-high month-on-month increases seen at the beginning of the year, they are still expected to rise 10%-20% quarter-on-quarter in Q4, with absolute increases remaining at historically high levels.
The market's concern that "price increase slowdown = cycle peak" is more a base effect and the apparent deceleration caused by long-term contract price ceilings, rather than a true weakening of demand. Hyperscale cloud providers and emerging GPU clouds are still expanding massively, with storage being one of the bottlenecks in AI infrastructure.
The market fluctuations mainly stem from valuation digestion and macro sentiment. After significant gains, storage stocks are highly sensitive to risks such as interest rates, the sustainability of AI capital expenditures, and competition from China, leading to amplified short-term volatility. However, as long as AI server shipments and data center construction do not experience systemic contraction,
what deserves more attention currently is: HBM iterating towards higher bandwidth and capacity, enterprise-grade NAND penetration in AI inference, and the cost advantages of hard drives in tiered hot and cold data storage. #AI需求支撑存储市场，芯片股行情反复 The current macro environment of the US stock market is indeed under the dual pressure cycle of high costs and high interest rates.
Trump's repeated statements between war and peace have created high uncertainty in geopolitical risks.
Brent crude oil standing at $116 directly pushes up inflation expectations and corporate operating costs.
Gold and US stocks are falling simultaneously, reflecting that the market is experiencing a liquidity squeeze.
At the same time, AI tech stocks are facing the combined test of performance realization and rising costs.
Next, do not blindly guess when this round of adjustment will end; focus on two variables: first, whether the Middle East situation substantially de-escalates leading to a drop in oil prices, and second, the earnings guidance of core AI companies.
Before the macro pressure is lifted, maintain a neutral to low position, watch more and act less. SpaceX's performance is very good, but the stock price does not necessarily move solely based on profits and merger rumors. When early shareholders can sell also depends on other factors.
The recent phased unlocking has allowed some previously restricted shares to enter the tradable range. Unlocking does not mean these holders will immediately sell all their shares, but the potential supply the market faces has indeed changed.
This is why I don't simply interpret the temporary pressure on the stock price as everyone rejecting the financial report. Some early investors, even if they remain optimistic about the company, may need to raise funds or adjust their portfolios.
Conversely, the buying interest brought by merger speculation may not last forever. If the official plan stalls, short-term trading funds may have much less patience than long-term shareholders.
Currently, this stock is influenced simultaneously by operational performance, trading expectations, and changes in share circulation. Explaining all price fluctuations by just one factor is overly simplistic.
I am interested in SpaceX's business progress but will consider the company's operations and market supply and demand separately. The former determines how much long-term value can be created, while the latter may significantly affect the price during certain periods.
Especially when the financial report has just been released and merger discussions are heated, people tend to feel that with so many good news, the stock price should rise. However, as more people become able to sell the stock, the market still needs time to find a new transaction price. This process is not necessarily comfortable.
#SpaceX业绩大增，与特斯拉合并能否推进？ Chip stocks have recently started to "fluctuate repeatedly" again, especially in the memory sector. On one hand, AI demand has almost sold out HBM, DRAM, and NAND; on the other hand, stock prices overreact to any negative news, a typical case of "strong fundamentals, weak sentiment."
After Micron's earnings report, the guidance was strong, with data center SSD revenue surging; SK Hynix and Samsung posted record profits; Seagate, Sandisk, and others also clearly face tight capacity. Long-term agreements have locked volume and prices tightly through 2027, theoretically providing high visibility. But the market just doesn't buy it—the slowdown in price increases quarter-over-quarter is interpreted as a cycle peak, and concerns about interest rates, financing costs, and the sustainability of AI spending have caused the sector to pull back sharply from highs, pushing valuations down to historically low ranges.
This kind of fluctuation is actually very common. Memory is a strong cyclical industry; the AI supercycle has extended the upward phase but also amplified expectation gaps. In the short term, prices are extremely sensitive to "slowing growth" and "when supply will balance." Once sentiment turns, capital withdraws quickly; conversely, once concerns ease, rebounds are fierce.
