Post

峰哥的交易日记
峰哥的交易日记
Show original
闪迪跌近10%,美光、希捷跌超7%,西部数据跌超6%,SK海力士ADR跌超5%。费城半导体指数盘中扩大跌幅至5%。 SK海力士Q3营业利润83.33万亿韩元,营业利润率80%,历史新高。CEO郭鲁正在电话会上说了一句让所有空头沉默的话——“未来三年直至2029年,DRAM和HBM产能实际上已经近乎售罄”,客户需求量是公司供应能力的两倍以上。 三星2027年HBM产能全部售罄。铠侠2027年NAND产能全部售罄。 利润创历史新高。产能卖到四年后。供不应求延续到2030年。 然后股价跌了10%。 这不是基本面出了问题。这是得州一个州长选举的结果,暂时压倒了四年的订单簿。 11月4日,民主党候选人Gina Hinojosa赢下得州州长。她竞选时的核心承诺之一:上任后立即暂停数据中心项目审批,重新审查这些项目对电力、水资源和当地基建的冲击。 高盛当天出了一份报告,给了一个数字:地方层面的审批冻结,可能影响美国约45%的规划中数据中心新增产能,而得州一州就占了约25%。 市场的逻辑链条瞬间打通:如果数据中心不建了→芯片卖给谁?→先卖了再说。 摩根大通在暴跌前一天发布的存储行业报告写得清清楚楚:三季度DRAM价格环比涨了23%,高于预期;HBM4已经开始出货;厂商一致表示2027年以后需求仍将超过供应,2027财年超过75%的产能已经锁定。小摩维持对SK海力士、三星、铠侠的“增持”评级。 高盛和摩根士丹利同期做了一件事:把存储股的估值框架从市净率改成了市盈率,因为“周期股”的标签已经配不上这张订单簿了。 换句话说:华尔街不是不看好存储。华尔街是在用卖周期股的方式,卖掉一只已经不像周期股的股票。 当一只股票因为“可能少建几个数据中心”而下跌7%,而它的HBM产能已经卖到了2029年、客户在排队等货的时候,市场到底在定价现实,还是在定价一个尚未发生的政策情景? 监管风险是真实的。审批延迟会带来短期噪音。但延迟的是节奏,不是需求。得州暂停审批,改变不了微软、亚马逊、谷歌已经签下的长期算力合同,也改变不了HBM供不应求到2030年的产业事实。 存储股自2026年高点已经回撤了约45%,估值压到了2027年预期市盈率不足3.5倍。这个价格里隐含的假设是:AI资本开支要出大问题。 但财报说的正好相反。 暴跌那天,股吧里有人打了一行字:“科技泡沫开始爆破了”。 泡沫爆破的时候,产能是卖不到2029年的。 你信一个州长竞选承诺带来的情绪定价,还是信一份签了字的、锁到2029年的订单簿? $SNDK $SKHY $AMD
峰哥的交易日记
峰哥的交易日记
Can you still hold onto your $93 SOL? ETF has seen continuous net outflows, with $71.6 million running off in one week, down 68% from its all-time high, and the whole screen is full of people calling SOL trash—but just as you're about to cut losses, Samsung Wallet quietly integrated Solana, Schroders moved bond funds onto the chain, and the Breakpoint conference kicks off next week in London. Is this the last chance to escape before a crash, or just a classic panic washout orchestrated by whales? Let's look at the surface first: bad news bombardment, but the price just won't break 90. After weeks of steady decline, the daily RSI dropped to 28-30, technically oversold. Trading volume remains active over 24 hours, with 89.8 stubbornly holding. The candlesticks tell you: the price is clinging just above the 200-day moving average (around 89), and the upward trendline since June is right beneath. All indicators scream oversold, but retail only sees "down." First thing: ETFs are running, but institutions are doing something else. In one week at the end of October, SOL spot ETFs saw net outflows of $71.6 million, with inflows near zero for many days. Sounds scary? But look at another set of data: tokenized asset size broke $5 billion to a new high, stablecoin supply is $14.6 billion, Samsung Wallet plans to support USDC cross-border transfers on Solana, and Schroders—a giant asset manager with nearly a trillion dollars under management—has moved bond funds onto Solana. In plain terms: retail is selling ETFs, institutions are building infrastructure. The same script played out before SOL rose from $8 to $293 in 2023. Back then, ETFs didn’t even exist, and everyone called it a "downtime chain," but then Visa and PayPal came in, and the price multiplied 30 times. Remember this: ETF outflows are sentiment, institutions going on-chain is the trend. Sentiment can deceive, trends don’t. Second thing: Breakpoint kicks off next week, guess what news will drop? November 15-17, London, Solana’s biggest annual conference. Alpenglow upgrade is live, greatly improving finality. Validators are voting on inflation and fee proposals. Tokenized stocks, payment infrastructure, RWAs—all potential topics at this conference. Do you know what SOL does around every Breakpoint? In 2023, SOL launched a major uptrend before and after Breakpoint. In 2024, ecosystem projects collectively exploded during the conference. I’m not saying history will definitely repeat. I’m saying: cutting losses at 93 due to ETF outflows one week before a major catalyst—do the math yourself if that’s worth it. Third thing: technically, SOL is at a "do or die" position. 90-91 is the upward trendline and recent low since June. 88-89 is the 200-day moving average. These two levels together form SOL’s last mid-term defense line. Daily RSI 28-30, oversold. MACD is still below zero, bears aren’t dead yet but stretched to the limit. Short-term rebound signs appear; 95-98 above is a must-pass hurdle. Holding above 100 and reclaiming the 50-day moving average (around 108) would mark a mid-term trend reversal. But the first rebound battle starts right at 90-91. Bull vs. bear showdown, you decide: On the bullish side: RSI 28-30, technically oversold to the extreme Double support at 90-91 trendline + 88-89 moving average Breakpoint conference catalyst next week Samsung, Schroders, Visa, PayPal continuously onboarding RWA breaks $5 billion new high, stablecoins at $14.6 billion On the bearish side: ETF outflow of $71.6 million in one week, institutions cooling off short-term 68% retracement from ATH, believers worn down 50-day moving average at 108 pressing down, mid-term trend bearish Tightening macro, December rate hike priced in Binance security incident rattling sentiment Key levels Resistance above: 95-98 → 100 (psychological) → 105-110 (50-day MA zone) → 120 Support below: 90-91 (trendline) → 88-89 (200-day MA) → 85 (breakdown abyss) Trading strategy Short-term players: Wait for a pullback to 90-91 with hammer or engulfing pattern + volume surge, go light long, stop loss below 88, target 95-98. Add positions if volume holds above 98, eyeing 100-105. If rebound stalls at 95-98 and falls back, reduce or lightly short, target retest 90. Swing traders: Build long positions in batches between 88-93, leverage controlled within 5-10x. Stop loss: daily close below 85-88. Targets first at 100-110 recovery, Breakpoint catalyst + ETF turning positive could push 120+. Long-term believers: DCA below 90. SOL is down 68%, but the ecosystem is more than twice as strong as in 2024. Target 200-300 by 2027, betting on RWA + payments + institutional adoption for a long-term positive cycle. But remember—don’t rush to fully load if 100 doesn’t hold. SOL now is like SOL in August 2023— Everyone was calling it a "downtime chain with no hope," then two months later Visa came in and it went from 20 to 100. $BTC $ETH $SOL

Disclaimer: OKX Orbit content is provided for informational purposes only. Learn more

Replies

No comments yet. Be the first to reply!