Orbit Post Sitemap
$BTC hit a low of 62,530 last night.
My long stop loss at 62,500 was triggered just 30 dollars away. Cold sweat.
What scared me more was the data from Coinglass: breaking below 62,000, the cumulative long liquidation intensity on major CEXs reached $803 million. What does that mean? If this level breaks, $800 million worth of long positions will be forcibly liquidated, causing a chain reaction crash, directly down to 61,000 or even lower. Conversely, breaking above 64,000 triggers short liquidations of $888 million, pulling the price back up to 65,000.
The current market is hanging in the balance between these levels. The reason it can't go up is because ETFs have had outflows for three consecutive days: $144.6 million on August 10, $61.1 million on the 12th, and $131.1 million on the 13th — a total of $336 million out in three days. Last week there was an inflow of $860 million, but this week nearly half was given back in just three days. Institutions are withdrawing, so we shouldn't bottom-fish.
The reason it can't go down is because whale wallets holding over 1,000 BTC have reached a new 2026 high of 3.06 million BTC. Smart money is quietly accumulating; accumulation is a long-term logic, while dumping is short-term logic.
My plan: watch around 62,500, no bottom-fishing. If it can't break 63,500-63,800, lightly try shorting. Stop loss at 64,200, target 62,500-62,000.
Liquidity is low over the weekend; a single large order can either crash or spike the price. Position size at 30%, zero leverage. Avoid heavy positions over the weekend.
#BTC #liquidation #现货ETF资金回流，BTC与ETH能否接力？ In recent days, looking at Dogecoin, it appears to be sideways on the surface, but in reality, it feels more like a silence after an emotional retreat.
As of August 15, DOGE is priced around $0.07, with only slight daily fluctuations; market cap is about $11 billion. More notably, recent trading volume has gradually fallen from nearly 576 million coins on August 11 to about 254 million coins. The price hasn't dropped much, but volume is contracting. This signal does not mean "the bottom is stabilized," but more likely indicates the market has temporarily lost active buying.
I don't think the most pressing question about DOGE right now is "will it suddenly pump," but rather: does it still qualify as the preferred emotional asset for capital?
DOGE's greatest value has never been technology or on-chain applications, but its ability to absorb retail investor sentiment during bull markets.
When market risk appetite rises, BTC and major coins become "too expensive," and no new narrative consensus forms, capital turns back to find an asset everyone knows, with sufficient liquidity and a simple enough story. DOGE has been repeatedly chosen in the past, not because it is superior, but because it is best suited as an "emotional currency."
But the problem is, the current meme market is no longer dominated by DOGE alone. New coins emerge endlessly, with more exaggerated short-term gains, fragmenting attention. DOGE's advantages remain its recognition and liquidity, while its disadvantage is a lack of freshness $DOGE Every major BTC rally begins at a macro liquidity inflection point.
March 2020 — pandemic crash, Federal Reserve unlimited QE. BTC rose from 3,800 to 69,000.
Early 2023 — rate hike pace slows, market starts pricing in a “pivot.” BTC rose from 16,000 to 70,000+.
What about this time?
July 29 FOMC, Federal Reserve held rates steady for the fifth consecutive time at 3.50%-3.75%.
The key is — rate hike expectations are collapsing.
Early August, the market priced a 55% chance of a rate hike in September.
After CPI release, it dropped to 44.1%.
By August 15, CME data showed the probability of holding rates steady in September had risen to 67.5%, with rate hike odds down to 32.5%.
From 55% to 32.5% — this is not the end, but a signal that the Fed’s narrative is starting to loosen.
Short-term traders see “BTC hasn’t risen.”
Long-term holders see “the spark has been lit.”
The drop in rate hike probability from 55% to 32.5% is not the end, but a precursor to the Fed narrative beginning to collapse.
Consumer data shifting from “strong” to “unexpected decline” is not volatility, it’s a trend.
The trend is set, only awaiting Fed confirmation.
And once confirmed — BTC’s breakout always starts when most are still hesitating. $BTC US Stocks Look Calm. The Risk Isn’t.
The S&P 500 and Nasdaq barely finished higher this week, while the Dow slipped.
On the surface, that looks like normal rotation.
Underneath, I see something more fragile: market gains are becoming increasingly concentrated.
Money is crowding into a handful of mega-cap tech names while the broader economy is showing cracks.
US retail sales fell for the first time in nine months. Oil is rising. Treasury yields are climbing. Consumers are feeling the pressure of higher prices.
And then there’s AI.
The AI story is still powerful, but expectations are becoming enormous. When future revenue projections start doing most of the valuation work, the market leaves very little room for disappointment.
That’s where leverage becomes dangerous.
Leverage accelerates gains on the way up, but it can turn a small correction into forced selling on the way down.
The key question now isn’t:
“Can US stocks keep rising?”
It’s:
“How long can concentrated AI-driven growth hold while consumers weaken and yields rise?”
I’m watching the Fed closely.
If consumer weakness becomes clearer, rate-cut expectations could return.
But if inflation and energy prices stay sticky, the Fed may have little room to ease.
That creates a very delicate setup.
A pullback toward the 200-day moving average wouldn’t necessarily be bearish. It could simply be the market resetting expectations before the next major move.
The indices may look stable.
The underlying balance is anything but.
#WeakConsumptionFedSplit #OpenAIAnthropicRace #SKHynixCapexSurge $BTC In the past couple of days, a major piece of news has emerged in the DeFi sector: Strategy Inc. (MicroStrategy)'s STRC preferred stock yield product has been officially integrated into the Solana chain by Solstice! As a product on Solana that emphasizes "institutional-grade Real-Yield," Solstice even launched the strcUSX vault with advanced and sub-tiered layers, steadily surpassing the $400 million mark (TVL). However, looking at the market, SLX (Solstice) is still firmly holding near $0.077 (a full 58% retracement from the July high of $0.1854). While fundamentals are soaring, coin prices are falling in despair—what logic is behind this? 1. Typical "Positive Lagging Effect" Retail investors often only look at candlestick movements when watching cryptocurrencies; But institutions and big money focus on projects, focusing on ecosystem implementation and TVL support. In the crypto market, favorable fundamentals (such as integrating traditional US stock preferred yields) often take time to settle. After the price bottoms between $0.074 and $0.076 to end the desperate stamp, this bullish trend will become the strongest bottom support for subsequent rallyes. 2. Chips Washed Dry Amid Extreme Pessimism Daily Trading Volume Remains at the high turnover level of $6.5 million - $8.5 million. The circulating market capitalization has been#SK Hynix Accelerates Expansion, Can Capital Expenditure Deliver Returns?
Let's talk about something that's been hotly debated in the community these past couple of days—the massive SK Hynix expansion deal worth 54 trillion KRW (about $38 billion). Approved by the board on August 7, two wafer fabs, Yongin Y2 (DRAM) and Cheongju M17 (NAND), are being launched simultaneously. Honestly, my first reaction seeing this number was: these semiconductor folks really spend money without blinking.
But as crypto traders, don't just watch the spectacle; this matter relates to the narrative logic of those AI-related altcoins we hold, and even the entire Crypto market.
First, some background. SK Hynix's Q2 earnings this year were quite strong—revenue of 79.32 trillion KRW, up 257% year-over-year, and operating profit of 60.54 trillion KRW, soaring 557% year-over-year. But the problem was market expectations were even higher, so the stock price still fell after the results. Then the company said this year's capital expenditure would reach a high range of 40 trillion KRW, and in August they dropped another 54 trillion KRW to build two new fabs. The Yongin cluster's overall completion date was moved up from 2045 to 2033, cutting 12 years off. This pace is urgent.
So the question is: is this money well spent?
SK Hynix says memory has shifted from being a "common component" to a "core infrastructure determining AI performance." In plain terms, it means—before, people bought memory sticks based on price; now AI giants care about whether you can deliver on time and in full. The company's executives said, "Technical competitiveness alone is no longer enough to maintain an advantage; the real competitiveness lies in delivering sufficient products at the customer's required time." Sounds reasonable, right?