A more pragmatic trading view is: don't treat every pullback as a trend reversal, nor chase highs during emotional peaks. Focus on two signals—first, whether leading manufacturers have materially downgraded their order coverage for HBM and data center storage in 2027; second, whether cloud providers' capital expenditure guidance shows clear contraction. So far, neither has given a clear systemic weakening signal. #AI需求支撑存储市场，芯片股行情反复 [6 Billion USD Stockpiling 73 Oil Tankers, Korean Shipowner Wins Big in Hormuz]
Sinokor Shipping Group, behind Korea's Sungaya Shipping, originally operated container and tanker transport, with its subsidiary Changjin Shipping specializing in large crude oil transportation. This year, it spent about 5.98 billion USD to buy 73 second-hand ultra-large oil tankers in one go, a scale almost equal to the combined total of the other eight major buyers worldwide.
This big gamble coincided with the most exciting market conditions. After the outbreak of the Iran war, shipping risks in the Strait of Hormuz soared, and many shipowners dared not take orders, while oil-producing countries were eager to export crude oil $BZ. Sinokor's tankers thus absorbed a large amount of transport demand; in September, they transported 105.6 million barrels of crude oil through the strait, about 37% of the total volume, exceeding the combined transport volume of the three main competitors Saudi Arabia, Kuwait, and the UAE.
While others were busy withdrawing ships, Sungaya Shipping had enough capacity to take orders. However, the near-monopoly is only for transport share during specific periods, and it is far from a guaranteed profit. Several related tankers have already been attacked, with insurance, crew safety, and subsequent freight rates all being costs. The more dangerous the strait, the more attractive the freight; once shipping returns to normal, whether these 73 ships can continue to print money is the real outcome of this big gamble.$SMCI Personally, I think the expected bullish trend has ended, and the current decline is just beginning. This price is not enough to support its slowing performance. It is expected to drop to $23, very soon.October pause: smoke is not fireworks (￣▽￣)~*
The market is hyping a 92.8% chance of a pause, but PMI sales price growth is the fastest since August 2022, and inflation shifting gears is more worrisome. Staying still in October seems like giving data some face, not a free pass for risk assets; the door for December isn’t closed, the Fed is taking one step while looking three steps ahead (¬‿¬).
BTC is consolidating, waiting for a signal. ETF inflows are pulsed, gains can be mostly given back in a day, trading volume is not allocation volume. The overhead resistance zone stacks the yearly opening price and moving averages, without sustained net inflows it can’t be broken; only the cost zone below has real buy orders. I’m not chasing, waiting for the FOMC to release December wording before moving (・∀・).
ETH is BTC’s shadow, and the shadow breaks first. ETF outflows are harder to stop, the ecosystem lacks new narratives, and off-exchange funds aren’t coming in. ETH pretends to be stable when BTC consolidates, but falls fastest when BTC weakens; consolidation ≠ stabilization (´･ω･`).
Don’t just treat gold as a safe haven or a rising asset. It can bounce when US Treasury yields fall, but if high yields persist, the rebound will be short-lived. Support depends on central bank buying and geopolitical lines; if the market seriously prices in that the fiscal side can’t handle high yields, gold will run an independent trend, don’t follow BTC’s rhythm (￣ω￣).
$BTC $ETH $XAU
#10月FOMC临近，美联储暂停加息还有哪些变数？
#中东原油供应风险再起，油价持续走强
#BTC现货ETF资金反复，市场买盘能否企稳？ At the beginning of this year, 1 US dollar could be exchanged for about 6.99 RMB, but now it can only be exchanged for around 6.68 RMB.
This means that if you hold 10,000 US dollars, its value in RMB has decreased by about 3,100 yuan compared to the beginning of the year…#10月FOMC临近，美联储暂停加息还有哪些变数？
The October FOMC (October 27-28) pause in rate hikes is now the mainstream expectation, but it is not without uncertainties; the market focus is shifting from "whether to pause" to inflation assessment, the subsequent path, and whether there will be another hike in December.
• The Federal Reserve will hold the FOMC meeting on October 27-28, with the decision expected to be announced on October 28 (Eastern Time).
• The market generally prices in no change in rates in October (Federal Funds target range 3.75%-4.00%), with CME FedWatch and forecast markets showing a pause probability of about 88%-92%.
