Orbit Post Sitemap
Can't predict ETF rises and falls? DWF Labs: Fund flows are more like a "rearview mirror" of the market!
A set of data directly shatters the illusion:
📉 Out of 8 times BTC dropped more than 15%, 7 times ETF funds had net outflows!
📈 Out of 9 times BTC rose more than 15%, ETF funds all had net inflows!
But the real key is:
• 0.54: Correlation coefficient between ETF fund flows and BTC's past 5-day performance.
• 0.03: Correlation coefficient between ETF fund flows and BTC's next 5-day performance, almost no correlation.
In other words, funds only chase in after the market rises; funds only withdraw after the market falls. ETF fund flows are more like a trend confirmation indicator, not a leading indicator.
Look at how severe the two rounds of sell-offs were:
💸 From October 2025 to February 2026: net outflow of 6.9 billion USD.
💸 Then from May to June: another outflow of 8.2 billion USD.
A total withdrawal of 15.1 billion USD in two rounds!
However, the market has also changed: the number of times BTC experiences a drawdown of more than 15% per year has dropped from about 8 times in 2021 to about 3 times per year after the ETF launch.
Conclusion: Don't take ETF net inflows as a signal of price increase, nor take net outflows directly as a bottom-fishing signal. Fund flows may just follow prices, not lead them.Justin Sun’s accumulation of $PEPE has become the week’s most instructive meme trade. Starting on the 18th, the token climbed 22% in three days without any public call to buy. That silence is the story. $DOGE, despite Elon Musk’s megaphone, could not move, and $TRUMP, despite a party thrown by Donald Trump himself, could not rally. Attention is no longer the scarce input; conviction among large holders is. The structure explains the outperformance. PEPE’s supply is unusually concentrated, with t【KOSPI Drops Over 6% in Two Days, Is the AI Chip Rally Fizzling Out?】
The South Korean stock market has taken a sharp hit. After a 3.29% rebound on October 21, it plunged 3.52% on the 22nd, and the decline widened to 3% after the market opened on the 23rd. The previous day's gains were almost completely erased, with SK Hynix ($SKHY) and Samsung Electronics ($SAMSUNG) leading the losses again, showing the market is clearly still unsettled.
Chip demand hasn't suddenly disappeared. South Korea's semiconductor exports in the first 20 days of October rose 222.5% year-over-year, and industry data remains strong. What investors worry about is something else: with AI giants pouring so much capital expenditure, how much profit can they actually recoup in the future? When high expectations meet return anxieties, even strong earnings may not be enough to support stock prices.
What needs attention is that Hynix's 40 trillion KRW buyback plan has ended, removing a key support for the stock. The upcoming earnings report on October 28 will be a critical test: if profits and future guidance are strong enough, there’s a chance to stabilize sentiment; if positive news can’t save the stock price, this round of decline may be more than just short-term profit-taking.Overnight U.S. stock market closing trends showed severe divergence, with the Dow Jones nearly flat with a slight rise, while the Nasdaq suffered a heavy decline.
Dow Jones: +0.02%, closed at 52867.98
S&P 500: -1.21%, closed at 7663.43
Nasdaq 100 fell 2.9%, marking the largest single-day drop since June
AI chips and memory sectors were collectively sold off, with the Philadelphia Semiconductor Index plummeting 6.72%. SpaceX dropped over 9%, AMD, SK海力士, and SanDisk fell over 8%, Meta nearly 7%, Nvidia, Broadcom, and Tesla also plunged; only Apple bucked the trend with gains. Funds fled from overvalued tech stocks, shifting to defensive consumer sectors, with Procter & Gamble, McDonald's, Coca-Cola, and other essential consumer goods surging.
The core reason behind this: during earnings season, the market began to worry that AI capital expenditures would fall short of expectations, compounded by rising oil prices causing inflation rebound pressure, leading to a concentrated valuation sell-off in growth stocks.
$BTC $SNDK $SKHYNIX
#SK海力士Q3财报在即，HBM4能否提振盈利？ Saudi Arabia was hit by Houthi drones, Brent crude reached a new high of $113 this year, so inflation expectations have returned, and the probability of a rate hike before the end of the year has risen to 87%.
The crypto market got hit along with the Nasdaq, and the liquidation direction has also reversed. Last night, out of $465 million in liquidations, $427 million were long positions, with 103,000 people liquidated.