But let's be realistic. SK Hynix's HBM market share in Q1 this year was 58%, with Samsung and Micron each at 21%. The three major manufacturers have already tilted 70% of new production lines toward HBM. Yet, the HBM supply gap is expected to widen from 5% in 2025 to 6% in 2026, and 9% in 2027. Not a big gap? Don't forget the HBM market size is expected to grow 58% this year to $54.6 billion, nearly 40% of the DRAM market.
More absurdly, Samsung, Micron, and SK Hynix have fully allocated their DRAM and HBM capacity for 2027, and most customers end up receiving only 60% to 70% of their initial orders. This supply-demand relationship is fiercer than some altcoin buy orders we see.
But what's the flip side?
Since SK Hynix's US listing, its stock price has dropped nearly 21% from the July high, wiping out over $500 billion in market value. What is the market afraid of? It's afraid that this massive capital expenditure will ultimately become excess capacity. TrendForce data shows that although memory industry capital spending will increase in 2026, its contribution to bit output growth is limited—in other words, a lot of money is spent, but capacity may not be released as scheduled.
There's also an interesting point. In June, SK Hynix actually adjusted its capacity layout, slowing down the HBM4 expansion pace and shifting more resources to general DRAM. This move is somewhat like when we trade and reduce positions in a coin that has surged too much, reallocating to relatively undervalued assets. Big manufacturers are also doing asset allocation.
Now, something closer to us.
SK Hynix went public on Nasdaq in July, raising $26.5 billion. Then it signed a $500 billion cooperation agreement with Nvidia. Jensen Huang personally said this money includes Nvidia's purchases of memory chips and supercomputers. The AI computing infrastructure arms race is visibly accelerating.
What does this mean for Crypto? My understanding is—the big money narrative in the AI track won't fade in the short term. Projects related to AI computing power and distributed storage still have stories to tell. But don't expect these semiconductor giants' expansions to directly benefit any single coin; the industry chain transmission isn't that fast. A more realistic impact is that if HBM capacity really comes online next year and prices stabilize, AI training costs will drop, potentially stimulating more application deployments, which is good for the entire Web3+AI ecosystem.
Finally, some practical words.
SK Hynix's 54 trillion KRW investment means the first cleanroom won't be operational until the end of 2028, and full production is at least three to five years away. Discussing "whether it can deliver returns" now is like guessing next year's Bitcoin price—no one knows. But strategically, the company is betting on AI infrastructure continuing to grow over the next five to ten years. If they're right, this $38 billion is a bottom-fishing opportunity; if wrong, it's another classic case of "cycle-top expansion."
We're not taking sides, just watching the show. After all, it's not my money (doge face).
Let's chat in the comments: do you think SK Hynix's expansion is a brilliant move or a blunder? Anyone holding related assets keeping an eye on this?
$BTC $ETH $SNDK I think SanDisk will have a hard time breaking through 1800, 1700 is possible, and the Nasdaq will still rise sharply next week.
I’m not optimistic about this rally in SanDisk. Although I went long, I also closed my position last night and tried to short but failed 😕
Actually, it’s very simple. The news behind this rally is even more serious.
It’s rumored that a Singapore billionaire foundation is going to invest heavily in SK Hynix, and SanDisk’s meeting on the 13th indicated profit returns to shareholders.
But in reality, the foundation needs to negotiate long-term cooperation with the national government, which is a very lengthy process, and the implementation might take three to five months of negotiation.
And after SanDisk returns profits to shareholders? It still needs to prove its own value to the market. The company is good, and the industry is good, but the final result is what matters to be realized. If the performance does not meet expectations, market pressure will be even more severe, but its financial report has already been released.
Also, I’m not optimistic about the upcoming Nvidia earnings report.
Huang (Jensen Huang) is expanding production, but the scale is too large, so short-term growth will inevitably lag behind. Therefore, its earnings report expenses will likely have the same problem as Google: expenses exceeding income. Nvidia needs to show the market when and how the money spent will come back.
So storage and computing power will be affected, and a drop is inevitable.
Additionally, Musk’s speech yesterday was very interesting. He talked about sensing and computing, no electricity on the ground, and that photovoltaics are underestimated. This speech should be brought up for hype when storage heat cools down in the second half of the month.
Looking at the second half of the year, the first sector to be hit will be computing power, and the one mentioned is electricity.
Also, the relationship between the US and Iran is not actually that bad. Iran hates Trump, and if he really gets impeached in September, reconciliation should begin.A classic historical bottom support is marked by spot volume expansion, combined with neutral or even negative futures funding rates, but the current structure is exactly the opposite.
Spot relative trading volume hovers around 0.75-0.8, at a historical low within the past five years. It wouldn't be an exaggeration to say that crypto is currently "unattended."
Therefore, marginal pricing power is basically handed over to the derivatives market.
Since August 9, perpetual contract open interest (OI) has been continuously rising, suddenly surging to 524,000 BTC on August 14, the highest level in nearly three months.
At the same time, the 7-day average long premium reached $242,000 per hour, returning to the high rebound zones seen in January and May this year.
The difference is: in the previous two instances, the premium reached this level only when the price hit a stage high; this time, the premium is already maxed out while the price is still in a correction.
In terms of relative price performance, the crowding of longs is more severe than the previous two times, with overextension occurring earlier.
Price falls, OI rises, longs continue to pay premiums, indicating leveraged longs are bottom-fishing and adding positions against the trend, while counterparties are firmly selling.
This high-density adversarial position building and directional divergence accumulation will ultimately be resolved in a "do or die" manner. 📊 全网都在刷$OKB，我扫了一眼K线，好家伙，已经站上100美元了。这画面太有冲击力了，隔壁坐着$BNB，价格高高在上地挂在500美元上方，两者一对比，OKB显得格外“亲民”。这也就是为什么市场情绪开始沸腾的原因，散户的逻辑很直接：都是平台币，一个600，一个100，那不还有5倍空间吗？买！但真实的市场逻辑从来不是这样算的。 先别急着按下单键，我们把这个对比拆开看。100美元和600美元之间的差距，看似是“潜在涨幅”，但本质上是市值、流通深度、生态成熟度和市场定位的综合投射。BNB的500美元是用数年的生态建设、BSC链的庞大用户量、以及持续不断的通缩机制一步一步堆出来的，它背后是整个币安帝国的现金流。而OKB呢？说实话，我翻了一圈信息，大家嘴里的叙事来来去去就那几个词：代币销毁、总供应锁定、X Layer生态。听起来很性感，但你问十个人，可能有九个人说不清楚这些叙事到底对应多少真实的链上数据、新增了多少活跃地址、贡献了多大规模的手续费收入。 这就是典型的FOMO叙事驱动行情，价格在涨，故事在传，但真正能动手验证的基本面验证，其实没多少人做过。群里流传的“已经销毁了”“供应全锁了”$APR Brothers holding long positions! I bet your arms are sore from all the pumping.
Yesterday I saw some showing off their profits, still holding with more than tenfold gains.
Just holding is one thing, but being short is inherently painful.
Yet those people keep mocking and ridiculing the short sellers.
Well, this time, this big drop has given me a comfortable fall.
All I can say is: what goes around comes around.
You all know shorts lose money so longs can make money, but how come you don’t realize that if the main players don’t make money, then others won’t either?
Yesterday I posted several times expressing that the spot trading volume of this coin is artificially inflated.
Because on-chain there are always a few addresses trading back and forth with tens of dollars, the highest volume activity is very obvious.
In reality, no real capital has entered the market.
In the futures market, high leverage funds have pumped it up.
And a leverage-driven rally is inherently unstable.
So this big drop is quite justified.
Since the drop turned out like this, it’s probably impossible to recover.
The reason is, a couple of days ago at this level it rose above 0.5, with futures trading volume close to 800 million,
Now if it tries to go back up, I doubt any major player is willing to act as the "liberation army" for the longs.
0.6 is probably impossible now, but 0.06 is just a matter of time. The biggest problem for $BTC bears, aside from expecting a crash in October like carving a mark on a boat, is that if the mark is firmly made, with the S&P 500 index breaking upwards and the BTC-S&P 500 trading pair showing a potential bottom, the math is completely against them.
There are only 7 weeks from early October until now.
In the past four years, every time the S&P 500 index broke through, its rally lasted between 7 and 29 weeks.