• The latest S&P Global Flash PMI: Composite PMI preliminary at 57.6 (still in a strong expansion range, the second highest in recent years), Manufacturing PMI rose to 56.0 (about a 53-month high), and the increase in business selling prices is the fastest since August 2022, indicating persistent inflationary pressure.
• As the October pause becomes consensus, the focus has shifted to the Fed's assessment of inflation, the subsequent interest rate path, and whether there is still a possibility of a rate hike in December. $SPCX Negative News: SpaceX AI Financing Hits a Roadblock, AI Infrastructure Frenzy Faces Harsh Reality
My first reaction upon seeing this news is that the grand narrative of AI has finally hit a hard stop in the credit market.
SpaceX plans to raise $40 billion for AI, even attempting to use an SPV (Special Purpose Vehicle) to move liabilities off the balance sheet to protect its rating—a very clever tactic.
But Wall Street isn’t buying it anymore; the credit market is highly volatile, and borrowing costs are simply not negotiable.
This perfectly aligns with the “massive capital expenditure” risks exposed in Meta and Google’s earnings reports.
Investors are re-evaluating the risks of large-scale AI infrastructure, fearing that all this money might be spent with no returns.
This is directly negative for the crypto market, especially the AI sector.
Just as Musk was still painting dreams of flying cars, the financing of hundreds of billions of dollars has stalled, indicating that in a high-interest macro environment, institutional hot money has dried up, and no one wants to be the sucker to take on a bottomless heavy asset.
BTC and ETH are already constrained by the weakness of US tech stocks, struggling around 81,000 and 2,400 respectively.
Now even the top tech giants are hitting financing walls, which will only deepen the risk-averse sentiment and further drain liquidity in the market.
Without fresh capital inflows, the market can only continue to weaken and grind lower. Hold your hands, don’t try to catch a rebound at this time. The blood oxygen curve on the monitor is still climbing, but the right ventricle has already started to dilate—the first impression SanDisk's earnings report eve gave me is this kind of "numbers look good, but the myocardium may not" compensatory state. Revenue guidance of $10.3 billion to $10.8 billion is like the ejection fraction in a preoperative ultrasound—pretty, but you have to ask: is this the myocardium contracting on its own, or is it pushed by drugs? AI computing power center demand is a load test, NAND price increase is afterload elevation; if growth mainly relies on price rather than bit shipments, that is preload dependence, and if margin improvement comes from inventory revaluation, it’s more like a diuretic-induced pseudo weight loss.
I view XIBM’s linkage as coronary collateral circulation: the main vessel is blocked, collateral opens making people mistakenly think perfusion is still okay, but collateral cannot replace local myocardial biopsy. No matter how sensitive cross-market neural reflexes are, they cannot prove the storage cycle lesion has been removed. What really needs monitoring are gross margin, free cash flow, inventory turnover, and capital expenditure—inventory cycle is arrhythmia, capital expenditure is myocardial hypertrophy, demand front-loading is silent ischemia. If after the earnings report prices fall but bit shipments don’t pick up, that is reperfusion injury.
The monitor can lie, but the myocardial enzyme profile won’t. #sandiskq1storagegrowth📉 Oil Prices Soar and Gold Falls Below 4000: Funds Are Withdrawing Indiscriminately!
The current market is experiencing extreme liquidity squeeze. The Middle East situation is pushing oil prices up, reigniting inflation concerns, and even spot gold can't hold, plunging straight below the 4000 USD mark.
Tightening macro liquidity is causing the crypto market to face pressure simultaneously. BTC is around 80800, breaking key support, with ETF outflows exceeding 600 million USD in a single day, reflecting strong risk-averse sentiment.
ETH is around 2379, struggling weakly. After following the broader market down, the buying strength is average, lacking the confidence to strengthen independently.
ZEC is around 1165, down over 5%. However, amid the weak market, spot ETFs have recorded a net inflow of over 20 million USD against the trend, and even a whale has opened multi-million level long positions. This indicates that under panic sentiment, funds are seeking defensive assets with independent narratives.
Practical insight: Before the macro situation stabilizes, funds are indiscriminately selling off. Don't blindly bottom-fish weak altcoins; observe the main force's defensive direction (such as ZEC's capital support) before making a move.