The ETF situation is even more ironic: the $BTC spot ETF saw inflows of $1.07 billion over three days, with $528 million on Wednesday alone, yet the price dropped from 85,000 to 81,000.
Money is coming in but can't push the price up, indicating that sell orders are more urgent than buy orders. Two key levels for BTC to watch: the mid-term holders' cost is 86,700, a major resistance level; below that, 80,000 is the cost zone for ETFs and treasury companies, and if this zone is lost, it will likely head straight to 75,000.
The $ETH ETF finally saw some return money, with a net inflow of $14.9 million yesterday. Technically, RSI is 35.4, MA20 at 2495 is pressing from above, and the 200-day moving average at 2302 is the last line of defense. Today's low of 2360 has already touched the doorstep.
For $ZEC, out of the $13 million liquidations last night, shorts accounted for 76%, but in the last hour, it reversed with longs making up 95%. Even more intense, a giant whale short took a $35 million loss to close out.
RSI is 44.9, MA20 at 1222 is right underfoot, and MA30 at 1302 is the ceiling.Q3 book profits have a good chance, but the stock price's winning point is not in the current profits, but in the HBM4 ramp-up guidance + capital expenditure.
✅ Bullish logic
1. AI inference demand drives HBM4 shipments; HBM unit price is high, and gross margin is much higher than ordinary DRAM/NAND, making it the main profit engine; traditional consumer memory prices are recovering, providing a basic support floor.
2. Long-term contracts with top customers like Nvidia lock in orders, with strong order certainty; market expects Q3 profits to potentially hit new highs.
⚠️ Reasons for today's decline
1. Biggest variable for HBM4: Nvidia's new chip Rubin delayed, slow customer certification and tuning, dragging down HBM4 volume ramp-up pace; yield ramp-up below expectations is the biggest risk; Samsung's HBM capacity catch-up increases share pressure.
2. Capital expenditure dilemma: raising expansion budget suppresses free cash flow; conservative expansion raises concerns about losing HBM market share.
3. Ordinary memory (PC, mobile DRAM) recovery is weak, unable to support valuation alone, considered a "dragging business."
📌 Key financial report points to watch:
① Actual HBM4 shipment volume, yield, ASP (selling price) guidance
② Capital expenditure plans for second half of 2026 & next year
③ Updates on customer long-term contracts and order renewals
#SK海力士Q3财报在即，HBM4能否提振盈利？ 🔥Key on-chain signals! These BTC holders are basically no longer dumping⚠️
In the past 30 days, only 4.4% of coins from native Bitcoin holders in the 6-12 month cycle have been transferred to exchanges.
Simply put, most people who entered at high levels earlier have mostly stopped selling, relieving selling pressure.
But for the market to rally, new funds need to come in to take over; currently, new buyers have yet to appear.
With less selling pressure but no incremental funds, do you think BTC will first consolidate or start to rise?
❗This is only an on-chain data share and does not constitute any trading advice
#机构资金转弱，BTC与ETH谁先迎来回流？ #SK海力士Q3财报在即，HBM4能否提振盈利？
【Profit hitting new highs is not enough, SK Hynix must prove HBM4 can continue to generate profits】
SK Hynix ($SKHY) will release its Q3 earnings report on October 28. What the market really wants to see is no longer just whether profits can hit new highs. The company posted Q2 revenue of 79.32 trillion KRW and operating profit of 60.54 trillion KRW, demonstrating remarkable profitability. The continuous rush for high-end storage in AI servers has indeed given it the confidence to raise prices and make money.
HBM4 is the next big battle. Samsung ($SAMSUNG) announced mass production and shipments back in February, and SK Hynix is also advancing delivery of its next-generation products. The competition is shifting from technology showcases to battles over orders and market share.
The market currently expects Q3 revenue to be about 100 trillion KRW, a roughly 26% increase over Q2. This means that even if the earnings report hits new highs again, it may not be enough to push the stock price higher, since high growth is already priced in. What can truly encourage bulls to increase their positions is HBM4 shipment ramp-up, yield, and follow-up orders exceeding expectations.
The key focus of this earnings report is whether profits exceed expectations and management’s outlook on demand and competition in 2027. If profits hit new highs and HBM4 orders remain strong, the rally has reason to continue; if it only delivers a solid report that meets expectations, the risk of the positive sentiment being fully priced in cannot be ignored.On October 28, the Federal Reserve will hold a meeting.