But even if the S&P 500 rally lasts only 7 weeks, even during a pullback, it will not fall below the starting point of the rally for at least 12 weeks.
Therefore, bears now need something that has never happened in the past 4 years: the S&P 500 falling back after a breakout.
And even if the BTC-S&P 500 has not yet bottomed and pulls back 5% below the low (many bottom signals have already flashed), a 5% rise in the S&P 500 would offset a 5% drop in the BTC-S&P 500.
Many bearish perspectives heavily rely on the S&P 500 falling back from consolidation because people think this is a midterm election year, but on the contrary, it has already broken through, which is more consistent with a macro risk bull market, and the macro risk bull markets of the past 30 years have never coincided with midterm election years Tencent's new AI product in the optical segment burns over 100 million yuan a day. What exactly is Tencent betting on with this money?
Even though all three giants are pouring money in, their approaches have diverged. Alibaba treats computing power as a business, earning revenue by renting out GPUs; ByteDance is all in on its own Doubao and AI applications; Tencent's choice is the most unique—selling computing power is ranked third in capital expenditure by management.
For Tencent, Goldman Sachs, UBS, Jefferies, and JPMorgan all maintain positive ratings but have lowered short-term profit forecasts and target prices. Goldman Sachs cut from HKD 700 to 670, UBS from 780 to 770, Jefferies to 750, and JPMorgan to 690.
JPMorgan's view aligns most closely with current sentiment: "Valuation is not high, but without clear catalysts and auditable AI revenue, the stock price is unlikely to be quickly re-rated."
Regarding the computing power chain, Tencent raised its full-year capital expenditure forecast from 170 billion to 250 billion yuan. This figure itself is a barometer of AI computing power demand—when it increases its stake, the demand for servers, optical modules, and GPUs along the chain is assured.
The AI application line will be collectively undervalued in the short term until someone first delivers auditable revenue.
Four key things to watch next:
Whether the year-end HunYuan Hy4 can make an impact;
Retention and payment rates of WorkBuddy;
When WeChat Xiaowei will move from grayscale to scale;
Whether Tencent Cloud TokenHub's daily token volume (already 25 trillion, up 5x in two months) can continue to multiply. Gold — $4,380, took a breather after surging to 4,400
Spot gold is at $4,376-4,380 per ounce, having surged to 4,400 and even 4,450 earlier this week before pulling back. U.S. July retail sales unexpectedly dropped 0.6%, cooling Fed rate hike expectations, and a weaker dollar gave it a boost. However, profit-taking concentrated between 4,400-4,500 is the main reason for the pullback. Domestic gold jewelry prices have all broken 1,300 yuan/gram. The world's largest gold ETF holdings increased from 999 tons on July 17 to 1,025.81 tons on August 12.
Fifty percent are digesting in the 4,300-4,400 range; thirty-five percent pulled back to 4,250; fifteen percent held above 4,400 and pushed to 4,450+. $XAUT $BTC is still hovering around $63,000. US inflation and PPI are both cooling down, so risk assets should theoretically get support, but $BTC still can't hold above $64,000.
This indicates the macro environment has indeed improved, but there isn't enough capital truly willing to enter the market yet.
Looking at $ETH, the situation is a bit different.
In July, the US ETH spot ETF saw net inflows of about $350 million, and since August, about $240 million more has flowed in. Although the record of five consecutive weeks of net inflows was interrupted this week, during the same period BTC ETFs saw outflows of about $330 million, while ETH only had a slight outflow of about $3 million.
So what I see is not "ETH is about to take off," but that funds are temporarily showing more resilience than BTC.📌
Next, we really need to wait for two confirmations:
BTC firmly reclaims $64,000, with spot trading volume picking up;
ETH ETF funds resume inflows, and the price starts to generate positive feedback from the capital.
Only when funds and price move together can rotation truly begin.
Liquidity can be positioned in advance, but without price confirmation, I won't rush in just for a new narrative.⚠️#OpenAI与Anthropic估值竞赛升温
The private valuations of the two leading AI companies continue to surge, with Anthropic's valuation surpassing OpenAI's. The capital competition is intensifying, and both companies are advancing IPO preparations. The AI primary market valuations have reached extremely high levels.
Bullish logic: Enterprise AI demand is exploding, revenue is growing rapidly, and institutional funds are pouring crazily into the large model track. The arms race in computing power continues to escalate, driving prosperity in the chip and server industry chains, which forms a positive sentiment for crypto AI computing power themes.
However, the risks behind the high valuations are also prominent. Current valuations are based on long-term revenue expectations and are set by primary market private placements, not verified by the secondary market. Both companies maintain large capital expenditures, and if revenue growth falls short of expectations, there is a risk of significant valuation corrections. At the same time, the massive concentration of funds in the AI track may siphon capital from risk assets.
Personal view: The long-term direction of the AI industry is sound, but current valuations have already priced in many optimistic expectations. Reflecting on the crypto market, this is only suitable as a sentiment reference; do not blindly chase high AI concept tokens.
Going forward, the key focus is to observe the secondary market pricing after the two IPOs. Whether the high valuations in the primary market can hold steady in the public market is the key test of the AI bubble.#加密估值转向收入，BTC如何定价？
MVRV is approaching 1, and the true bottom of this cycle may not have appeared yet.
Looking back at the past three cycles:
📌 2015: Bottomed 14 days after dropping below 1
📌 2018: Bottomed 31 days after dropping below 1
📌 2022: Bottomed 94 days after dropping below 1
The MVRV at cycle bottoms has been rising each time: 0.54 → 0.69 → 0.75—0.80
This data reveals a change: BTC's extreme discount is weakening, but the process of finding the bottom is becoming longer.
In other words, future bear market bottoms may not be deeper, but they could be more drawn out.
Currently, MVRV is about 1.21, BTC is around $63000, and the realized cost across the network is near $52000, about 20% away from the market overall entering unrealized losses.
If this cycle continues to follow the "converging decline, extended time" pattern, then according to this projection:
1️⃣ October–December 2026: MVRV falls below 1, BTC enters the $52000—$56000 cost zone.
2️⃣ January–March 2027:
MVRV dips to 0.82—0.90, BTC searches for the cycle low around $44000—$49000.
Of course, this is not a precise prediction but a baseline path based on previous cycle structures; MVRV falling below 1 is not the bottom, but the market beginning to price in the bottom.昨晚，美国商务部公布数据：7月零售销售环比-0.6%，市场预期可是+0.1%啊。 6月还是正增长0.2%，一个月时间直接翻脸。 消费占美国GDP的70%，这东西一垮，整个经济 narrative 都要 rewrite。 同一天，密歇根大学8月消费者信心指数初值51.0，预期54.5，7月还是55.2。三个月来首次下滑，环比跌了7.6%。 美国人不但没钱花了，连“觉得未来有钱花”的信心都没了。 咱们把过去一周的牌摊开看看： 7月CPI同比3.4%，低于前值3.5%，核心CPI同比降至2.5%。通胀在降温。 7月PPI环比0%，预期可是0.2%。生产者价格原地躺平。 7月非农就业减少2.3万人，预期可是增加8万。5月和6月的数据还被累计下修了10.3万。 四箭齐发：CPI降温 + PPI躺平 + 非农崩溃 + 零售暴跌。 消费熄火了，就业垮了，价格涨不动了——美联储拿什么理由继续加息？ 8月5日，CME FedWatch显示9月加息概率还有58.4%。 8月7日非农出来，掉到55%。 8月12日CPI出来，掉到48%。 8月13日PPI出来，掉到38%。 一周时间，加息概率从58%跌到38$DOOD
Not looking good
Basically all withdrawals to exchanges for selling
Among the top ten addresses, 2 have been continuously selling, the others are exchange hot wallets
No hope of breaking even🙀 Panic: $BTC hiding sell pressure? Exchanges have accumulated 130,000 coins in three months, what's going on
I just saw that Glassnode chart. From May to now, in three months, $BTC reserves on Binance have risen back up, reaching 667,000 coins in August, the highest in nearly half a year.
Coinbase and OKX have also each added tens of thousands of coins. Someone in the group directly shouted to dump, but I think don't rush.