This article is for market analysis only and does not constitute investment advice. #10月FOMC临近，美联储暂停加息还有哪些变数？ #现货黄金跌破4000美元关口 #中东原油供应风险再起，油价持续走强 $BTC $ETH $ZEC #谷歌云业务增长82%，AI投入回报如何？
Google Cloud's Q2 2026 revenue surged 82% year-over-year to $24.8 billion, with returns on AI investments clearly realized on the revenue side, but the costs are also significant.
Returns evident: Cloud business operating profit jumped from $2.8 billion in the same period last year to $8.8 billion, with a significant improvement in profit margin. Growth is directly driven by enterprise AI infrastructure and solution demand, with cloud business backlog reaching $514 billion, providing visibility for future revenue.
Heavy cost: Capital expenditures for the quarter doubled to $44.9 billion, leading to free cash flow turning negative for the first time (-$5.9 billion). The full-year capital expenditure guidance has been raised to $195 billion to $205 billion. Market concerns focus on whether the "burn rate" of AI infrastructure spending is squeezing cash flow and if the return cycle can keep pace with the investment rhythm, which remains an open question. Crypto markets are drifting through a dead zone, and the calm is misleading. After a sharp selloff overnight, prices have stabilized into what looks like a recovery but behaves like quicksand: macro funds are parked ahead of the FOMC decision, liquidity has thinned, and the tape now rewards patience rather than conviction. Every bounce is being sold into, every dip met with indifference. This is not accumulation. It is a waiting room. $BTC tells the story most clearly. Pinned near 81,500, the leA completion acceptance certificate that only takes effect on January 18, 2027, is now pressing down on the load-bearing walls of every stablecoin skyscraper. The rules finalized on October 21 are not a renovation notice; they are a structural modification permit, issued more than two years in advance to the general contractor.
Let's first look at the foundation. The issuer's reserves are the foundation, and the redemption clause is the load-bearing column. For any project, if the reinforcement ratio in these two areas is insufficient or the concrete grade is inadequate, no matter how fancy the upper structure is, a small liquidity drizzle can topple the whole thing. Fixing these two items on the blueprint is a good thing; what I fear most in this industry is a stunning rendering with no structural calculations. The white paper is always just a design plan; what truly determines whether this building is livable is the underlying architecture, construction quality, and long-term scalability.
The real variables lie in the cantilevered parts. In the past, third-party rewards were treated as illegal additions and were all removed; now, this "presumed opposition" beam has been removed, effectively leaving an opening for additions outside the main structure. But leaving an opening does not mean exemption from inspection. You can cantilever on the facade, provided every load transfer is clearly accounted for: from which layer the funds come, through how many beams, at which node it converts into rewards, and whether extreme market conditions might resonate with redemption waves. Compliance boundaries are not drawn on land-use red lines but on seismic resistance levels.
The American Bankers Association's opposition essentially reflects concerns about foundation settlement. If deposits are withdrawn from this load-bearing system to support another structure, the stress balance of the entire block will change. The Treasury Secretary's statement is equivalent to the planning department saying: innovation is allowed, but acceptance standards will not be lowered. These two statements are not contradictory; they belong respectively to the geological survey report and construction codes.
Therefore, what the platform receives is not a pass but a renovation permit. If rewards can be made into secondary beams that participate in load-bearing rather than decorative curtain walls on the facade, this addition is valid; if made decorative, a strong wind will peel off the entire section, and those hit will be the people downstairs. The more than two years before the effective date is the reinforcement window: whoever completes the structural strengthening will get the qualification stamp on delivery day; whoever just changes the curtain wall color will be exposed in the first stress test.
As for that linked asset, don't just look at its facade rendering; look at its load transfer path—where exactly the reward funds come from which reserve pool, how many layers of structural conversion it passes through, and ultimately which foundation it presses on. I never sign off on buildings whose blueprints do not specify seismic resistance levels. #geniusstablecoinrewardsRambling thoughts
Analyzed my positions and strategy with Gemini. Gained some insights I had never considered or had overlooked.