The coffin shop has already opened early.
In the past 24 hours, the entire network liquidated $221 million, with long positions liquidated at $207 million, accounting for 93.7%. BTC was smashed from $84,789 down to $81,331, losing $3,460 in one day.
Before Waller even spoke, those chasing longs were already carried away.
---
Then you open CME FedWatch and see a different picture.
The probability of maintaining the interest rate in October exceeds 85%. The probability of a rate hike is less than 15%. And before Waller's speech at the IMF Annual Meeting in Bangkok on October 16, this probability was around 27%.
What did Waller say? Almost nothing.
"I will not predict the outcome of the meeting, and I don't think I should say what actions I might take at the remaining meetings this year." — This was his exact wording. Goldman Sachs directly interpreted this as: no change in October, next rate hike in December.
But the most critical thing is not that he "dove." The most critical thing is that he changed the rules of the game.
Waller clearly announced: The Federal Reserve is entering a new phase of policy communication, no longer revealing decision tendencies to the market in advance. This means — in the Powell era, the market could still parse wording and price in advance; now Waller has closed that door. You can only wait for the decision to come out, with no preview.
#10月FOMC临近，美联储暂停加息还有哪些变数？ Western Digital Earnings Report: Storage Demand Remains Strong, Market Begins to Worry How Long Profits Can Last
Western Digital's earnings report sends mostly positive signals for the storage industry: this quarter's revenue was $4.381 billion, a 55% year-over-year increase; adjusted gross margin was 58.1%, with next quarter expected to rise to 60%–61%. At least based on the company's guidance, near-term demand and profitability show no obvious weakening.
What’s truly worth watching is the growth structure. This quarter, delivered capacity grew about 22% year-over-year, but revenue grew 55%, indicating significant contributions from improved pricing and product mix. Storage manufacturers are making money, but this profitability increasingly depends on pricing power.
This explains the market divergence: orders are strong now, but will new supply in the future suppress prices? Recent concerns around Toshiba's capacity expansion have already affected hard drive stocks. However, the expansion targets are still some way from actual delivery, so it cannot be directly concluded that supply and demand have reversed.
For Seagate, this is a relatively direct validation of industry health; for SanDisk, it signals that data center storage demand remains strong, and HDD price increases cannot be directly applied to NAND. Whether SanDisk can continue to raise profitability depends on whether enterprise SSD revenue, NAND prices, and shipment volumes can all be realized together.
My judgment is: this earnings report supports continued attention to storage bull opportunities but is not sufficient reason for an immediate bottom-fishing. The most important thing going forward is to observe whether the stock price can stop falling after strong guidance and whether peer earnings reports continue to confirm demand.
If capacity growth, firm prices, and profit realization occur, then a pullback will be more worthy of study; #存储芯片股反弹，AI扩产与竞争并存 $BTC and $ETH: A Short-Term Breather Amid Expectations of No Rate Hike
With the midterm elections approaching, the Federal Reserve is clearly hesitant about raising interest rates. Although CPI remains elevated, employment and real estate data have shown signs of weakness, and the market currently prices in over an 80% probability of holding rates steady. A hard rate hike would hurt both the economy and political support, increasing the likelihood of a policy shift toward a wait-and-see approach.
On the charts, BTC struggles repeatedly around 82,000, while ETH has weakened further after breaking below 2,410, showing a clear high-beta characteristic. If a no rate hike is confirmed in October, it would be like a short-term lifeline for the long-dry market: BTC is expected to stabilize above 80,000, and ETH could find support near 2,400.
However, it is important to stay clear-headed: persistently high rates without cuts still act like a dull knife cutting into the market, and without fresh large capital inflows, it will be difficult for the two major mainstream coins to stage a unilateral major reversal just from a pause in rate hikes. A short-term breather is expected, but a trend reversal will require more incremental signals.
#10月FOMC临近，美联储暂停加息还有哪些变数？ #特斯拉Q3汽车毛利率回升，自由现金流转正 X Layer has been quite active recently, but I'm more concerned about whether OKB can truly benefit.
🔍 Recently focusing on OKB, I found that X Layer is trying to change its development direction.
On October 13, the official announcement introduced a trading competition centered around community-certified tokens, planned to start on October 19, aiming to increase trading participation and liquidity of ecosystem assets.
This seems like just an event, but combined with the previously launched Exchange OS, the underlying idea is worth noting.