I checked on-chain, and many large deposits are not for selling at all; some are locked in financial products, some are market makers flipping, and some are coins migrated back after hardware wallets were hacked.
CryptoQuant also said net inflow shows no extreme signals, the selling pressure hasn't actually been released.
But we can't pretend to be blind either. BTC has been stuck around 63,000 for almost two months, 30-day net spot demand is negative, futures leverage is pushing up, the rise is hollow. Glassnode says spot trading volume has dropped to the lowest since 2019, buyers and sellers are staring at each other, like frozen.
This pile of coins fished back in these three months is like a knife hanging over our heads. Before they change hands, the higher the bounce, the easier it is for someone to use it to sell.
I don't guess when it will dump, I just wait. $BTC waits for a 40,000 dip or the bell on October 5, holding $OKB at the bottom firmly. When bored, I take some small change to open a small contract for fun, run after making a profit, never fight hard against this frozen period.这两天一组ETF数据搅动市场情绪： 8月12日，美国现货比特币ETF净流出约6116万美元； 8月13日，流出规模进一步放大至1.31亿美元 就在这波流出之前 市场刚经历9个交易日连续净流入 累计吸金超10亿美元 资金前脚持续进场，后脚连续流出，这让市场情绪迅速转向： 有人开始担心，机构是不是已经开始砸盘离场 但事情并没有那么简单 前面9天流入，说明增量资金确实在低位布局BTC； 连续两日流出，更多代表短线资金在上涨后兑现利润，不能直接等同于机构全面撤退。 尤其是8月12日，BTC当天并没有因为ETF流出而大跌，盘中甚至一度冲到64500美元。 这反而说明，当前机构资金更像是在做动态仓位管理： 涨起来，就适度减一点仓； 跌下来，再找位置接回去。 所以现在真正值得盯的，不是单日流出了多少钱，而是三个关键信号： 📌 ETF净流出会不会持续扩大； 📌 BTC能不能重新站上66885美元； 📌 60965美元这个关键支撑会不会失守。 如果ETF只是短暂获利了结 BTC又能守住关键支撑 那这次流出反而可能给下一轮资金进场留出空间 但如果资金持续大额撤离 同时价格跌破60965美元 市场逻辑BTC is on the verge of resistance at $64,000 and ETH at $1,900, yet no chase buying has appeared. Why are good inflation indicators failing to turn into bullish factors? The U.S. CPI for May slowed to 3.4% year-on-year, and the PPI also showed signs of cooling. Despite the macro environment where expectations for rate cuts seem to be reviving, BTC is fluctuating around $63,552 and repeatedly rejecting the $64,000 breakthrough. ETH has hovered around $1,886 and tested $1,900 several times, but has not led to a convincing breakout. The way the market moves is not headlines, but whether expectations are reflected in advance. It is highly likely that the slowdown in inflation itself has already been largely reflected in prices. Traders who participated in pre-priced buying before the indicator release are now taking profits rather than entering new stocks. In other words, the positive factor of slowing inflation has already been reflected in the position, and currently, it is in the stage of digesting the news and exploring the next direction. Here, about 100 million won tonight$BTC
【Long-Term Cycle Analysis】Historical-Level Resonance Between CVDD and NUPL: Does the True Cycle Bottom Still Require One More Dip?
From the perspective of Bitcoin's macro monthly chart over more than a decade, the linkage between CVDD (Coin Days Destroyed Value bottom line model) and NUPL (Net Unrealized Profit and Loss) has consistently maintained a highly tight synchronization. Comparing every deep bear market bottom in history (2015, 2018, 2022), this resonance pattern once again provides clear guidance for the current market evolution: the true macro cycle bottom is often accompanied by a deeper probing release.
1. Resonance Characteristics of Historical Bottoms (Green Highlighted Areas)
NUPL's “Extreme Panic Zone” (< -13.000): At every absolute bottom of the cycle in history, NUPL without exception has fallen below the lower baseline into an extremely oversold negative zone (green shaded area in the chart), completing a full capitulation and clearing of positions.
CVDD Double Track Bottoming: The price must deeply retrace and embed into the CVDD lower support band ($48,000–$57,000 range) on the monthly level, forming a solid structural hard bottom before starting a new long bull run.
2. Current Structure Comparison and Logic for a Lower Bottom
Current position still appears relatively high: Observing the far right current status, Bitcoin's monthly price still hovers above the CVDD upper-middle track (~$62,983), while the lower NUPL currently only stagnates around 16.9, far from reaching the historical-level bottom clearing line (below -13).
Synchronization inevitably requires a lower bottom: To maintain the astonishing “synchronized bottom” iron law of these two major indicators over more than a decade, the current sideways movement is only an intermediate adjustment. The market must experience an accelerated dip, pushing the price closer to the CVDD bottom track (around $48,000–$50,000 or lower), while forcing NUPL to break below zero and even deeply pierce the lower track, to truly complete the cycle-level turnover and bottom formation.
(This is not investment advice, for reference only) SK Hynix Warning: Storage Shortage Will Worsen, Establishing a Super Storage Cycle
Global storage giant SK Hynix publicly warns: next year’s chip shortage may worsen, facing the most severe supply-demand tension in history. This statement breaks the market’s mild expectation of "stabilization next year" and confirms the logic of a super storage cycle.
1. Core Issue: Structural Mismatch
It’s not a capacity shortage, but a conflict between AI and consumer electronics demand. AI servers and large models cause a surge in demand for HBM and high-end DRAM, with leading manufacturers prioritizing limited capacity for the AI track; meanwhile, traditional demands like smartphones, PCs, and automotive remain stable, continuously squeezing supply, creating a "double shortage." Coupled with a 1-2 year expansion cycle and equipment controls, new capacity struggles to meet demand, making shortages hard to resolve next year.
2. Sellers Strongly Dominate
SK Hynix admits price increases exceeded expectations. Downstream buyers scrambling for goods has become the norm, and the industry has entered a seller’s strong cycle.
3. A-Share Opportunities and Strategies
The sector’s strength relies on solid performance, not speculative themes. Key sub-themes include: ① HBM high-end storage (AI essential, highest prosperity); ② General DRAM/NAND (clear price increases); ③ Equipment/Materials/Packaging & Testing (orders continuously landing).
In terms of operations: focus on solid targets with core capacity, reject concept chasing; rely on moving averages to buy on dips, avoid chasing highs; this cycle will last at least until next year, patience in holding beats frequent stock switching. Based on fundamentals, one can seize the tech dividend.
#海力士扩产提速，资本开支能否兑现回报
$SKHYNIX July retail sales hit the largest drop in over a year, consumer confidence fell for the first time in three months, and CPI cooled moderately — then the probability of a rate hike in September dropped from 67% to 36%, only to bounce back to 45%. I stared at these sets of data and laughed for a long time, confirming one thing: Murphy's Law never misses in macro data; it just likes to flip the table when you are most certain. 📊 Let's look at the data: three cards, each more twisted than the last. The first card: consumption suddenly "brakes." U.S. retail sales in July fell 0.6% month-over-month, the largest drop since May last year. The market expected a 0.1% increase — a nearly 0.7 percentage point gap from expectations. Excluding autos and gasoline, sales still fell 0.2%, indicating the consumption weakness is not caused by a single category. The University of Michigan's preliminary consumer sentiment index for August dropped to 51, below July's 55.2 and economists' expectation of 55. One-year inflation expectations actually rose from 4.2% to 4.3%. The second card: inflation cools moderately. July CPI rose 3.4% year-over-year, matching expectations and below the previous 3.5%; core CPI rose 2.5% year-over-year, also matching expectations. The data "perfectly" met expectations — no surprises, no shocks. The third card: rate hike expectations on a roller coaster. After July's nonfarm payrolls turned negative, the probability of a September rate hike once surged to 67%. After the CPI data release, the interest rate swap market cut the September rate hike probability from about 50% to less than 40%. CME data shows the probability of holding rates steady in September is 63.6%, with a rate hike probability of only 36.4% $BTC
BTC Current Situation Analysis
Historically, classic bottom support is characterized by spot volume expansion, combined with neutral or even negative futures funding rates, but the current structure is exactly the opposite.