First is the difference between coin-margined perpetual contracts and coin-margined leverage. I used to think coin-margined contracts earn coins without leverage. Only now do I realize leverage can also earn coins. Moreover, the funding rate annualized for coin-margined contracts is about 100%, which basically means holding for a year would almost completely erode the principal. In contrast, leverage interest annualized is less than 5%, allowing one to get through an entire bull and bear cycle. Coin-margined leverage is more suitable.
Second, the essence of perpetual contracts is as a tool for short-term and mid-term swing trading to navigate bull and bear markets. It’s not suitable for long-term use; it feels like using a cannon to shoot mosquitoes. It’s better used for intraday swing trading, separated from long-term positions.
Third, I reorganized my positions.
1. Too many positions, too messy. I decided not to add more to the original coin-margined ETH and BTC long contracts, and will fully convert to coin-margined leverage.
2. Continue funding in, keep the dollar-cost averaging plan unchanged, keep adding to spot positions, and use spot BTC as low-pressure collateral to borrow coins for leverage.
3. Update the spot order price points. Previously, I only focused on the 75K level. Now I place spot orders in batches on the left side between 73-76K. The last position is at 65K. If the close price continues below 65K, all adding positions will stop, the dollar-cost averaging plan remains, and leverage is conceded.
4. The overall leverage level of the account does not exceed 2x. All current positions are opened under the premise of a bull market starting, ensuring no liquidation during a bear market. I view this message as a headwind to bullish sentiment, somewhat bearish, but not enough to confirm a trend reversal. According to TechFlow, about $2.6 billion worth of $BTC flowed into exchanges in a single day, the largest scale since April; after the last similar inflow peak, BTC prices topped out within weeks.
My first principle for trend analysis is price confirmation first. Inflows to exchanges do not equal sales yet; if the price subsequently breaks key support and rebounds fail to recover, then in my view, this message gains price validation.
The second principle is to observe sustainability. No matter how large the single-day figure is, we need to watch if inflows continue consecutively. If selling pressure keeps increasing and price support weakens, I would judge that bullish sentiment is more likely to be eroded; drawing conclusions from just one day's data is premature.
The third principle is that history serves only as a reminder. The last inflow peak was followed by a top, but we cannot directly infer the same will happen this time. However, this association can increase holders' hesitation, and short-term sentiment may be impacted faster than the trend itself.
The trend is the best friend. My stance remains to wait for confirmation and not rush to guess the top. Facing this large inflow, what methods are you using now to judge the trend—looking at moving averages, volume and price, or changes in on-chain funds over several days?A cumulonimbus cloud worth $825 billion is looming over Meta's head, yet the market radar is only fixated on the false clear sky echo of 26%. Ground observation stations report quarterly revenue of $6.4601 billion, a 26% year-over-year increase, but the upper atmosphere detection data has already changed flavor—the future lease and irrevocable contract commitments have surged from nearly $700 billion to $825 billion, with moisture flux being injected into the troposphere wildly under the name of AI infrastructure.
This is not an ordinary frontal passage; it is a typical precursor of strong convection with warm, moist air intensifying and cold air poised to strike. Revenue is the ground temperature—looks good; but the weather system's trajectory is always determined by the upper air pressure field. The five-year credit default swap spread has widened to 116 basis points, equivalent to a sudden steepening of the pressure gradient in a very short time, and wind shear has already appeared; the 7% drop on October 22 was just the first strong echo cell on the radar, the real squall line is still coming.
The AI investment return cycle, as the forecasters say, is a "precipitation lag." Moisture has already risen, but when it will rain, where it will fall, and how much, the ensemble members of numerical forecasts differ greatly. What the market is doing now is discounting the certainty of precipitation in advance, but Meta's cloud base height is rising—the time lag between long-term capital demand and return realization is restructuring the entire evaluation model's pressure configuration. The $825 billion commitment is not the precipitation amount but the atmospheric precipitable water; if it hangs overhead without falling, it is sultry heat; if it falls, it is a downpour.