Spot relative trading volume hovers around 0.75-0.8, at a historical low within the past 5 years. It wouldn't be an exaggeration to say that crypto is currently "unattended."
Therefore, marginal pricing power is basically handed over to the derivatives market.
Since August 9, perpetual contract open interest (OI) has been continuously rising, suddenly surging to 524,000 BTC on August 14, the highest level in nearly three months.
At the same time, the 7-day average long premium reached $242,000 per hour, returning to the high point areas of rebounds in January and May this year.
The difference is: in the previous two instances, the premium reached this level only when the price hit a stage high, but this time the premium is already maxed out while the price is still in a correction.
In terms of relative price performance, the crowding of longs is more severe than the previous two times, with overextension happening earlier.
Price falls, OI rises, longs continue to pay premium, indicating leveraged longs are bottom-fishing and adding positions against the trend, while counterparties are firmly selling.
This high-density opposing position buildup and directional divergence accumulation will ultimately be resolved in a "do or die" manner. The entry of traditional banking channels establishes medium- to long-term liquidity absorption capacity, but the 2027 time window means that the current market dominance still depends on the resonant projection of macro interest rates and U.S. stock risk appetite.
Israel's leading bank Bank Leumi plans to access $BTC spot in 2027. Such pipeline connectivity belongs to the infrastructure dimension of long-term incremental growth. From the recent queue of capital, changes in U.S. Treasury yields, fluctuations in the U.S. dollar index, and the rebalancing rhythm of U.S. tech stocks occupy the top positions in immediate pricing.
When U.S. Treasury yields and the U.S. dollar index remain volatile, the safe-haven attribute of gold and the liquidity premium of crypto assets diverge. If the high interest rate environment persists for a long time, institutional compliant buying tends to hedge on the derivatives side, making it more difficult for the spot market to break through resistance levels upward.
The conditions triggering the bullish scenario are a clear expectation of Federal Reserve rate cuts and a strong rebound in U.S. stocks, while the U.S. dollar index falls below key support levels. If capital flows return from gold back to risk assets, the effectiveness of the spot market attempting to break resistance upward will be confirmed; if U.S. Treasury yields rise again, this bullish projection immediately becomes invalid.
The conditions triggering the bearish scenario are that the high interest rate environment delays the valuation recovery of U.S. stocks, intensifying deleveraging sentiment. Even with long-term positive support such as traditional banks entering in 2027, the short-term spot absorption capacity still struggles to withstand the selling pressure caused by macro liquidity tightening; if the U.S. dollar index quickly breaks below the lower band, the bearish projection loses its foundation.
The key to the market structure lies in the transmission speed of cross-market capital chains. If the U.S. tech sector maintains high beta volatility and the crypto market spot trading volume fails to expand in the same direction, it indirectly confirms that on-exchange funds are still in a stock game state.
In the next 7 days, focus on changes in the U.S. Treasury yield curve, the absorption strength of U.S. stocks at key support levels, and the repricing effect brought by the U.S. dollar index breakout direction on cross-asset liquidity.
#加密估值转向收入，BTC如何定价？ #CLARITY表决待定，SEC规则未落地 #AMD完成历史最大美元债发行：融资47.5亿美元$APR crashed sharply, luckily I switched back to short positions last night.
The reason is, I saw a trading event last night with $APR as the reward, and the reward amount was quite substantial.
Afterwards, I studied it carefully and think it might be hard for it to pump in the short term because there should still be a lot of chips in the market that haven't been collected by the whales.
However, in the long run, $APR still has a good chance to pump.
So, I will consider going long after it stops falling and stabilizes.The squeeze in storage is intensifying, and when everyone is bullish, beware of a double whammy of longs and shorts being crushed
$SNDK continues to surge, SK Hynix moves up in tandem, and bullish sentiment in the storage sector has completely spread.
After a big daily rally, investors did not quickly pull back; the capital support far exceeded expectations, with prices holding above the 1600 level, forcing countless shorts into continuous squeeze.
The market's unified bullish logic is very clear: AI brings massive storage demand, a tens-of-billions buyback supports valuation, and long-term performance targets open up imagination space.
The short-term violent rebound in Korean storage is rooted in the previous deep decline and the replenishment of funds after a large amount of leveraged chips were cleared; SanDisk's rise relies on corporate fundamentals and a long-term growth narrative, representing a valuation re-rating.
Both rises share the same source, but the driving forces are fundamentally different, so the sustainability of the market should not be confused.
Risks hide beneath the celebration: major manufacturers are successively putting new capacity into production, but can the current hot demand expectations be fulfilled as scheduled?
The current rally is trading on a long-term story, and there is still a long way to go before earnings reports are released; a large amount of optimistic expectations have already been priced in.
The market never has both longs and shorts rushing in simultaneously; what plays out repeatedly is the double whammy of longs and shorts being crushed.
Continuous rallies force short holders to stop loss and exit; when short positions are exhausted and bullish momentum fades, a rapid correction will firmly trap traders who chased the highs.
Some around me can't help but chase the rally and enter, while others hold positions against the trend, enduring floating losses and suffering.
I choose to maintain my position and watch quietly, neither panicking to cut losses nor arbitrarily adding chips. Even SK Hynix and Micron in the same sector have pulled back, but SanDisk is pushing upward without looking back. The core reason is the news from Investor Day: future earnings will be made only by necessary investments, 100% of buybacks and dividends distributed to shareholders. The market directly treats this as long-term returns, pricing promises and putting all short-term sentiment on SanDisk. But Ahua is right: this rally is mainly supported by news sources. Once the hype passes, a correction will happen when necessary. The fundamentals remain unchanged. Concerns about peak gross margins remain, and the fact of weak guidance hasn't changed. The market has just chosen not to look at these for now The hourly near 1700 is a short-term resistance level. This level can be used as a reference for short selling, but the position must be light. Set a stop loss above 1730, targeting 1650 to 1630. If SanDisk continues to increase volume and breaks through 1700 and holds steady, short positions must exit decisively rather than holding on. This is a news-driven independent market. Chasing long stocks fears taking over; short selling fears continued pressure. Both sides feel uncomfortable. Wait until the sentiment releases before reconsidering. Short-term short positions can be tested, but don't heavily gamble. Before the trend breaks, short positions are just touching the top, not following the trend with $BTC $ETH $SNDK #消费动能转弱, September policies remain constrained by inflation, #OpenAI与Anthropic估值竞赛升温 #海力士扩产提速 whether capital expenditure can deliver returns NVIDIA holds about $21 billion in SpaceX, with AI collaboration drawing attention, but SKHYNIX did not follow the rise. Current price is 1168.48, up only 0.4% in 24h, trading volume 208,673, funding rate 0, open interest 51,160, liquidity is very dry.
1-hour chart is upward, 3% below the high of 1205; 4-hour chart is downward, rebound is more of a correction. Order book shows obvious selling pressure (sell 14 vs buy 7), short-term resistance above.
Medium term remains bearish. Key levels: resistance at 1205, support at 1004, break below targets 902.
Strategy: Short on rebound at 1205, stop loss at 1220, target 1100. If it pulls back and stabilizes at 1004, a short-term long is possible, stop loss at 980, target 1168.
Risk points: low trading volume, large slippage; if NVIDIA news triggers AI sector volatility, spikes are likely. Keep positions light.
— This is only a personal opinion and does not constitute investment advice. Wish you smooth trading. —
#NVIDIA持有SpaceX约210亿美元，AI协同受关注 $SKHYNIX Why does BTC halving always lead to a price increase? Can't it go down? Is BTC definitely a high-quality asset for long-term growth?
When asked such questions, I always think of Buffett, who started investing at 11 and has been doing so for 84 years. Over these 84 years, people have continuously asked: Does dollar-cost averaging into the S&P 500 always make money? Will the U.S. economy keep growing indefinitely? Looking back at those discussions about the future, the answer was always: not necessarily.
The core principle of investing is faith. If Buffett didn’t believe in America’s destiny, he wouldn’t have been able to sustain his career until now. Faith runs through everything, and there’s no need to doubt it. Because if humanity’s long-term economy collapses, no business will matter anymore. Whether you believe or not, the ultimate outcome is bankruptcy.
For BTC, its long-term rise has little to do with halving or any short-term positive factors. It is the “gold” favored most by Generation Z, a “new asset” where young people have a pricing advantage. Whenever we have some spare money, we’re willing to buy some BTC.
The highest probability strategy for Generation Z is simply to have faith in BTC, to believe in Satoshi Nakamoto, even more than in the operators of U.S. stock companies. In the next 20 years, we have a chance to surpass gold.
Oh, and I haven’t yet explained why BTC rises long-term: 113 years ago, the Federal Reserve was established, and since then, the value of the dollar has lost 97%.Weekend liquidity is poor, and watching the market isn't very meaningful, so let's organize our thoughts together
First, let's talk about two pieces of news
SanDisk dropped a big move on August 13th Investor Day, returning 28-30% of profits after successful investments. The news caused an intraday surge of nearly 18%
But a week ago, when the Q4 earnings report came out, the mid-point guidance for the next quarter was 10.55 billion, lower than expected, causing a post-market drop of over 7%
Explosive performance but slightly missed guidance led to a sell-off; a shareholder return promise then pushed the stock up 13%. This stock moves based on expectation gaps and is extremely sensitive to sentiment
On NVIDIA's side, on August 14th, SEC filings disclosed holding about 123 million shares of SpaceX. These shares were most likely converted from the previous 10 billion investment in xAI during the acquisition, not recent purchases
Musk announced a goal to reach 10 gigawatts of computing power by the end of next year, with AI eventually accounting for 99% of SpaceX's value. It's exciting to hear, but whether the timing of the holdings disclosure is an endorsement of industrial synergy or a risk exposure of related-party transactions, the market still has differing views in my opinion
Back to my holdings, I hold a short position on $SNDK. Returning cash may boost the stock price in the short term, but the company is returning money to shareholders instead of reinvesting, so I think high growth may have peaked
Plus, the market is extremely sensitive to guidance; any negative news could trigger a panic sell-off
However, recently there have been too many shorts, which has caused valuations to keep rising and breaking through
Waiting for the end-of-month earnings report to see the computing power demand guidance. This $NVDA position has a compensation voucher anyway, so I'm not worried
#闪迪投资者日后股价大涨，长期目标待验证 🚨 ETF MONEY IS BACK — BUT HERE’S THE PART TRADERS ARE MISSING
Spot crypto ETFs are attracting serious attention again.
Recent data highlighted roughly $865M of weekly Bitcoin ETF net inflows, with BlackRock contributing around $694M alone.
That sounds extremely bullish.
But price action is telling a more complicated story.
$BTC remains stuck around the mid-$60K region instead of immediately breaking higher.
That creates an important divergence:
📈 ETF demand improving
📉 Price still struggling
💧 Market liquidity remains thin
👀 Altcoin participation remains selective
This is why chasing every green candle can be dangerous.
If ETF inflows continue AND BTC breaks resistance with expanding spot volume, the signal becomes much stronger.
Until then, capital may simply be rotating rather than creating a new market-wide expansion.
Watch:
$BTC $ETH $SOL $XRP $HYPE $SUI $BNB
ETF inflows matter.
But what price does with those inflows matters even more. 🔥
#WeakConsumptionFedSplit #OpenAIAnthropicRace 🚨 FORGET THE NOISE — THIS NUMBER COULD MATTER MORE THAN BTC HEADLINES
The U.S. 10-year Treasury yield is back near 4.67%, keeping financial conditions firmly in focus.
Why should crypto traders care?
Because yields influence the cost of capital.
Higher yields can pressure:
📉 Risk appetite
📉 Leverage
📉 High-beta altcoins
📉 Speculative assets
But falling yields can create the opposite effect:
💰 More risk appetite
💰 Easier financial conditions
💰 Stronger liquidity flows
💰 More appetite for crypto
That's why BTC can sometimes ignore seemingly bullish headlines.
The market isn't only trading crypto news.
It's trading global liquidity.
Watch:
$BTC $ETH $SOL $BNB $HYPE $SUI
If yields ease while ETF flows remain positive, the setup becomes much more interesting.
Liquidity first.
Narrative second. 👀
#Nvidia21BSpaceXStake #OpenAIAnthropicRace #WeakConsumptionFedSplit Zero-threshold configuration for US stocks? ACO native DEX's journey of RWA real-world asset tokenization 📈
Traditional investors wanting to participate in global premium asset allocation often face complicated account opening procedures, deposit and withdrawal restrictions, and high cross-border fees.
ACO native DEX introduces the RWA (Real-World Asset tokenization) native module:
🌐 US stock tokens trade 24/7: priced in mainstream stablecoins like USDT, enabling seamless buying and selling of premium US stock tokens without being limited by traditional stock market hours.
🔒 On-chain asset transparent anchoring: through decentralized oracles and multi-signature custody, ensuring a 1:1 mapping and transparency between real assets and on-chain tokens.
🔄 One-click cross-chain and circulation: say goodbye to complicated Web2 bank wire transfers; assets are available on-chain for instant use and exchange, balancing liquidity and flexibility.
Seamlessly integrating Web3 funds into global premium assets—this is the core empowerment brought by RWA.
#RWA #USStockTokens #ACO #DEX #DeFi On August 4th, a bomb was dropped on the Ethereum forum.
Six researchers jointly submitted EIP-8363, named "Progressive Issuance Burn," with the core idea in one sentence — when the staking rate reaches 50%, the new issuance rewards for validators will gradually be burned down to zero.
The current staking rate is already 34%, and at the current pace, 50% is expected around the end of 2027 to early 2028.
This proposal was specifically discussed for half an hour at the core developers' meeting on August 6th. On the same day, the community suggested removing it from the Hegota upgrade. It took only two days from submission to the suggestion of removal.
Opposition exploded immediately.
Aave founder Stani was the first to oppose, and SharpLink CEO Joseph Chalom published a long article listing four reasons.
Staking yields serve as the benchmark for all on-chain interest rates, and $35 billion worth of liquid staking tokens are the core collateral for on-chain lending.
If yields drop to zero, on-chain capital costs will rise, and collateral will flow to assets with yields. Independent stakers and small-to-medium operators will be squeezed out first. The native yield attribute is the core reason institutions choose ETH over BTC. Erasing this is equivalent to voluntarily giving up the competitive advantage when ETH outperforms BTC.
Supporters say staking centralization risk is becoming more serious; opponents say you are destroying ETH's narrative as an income-generating asset. Both sides have valid points. This proposal is still in draft form but has already sparked the largest economic model controversy since The Merge.
The market has not yet started pricing this in, but once it enters formal discussion, ETH's long-term valuation logic may be rewritten. $ETH The Glamsterdam upgrade has been postponed to Q4. This is the largest protocol overhaul for Ethereum since The Merge.
This upgrade has two core proposals. The first is ePBS (EIP-7732), which directly incorporates proposer-builder separation into the protocol. Currently, Ethereum block production requires off-chain third-party relays to match proposers and builders; ePBS aims to make this division of labor a built-in protocol rule. At the same time, it extends the validation window from 2 seconds to 9 seconds, significantly increasing the network's data capacity.
The second is BALs (EIP-7928), block-level access lists. The current processing method is like shopping blindfolded in a supermarket, confirming the next item only after touching one, so transactions must be processed sequentially. BALs are like preparing a shopping list in advance—identifying which accounts and contracts will be used. The system can then foresee which transactions do not interfere with each other and can be processed in parallel. In theory, this can greatly improve L1 throughput.
On August 11, Glamsterdam Devnet 8 launched as a temporary public testnet, allowing external node operators and validators to participate. The technical work is progressing, but market attention is currently elsewhere. It will get its turn once macro sentiment improves. Technical upgrades rarely drive prices alone during a bear market, but they determine Ethereum's form years from now. $ETH On August 13th, the Ethereum spot ETF saw a net inflow of $6.71 million, with Grayscale Ethereum Mini Trust contributing $6.47 million.
BlackRock's ETHA, on the other hand, had a net outflow of $560,000.
BlackRock is selling, Grayscale is buying, the directions are clashing.