Looking downstream, US stock token assets like $xPLTR are in the downstream confluence zone of Meta's weather system. When the upstream trough shifts, the jet stream guiding airflow changes, and the downstream precipitation probability must be recalculated. Meta's spending concerns are not an isolated high-pressure ridge; it transmits volatility energy downstream through the credit spread jet stream. The linkage of US stock token assets is essentially an upstream-downstream effect within the same weather-scale system—when upstream vorticity advection is positive, downstream must guard against cyclone formation. Now that the spread pressure gradient is increasing, downstream visibility will only worsen.
The forecasters' judgment is straightforward: revenue sunshine cannot mask the deepening upper trough; the $825 billion precipitable water hanging overhead, combined with a 116 basis point pressure gradient, is definitely not a "local shower" setup. When convective available potential energy is still rising and vertical wind shear has met the threshold, any trigger—even just a newly disclosed filing—can ignite deep convection. XPLTR's downstream radar echoes are waiting for the final positioning of that trapped front upstream.
With every radar rotation, the $825 billion of suspended moisture presses down another inch—this is not forecast divergence; the weather system has already locked in the landing zone. #metaaispendingconcerns#SK Hynix Q3 Earnings Report Approaching, Can HBM4 Boost Profitability?
On October 28, SK Hynix will release its Q3 earnings report, with consensus operating profit around 7.696 trillion KRW and a profit margin of about 78.4%, expected to hit a record high for the sixth consecutive quarter, though this is a 5.6% downward revision from three months ago. Citi is more optimistic, forecasting 8.13 trillion KRW, exceeding expectations by 8%.
But there is noise in reality:
- HBM4 shipments fell short of expectations, affected by Nvidia Rubin's delay and tuning, leading to a downward revision of the full-year outlook, with some capacity redirected to general DRAM.
- Samsung's Q3 HBM revenue has already surpassed SK's (Bernstein data: Samsung $8.7 billion vs. SK $7.6 billion), rapidly narrowing the market share gap, with yields reportedly near 80%.
- Some back-end equipment orders have been postponed to next year, raising doubts about the pace of capacity expansion.
In the short term, Q3 is still likely to be strong, supported by price increases and product mix to maintain high profits. The real test will be in Q4 and 2027: the ramp-up speed of HBM4, whether Nvidia's share can be maintained, and the catch-up efforts of Samsung and Micron. If the pace goes smoothly, profit elasticity will continue to be released; otherwise, profit margin expectations may be further revised downward.
Key points to watch in the earnings report: actual contribution of HBM4, 2027 capacity and capital expenditure guidance, and resource allocation attitude. The AI storage cycle is not over, but "who can better turn HBM4 into sustained high profits" has shifted from a single dominant player to a three-way competition.
#SKHynix #HBM4 #StorageChips #EarningsSeason Brent at $116.72: a six-month high, yet the Dow dropped 653 points.
On Friday, all three major US stock indexes closed lower, with the Dow leading the decline at -1.23%. Houthi attacks on Saudi Arabia, oil leaks reported in the UAE, and huge disagreements in US-Iran talks pushed oil prices from around the $100 mark to the highest settlement price since April within a week.
The real divergence is in the direction of money flow: American Express announced 3,500 layoffs, its stock fell -3.7%, and funds exited financials, industrials, and consumer sectors, turning sharply into energy. The S&P 500 Energy Index surged +2.5% in a single day to a record high, ExxonMobil +2.79%, Chevron +2.88%, both hitting 52-week highs.
I think this is not a restart of the energy rally, but rather money seeking shelter during an AI panic week: the Nasdaq only fell 0.21% in a day, the Philadelphia Semiconductor Index rebounded 0.44%, money hasn’t left the market, it just moved from "borrowing money to burn on AI" to "visible oil barrels."
The invalidation level is clear: if US-Iran talks make substantial progress or Brent falls below $108, the safe-haven logic for energy immediately collapses.
On Monday, what do you want to see more: oil prices continuing to surge to 120, or AI stocks pulling money back?
$XOM $CVX $USO
#crudeoil #MiddleEastEnergyRisksPushOilPricesUp Irys peaked at 16x in two days? Don't just look at the multiples, check the drawdown
Everyone is talking about Irys today, so I'll join the buzz and do some quick math:
Yesterday's low was 0.01980, rising to 0.07621, about 3.85x;
Today it went from 0.06031 to 0.32, about 5.31x;
If you held from yesterday's lowest point to today's highest point, that's nearly 16x.