Overall, ETFs still had a net inflow over the past week, but the scale of inflows is noticeably smaller compared to July. The total net asset value of the Ethereum spot ETF is $10.57 billion, with a net asset ratio accounting for 4.64% of Ethereum's total market cap. The pool is still growing, but the pace is slowing down. ETF data is improving, but the speed of improvement is not fast enough.
There is also a new signal on-chain. The number of Ethereum validators officially surpassed 1 million, but truly independent individual validators only make up a small proportion; the vast majority of nodes are hosted by large pools like Lido and Coinbase. The top five entities control over 55% of staked ETH. At the beginning of the year, this figure was 48%, rising by 7 percentage points in half a year. Validator numbers are increasing, but decentralization is decreasing. The more concentrated the nodes, the more vulnerable the network becomes. Wider acceptance of staking is good, but if it ultimately concentrates in the hands of a few entities, the fundamental logic of ETH needs to be reexamined. A single regulatory document can impact these major entities, and whether this will trigger a chain reaction in staked ETH is unknown. $ETH A few days ago, there was a technical discussion about the Glamsterdam upgrade, and I went over the transcript of the developer meeting speech several times.
The speaker mentioned that they are considering further increasing the Gas limit on the Ethereum mainnet, with the current discussion range being between "200 million to 400 million."
The current Gas limit is 60 million; 200 million is more than three times that, and 400 million is more than six times.
Each time the Gas limit is raised, the mainnet's processing capacity jumps significantly. If the Gas limit is increased to 400 million, the L1 throughput itself would improve enough to compete with some L2s.
This proposal directly points to a core issue—whether Ethereum should let L2 handle all transaction volume or enable L1 to regain the ability to process large-scale transactions. Once the mainnet Gas limit is significantly increased, the "necessity" of L2 will be questioned.
This technical route choice is more worthy of serious discussion than any short-term price fluctuations.
Especially against the backdrop of L2 tokens declining across the board and ETH mainnet revenue accounting for less than 5%, mainnet scaling is one of the most direct ways to address revenue loss.
But direct scaling also means the logic for L2's existence is weakened—if L1 becomes fast enough, L2's competitive advantage will narrow. $ETH EIP-8363 was proposed to be removed from the upgrade on the 6th, and then it sparked a week-long debate on X.
On August 13th, the proposal author said in a podcast, "If validators are unwilling to accept economic model adjustments, it means we need to rethink who is actually maintaining Ethereum's decentralization in governance."
This statement directed the debate to a more core issue — validators are the executors of network security and the direct beneficiaries of the economic model. If validators oppose any adjustments that are unfavorable to their own earnings, then all governance will become "validator supremacy." The founder of Aave opposed this proposal because it would affect the benchmark interest rate in the on-chain lending market. The CEO of SharpLink opposed it because their staking business would be directly harmed. The CEO of EtherFi opposed it because their LSD protocol would lose a large portion of underlying assets. The reasons for stakeholders' opposition vary, but the direction is consistent.
Both sides act in the name of protecting Ethereum, but their interests differ. Ethereum's governance is shifting from a technical issue to a matter of interest coordination. This process will repeatedly play out in every upcoming upgrade. $ETH As of August 15, the total assets under management of Ethereum spot ETFs have reached $10.57 billion.
In July, ETFs increased their holdings by approximately 136,500 ETH, and in the first week of August, they added about 53,000 more.
Fidelity has submitted an application to the SEC to add staking functionality to its FETH Ethereum fund. If approved, the ETH ETF would become a "yield-bearing" product, differentiating itself from BTC ETFs that only profit from price differences. This means investors buying ETH through the ETF can receive staking rewards without needing to run nodes themselves or use third-party staking services. For compliant funds, staking rewards represent an additional source of return, which will change the capital attraction of ETH ETFs.
Net inflows into ETFs have continued from July until now, and staking lock-ups are ongoing. Whether selling pressure still dominates short-term prices needs to be assessed in conjunction with market structure. $ETH ETH open interest contracts dropped to 13.3 million ETH this week, the lowest level since early May.
Futures market participation is declining, with fewer people willing to bet on ETH.
However, options open interest hit a record high of $8.11 billion, three times that of three months ago.
Futures are cooling down, options are heating up, and the market is shifting from "active trading" to "hedging protection."
Interestingly, ETH rose to $1901, but derivative participation actually decreased. The price rebounded, yet fewer people are involved. This indicates the rebound was not driven by new money entering the market, but by short covering combined with holders reluctant to sell.
The derivatives market cooling off is not inherently a bullish or bearish signal, but at least it shows market participants have become cautious. Trading volume is shrinking, open interest is falling, and fewer people are willing to bet at this level. Everyone is waiting for a clear direction.
$ETH Fidelity has submitted an application to the SEC to add staking functionality to its FETH Ethereum fund.
If approved, the ETF can directly stake ETH to earn yield, then distribute cash dividends to holders each quarter.
Grayscale has also changed its rules, setting ETH staking as the default operation.
Buying a BTC ETF only earns price difference, but buying an ETH ETF can also earn interest. A 2.6% annualized yield isn't high, but in an environment with a 4.7% risk-free rate, having an additional source of income is another selling point.
However, the SEC is also re-evaluating the classification of ETH and XRP. On one hand, they are promoting staking ETFs; on the other, they are reviewing classifications. Both processes are happening simultaneously, so the short-term direction remains unclear.
$BTC $ETH In Q2, the Ethereum application layer generated $1.79 billion in fees, but the mainnet only captured $88.4 million, accounting for less than 5%. L2 processes 1,270 user operations per second, while the mainnet only handles 20.4.
The busier L2 gets, the fewer fees ETH receives. Of the $1.79 billion ecosystem revenue, ETH only gets less than 5%, with the rest all taken by L2. Layer 2 TVL has also dropped to $5 billion, the lowest since 2023, significantly down from the 2024 peak of $35 billion.
ETH's scaling path relies on L2, but as L2 thrives, the value captured by the mainnet is actually shrinking. If L2 continues to absorb most of the revenue, ETH's long-term value capture logic will need to be reconsidered. $ETH ETH ETF structure → holding ETH → staking → generating additional yield → increasing attractiveness to institutional investors.
The cumulative net inflow of US ETH ETFs remains at about 11.46 billion USD.
ETH reserves on exchanges are about 15.12 million ETH, down from about 16.86 million ETH at the beginning of the year, equivalent to a decrease of about 1.74 million ETH.
I assess ETH currently as neutral with a slight bullish bias, but a new upward wave has not yet been confirmed. The brothers who chased in a couple of days ago must be very nervous, right?
This $CAP coin has multiplied several times in a few days, and the funding rate is now stuck at -0.94%.
With such a high holding fee rate, even if it doesn't drop, your money will still flow out.
I placed a short order at 0.09, waiting for it to spike.
In this kind of negative funding rate market, shorts are crowded. If the main force wants to harvest, they will most likely first pull a spike to blow out the shorts, then turn downwards.
The 0.09 level is reserved for that spike.
What's more interesting is that despite the negative funding rate, the long-short ratio is almost balanced.
The whole network's 24-hour long-short ratio is 1.0362, almost 1:1.
But when you break it down, you find some tricks: Binance account long-short ratio is only 0.7627, with more short accounts than long accounts;
However, Binance whales' long-short ratio is as high as 1.6556.
This indicates that whales are secretly increasing their long positions.
OKX is even more exaggerated, with account long-short ratio soaring to 1.21, and perpetual contract longs accounting for 77.12%.
On one side, retail investors are shorting; on the other, whales are betting long.
I think this is the scenario shorts fear the most.
Retail investors are desperately shorting, whales are quietly adding longs, and the negative funding rate is forcing you to pay rent every day.
If you cut losses, what if the spike hits, stops you out, and then reverses? You'll regret it badly.
If you don't cut losses, you lose money every day, and your account slowly drains like a leak.
Shorts now feel like they're being held underwater, trying to surface for air, but there's always a foot pressing on your head.
To be honest, my short order at 0.09 is not a blind short; I'm waiting to act after that spike.
If the main force wants to blow out shorts, they will definitely first pull it near 0.09 to clear out you shorts, then truly start smashing down.
So I placed my order at 0.09, waiting for it to fill me first, then follow the main force to catch the pullback.