Sounds tempting, right? But look at another number: the current price is 0.0830, down about 74% from today's high. Those who jumped in at 0.32 are now facing over 70% unrealized losses.
Also, a reminder: Irys is not a newly launched coin; it was issued on November 25 last year. After nearly a year of silence, it suddenly surged consecutively. This kind of movement raises the question: why now? Is there a real fundamental change, or is it just capital using the topic to manipulate? Until this question is clear, the multiples, no matter how attractive, are just someone else's story.
The rapid surge of an old coin is different from new coin hype: there are a lot of early low-cost holders waiting above; the higher it goes, the more people are waiting to cash out. The sharp drop in the past two days has already demonstrated this.
It's not that you can't touch it, but think clearly about which part of the profit you want to make: those with the edge and discipline capture the initial move; those who enter only after seeing "16x" are most likely providing liquidity for others.
Position size and stop loss are always more important than multiples. FOMO is the most expensive; don't use your own capital to pay for others.The US stock market has closed. Those who watched all night did not see a rebound, but a continuous slow decline. Don't close the software yet; the truly chilling thing is hidden in the long-short ratio👇 $BTC is currently above 81300, grinding along the bottom all night. The most glaring issue is not the price, but the 4-hour long-short ratio still as high as 2.29 — the more it falls, the more retail investors treat it as a discount and collectively add positions, stubbornly holding on. With buyers so crowded, the main players have no reason to lift the price here. $ETH at 2399, the 2400 level has been completely lost. Even more absurd is the 4-hour long-short ratio surging to 3.02, the screen full of retail investors bottom-fishing with long positions. This vehicle is too heavy; the script is almost written on the face: the market makers won't pump the price until the high-leverage positions below are cleaned out. $SOL at 103.75, weakening along with the broader market, the 4-hour long-short ratio is also as high as 2.64, retail investors still stubbornly holding, short-term remains weak. Putting these three numbers together, the conclusion is clear: this is not an ordinary pullback, but a stalemate in the chip structure. The more retail investors refuse to admit defeat, the harder it is for the market to find a true bottom. With ETF outflows exceeding $600 million in a single day and the October FOMC approaching, neither macro nor capital flows have given a signal for a turnaround, and the crowded longs themselves are the biggest downside fuel. My judgment is very clear — before the leverage cleanup is complete, any rebound lacks sustainability, the price direction remains bearish, and the real confirmation signal is a rapid drop in the long-short ratio accompanied by a volume surge liquidation round; that is the starting point of emotional clearing. So I continue to stay out and watch. Entering now to bottom-fish is equivalent to voluntarily putting your neck on the chopping block. Patience is key Illusion of the Heat Ranking: IRYS surges 1200% in one day, PEPE rallies against the trend, ATOM shorts are being hunted
The frenzy on the gainers list often masks the true structure of the chips. Breaking down the top three hottest coins today, each has a different logic
$IRYS surged over 1200% in 24 hours, with a 7-day increase of 2083%. But what drives it is not consensus, but short squeeze: about $21.4 million in short positions were forcibly liquidated, and open interest surged to $122 million. Circulating supply accounts for only 20% of the total, chips are concentrated, liquidity is thin, and the real test comes after the squeeze subsides and buyers step in
$PEPE rose about 12.7% in hours, ranking first in global viewership. However, Hyperliquid perpetual contracts have about $41.7 million in open interest, and funding rates show no extreme premium. The rise is not caused by crowded longs, and a chasing consensus has not yet formed
$ATOM is currently priced around $1.80, up 3.71% in 24 hours. Around $1.802, $18,600 worth of short liquidations were triggered, and conditions for a short squeeze are forming. But the daily trend has been bearish since the sell-off in October last year. Whether the $1.78-$1.80 support holds will determine if this is the start of a short squeeze or the end of a bull trap
Summary: For IRYS, watch if anyone picks up after unlocking; for PEPE, see how far the strong momentum without crowded funding rates can go; for ATOM, watch if the $1.78 support can ignite a short squeeze. Gains on the heat ranking never equal a margin of safety
#10月FOMC临近，美联储暂停加息还有哪些变数？