This is my strategy: not to tough it out at the spike tip, but to pick up the leftovers after the spike.
Retail investors fear the spike; I like the spike.
The fiercer the spike, the more meat there is afterward.
With CAP's structure, that 0.09 spike will most likely come; it depends on whether you can hold on until then.
Order placed, waiting for it to spike.
$BTC
$ETH
#消费动能转弱，9月政策仍受通胀制约 华尔街最大银行正在用真金白银绘制一幅加密资产配置版图。 资产管理规模约5.1万亿美元的摩根大通，在向美国SEC提交的二季度13F文件中披露了其加密相关ETF持仓的显著调整。 比特币：增持25%，期权结构转向看涨 截至6月30日，摩根大通持有贝莱德IBIT约1040万股，价值约3.557亿美元，较一季度的约830万股（近1.62亿美元）增持约25%。 更值得注意的是期权结构的变化： 看涨期权：增至约394万份 看跌期权：由约475万份降至约350万份 看涨期权增加、看跌期权减少，表明摩根大通正在减少对比特币下行风险的对冲，持仓态度更趋积极。 以太坊：暴增338%，但体量仍远不及比特币 摩根大通持有贝莱德ETHA近117万股，价值约1430万美元，较上一季度大增338%。 虽然以太坊持仓增速远超比特币，但其仓位价值仍不足比特币的二十分之一。这表明摩根大通将比特币视为核心配置，以太坊则更像是对生态的补充性布局。 XRP：一季度清仓后重新建仓 一季度将Bitwise XRP ETF持仓归零后，摩根大通在二季度重新建立了XRP敞口： Bitwise XRP ETF：113股，价值1,356美元 #消费动能转弱，9月政策仍受通胀制约
The market originally only expected US July retail month-over-month to be +0.1%, but the actual figure came out at -0.6%—the largest drop since May last year, with a forecast miss of nearly 0.7 percentage points. This "terrifying data" directly contradicted the narrative of "consumer resilience."
Logically, a retail collapse should lead to rising expectations of rate cuts and a surge in risk assets. But the majestic part this time is that in the same week, the University of Michigan's preliminary August consumer sentiment also fell to 51 (previous 55.2, expected 55), while the one-year inflation expectation actually rose to 4.3% (expected 4.2%). Consumption is retreating, inflation expectations are rising—a typical "stagflation signal," making September's policy even harder to navigate, caught in the middle.
CME data has already moved: the probability of the Fed keeping rates unchanged in September is 69.4%, while the chance of a 25BP hike dropped to 30.6%. Traders are also reducing bets on more than one rate hike before mid-2027. The US dollar index briefly fell to its lowest since May, potentially closing down six out of seven weeks. On the surface, this looks like "cooling rate hike expectations" pushing the dollar weaker, but what is really being priced in is that the market is starting to doubt this tightening cycle has ended, and that the next move is more likely to "hold steady" rather than continue hiking.
But the Fed itself hasn't closed the door. Harker clearly said he is "not confident inflation will continue to improve and that inflation needs to be pushed back to 2% faster," while Goolsbee said he wants to "wait another three to four months and only be confident inflation is returning to 2% if June's momentum continues," warning that "if retail sales keep declining, it will cause concern because consumption is a key pillar of the US economy." Together, their attitudes serve as the official footnote to this topic: weak consumption is worrying for the economy, inflation not falling means no easing—the September meeting will likely be about "how long to hold" rather than "raise or cut."
Interestingly, Goolsbee also noted that "if AI-driven growth cannot be sustained, the narrative around AI and monetary policy will need to be reassessed." This is the first time the AI capital expenditure narrative has been formally linked to monetary policy—which is why recently Fed officials have started to care about financial stability issues like "whether AI is in a bubble" and "large leverage used to buy Treasuries." While AI infrastructure stocks like SanDisk and Coreweave have surged for days, the policy side is already sounding a warning in case the "AI narrative loosens."
On the OKX front, it's very quiet: BTC is at $63,066, about -0.46% in 24 hours, with perpetual funding rate +0.01%. US stocks are pricing in a "peak rate hike cycle," but crypto is sideways—indicating the market is willing to pay for certainty assets like AI/semiconductors but hasn't yet translated "rising rate cut expectations" into broad risk appetite. Retail collapsed, but money hasn't fled.
So the real hook of this main storyline isn't "whether to hike in September"—that is basically priced in. What’s really worth watching is: with consumption continuing to weaken and the AI narrative cooling, will the underlying assumption of an "economic soft landing" be torn open before September? What do you think is more dangerous in the next month: retail continuing to deteriorate or the AI capital expenditure narrative loosening first?
#RetailSales #FederalReserve #ConsumerConfidence September rate hike probability falls below 40%: Is the liquidity inflection point for $BTC here?
Last night, the U.S. Department of Commerce released data: July retail sales month-over-month -0.6%, while the market expected +0.1%.
June still showed positive growth of 0.2%, but in just one month, it reversed completely.
Consumption accounts for 70% of the U.S. GDP; if this collapses, the entire economic narrative needs to be rewritten.
On the same day, the University of Michigan's preliminary August consumer confidence index was 51.0, expected 54.5, down from 55.2 in July. This is the first decline in three months, a month-over-month drop of 7.6%.
Americans not only have no money to spend, but they've also lost confidence in "having money to spend in the future."
Let's lay out the cards from the past week:
July CPI year-over-year 3.4%, lower than the previous 3.5%, core CPI year-over-year dropped to 2.5%. Inflation is cooling down.
July PPI month-over-month 0%, expected 0.2%. Producer prices are flat.
July nonfarm payrolls decreased by 23,000, expected an increase of 80,000. May and June data were cumulatively revised down by 103,000.
Four arrows fired simultaneously: CPI cooling + PPI flat + nonfarm collapse + retail plunge.
Consumption stalled, employment collapsed, prices stopped rising—what reason does the Fed have to continue raising rates?
On August 5, CME FedWatch showed a 58.4% chance of a rate hike in September.
August 7 nonfarm data dropped it to 55%.
August 12 CPI data dropped it to 48%.
August 13 PPI data dropped it to 38%.
In one week, the rate hike probability fell from 58% to 38%, a 35% discount. The probability of maintaining the current rate has risen to 59.9%.
One data point after another is dismantling the hawkish fortress brick by brick.
On August 14, BTC fell back to $62,773. Still hovering around $60,000.
QCP Capital bluntly stated: geopolitical risks, high oil prices, and global liquidity uncertainty—these macro headwinds have outweighed all the positive economic data.
In other words: it didn’t rise when it should have.
The rate hike boot is being pulled back, the liquidity inflection point is coming—but BTC just won’t fly.
The rate hike probability dropped from 58% to 38%, yet $BTC is still hovering around $60,000.
Either the market is wrong, or something bigger is brewing.
Personally, I lean toward the latter. #Consumption momentum weakens, September policies still constrained by inflation, but this "inflation" is not what you think.
July CPI was only 0.5%, half of May's 1.2%. Food prices fell 1.5%, pork down 13.3% year-on-year. This is not inflation heat, it's clearly demand cold. According to textbooks, such numbers should have triggered easing already.
But will there really be massive easing in September? Unlikely.
The bottleneck is not CPI, but PPI. July PPI rose 3.5% year-on-year, purchase prices up 5.5%—upstream has already turned positive, still the same imported pressure from crude oil and bulk commodities. If you aggressively stimulate demand now, PPI will keep rising, costs will pass downstream, and that little dip in CPI will be filled instantly. Not to mention the exchange rate is still pegged to the Fed; if they don't ease, do you dare to sharply cut unilaterally?
So September will most likely be a slow squeeze: targeted drip policies like consumption vouchers, trade-in programs, equipment upgrades will happen; there is room for interest rate and reserve requirement cuts but restrained. Don't expect a big wave to boost consumption.
The real problem is not the lack of tools, but broken transmission. People aren't refraining from spending because rates are high, but because income and confidence haven't recovered; money goes first to debt repayment and savings. Policy pours water into one end of the pipe, but it doesn't come out the other.
There will be moves in September, but they aim to "prevent further deterioration," not "immediate improvement." For consumption to truly pick up, household income must move first—that chain is much longer than just easing.