📊 Exchange OS allows developers to deploy spot, perpetual contract, and other trading markets, and participate in market operations by staking OKB.
I believe X Layer is trying to shift from simply attracting projects to settle in, to building a more complete on-chain trading ecosystem.
But here’s a question: can the short-term hype from trading activities retain long-term users?
⚠️ For OKB, ecosystem expansion is certainly promising, but with extremely low on-chain gas fees, increased trading volume may not directly translate into sufficiently large token demand.
Next, I’m more focused on three practical changes: whether active addresses can continue to grow, whether on-chain liquidity can be solidified, and whether Exchange OS can attract real trading markets.
X Layer is not lacking development plans; what it needs now are results that can be continuously validated.
$OKB
#XLayer #OKB #欧易 #PublicChainEcosystemSNDK dropped from 1409.5 to 1225.7, yet the perpetual positions increased by nearly 30% in one day. The bottom-fishing crowd here looks more and more suspicious to me.
I've written about this storage group twice before; the first two times focusing on price, this time on positions.
Actually, the whole group is deleveraging: WDC down 14%, STX down 10%, MU up only 4%.
Only SNDK is going the opposite way, with positions pushed from 75 million to 96 million.
That 11% strong rebound candle on Monday, not even fully covered in four days, now the price is slightly lower than before the rebound.
Long accounts increased from 78% on Wednesday by nearly 6 points.
To be clear, the fee rate was still zero the day before yesterday, 3.1 basis points this morning, now 3.4 for this period, annualized close to 40%.
My first reaction was someone was rushing the rebound, but seeing the fee rate changed my mind.
Anyone willing to rent chips at this price is probably blinded by losses.
This situation doesn't add up.
Last night, three hours after the US market opened, volume was 530 million, price fell from 1345 to 1281.
I won't catch this falling knife, nor will I short against the trend.
Eighty percent are crowded on the long side; a price shake can squeeze a wave. Let's see after tonight.
If it stands back at 1385 with no position shrinkage, I'll admit I was wrong; if it breaks below 1225.7, look down to 1216; let's wait for the US market close tonight.
Do you think this is bottom-fishing, or doubling down to bet on a rebound?
$SNDK $WDC $STX
#StorageChipStocksRebound, AIExpansionAndCompetitionCoexist
#TradingVoice: YourExperienceDeservesToBeHeard油价这根引线，正在点燃一场科技股多头的耐力测试。$BZ持续走高推升通胀预期，市场对降息的押注一拖再拖，美债收益率顺势上行，高估值成长板块首当其冲——AI与存储链条上的$SNDK、$SOXL同步承压，账户浮亏扩散，但买方阵营仍未举白旗。 这轮下跌的核心不是基本面塌方，而是宏观传导的连锁反应。HBM长单与存储需求的故事还在，可流动性预期一旦收紧，短期估值就只能被摁住。很多人把眼前的下挫定义为"宏观误伤"，拒绝在低位交出筹码，甚至越跌越买。问题在于，这种密集的买方持仓是典型的两面刃：行业尚无明确止跌信号，信仰越集中，后续的弹性与脆弱性就越同时放大。 真正被低估的变量，是即将到来的财报窗口。若Hynix业绩超预期、指引兑现，这批死扛的资金会瞬间从孤军变成合力，推动一轮像样的修复行情；反之，一旦财报或指引不及预期，浮亏中的信心极易集体瓦解，集中止损会把下跌斜率拉得更陡。换句话说，当前不是抄底信号，而是买方信仰与宏观利空的正面对撞。 我的判断是：市场把注意力放在了油价和利率上，却低估了财报作为"信心开关"的杠杆效应——它决定的不是估值贵不贵，而是这批被套多头还能不能继续扛。亏损仍在扩大，抵抗是否有Brothers, let's take a look at these three US stocks that come with built-in traffic. Don't just rush in by looking at the ticker symbols; the underlying logic of these three is worlds apart, mixing them up will get you hurt.
$TSLA around 386, the earnings report is the trump card, FSD is the engine.
Q3 earnings just came out: revenue 28.05 billion, EPS 48 cents, both beating expectations. Delivered 486,000 vehicles, undelivered orders hit a nearly four-year high, demand is indeed strong. But what really supports the valuation isn't car sales, it's FSD. Musk was very straightforward on the call: "FSD currently has no competitors," multiple European countries have approved it, waiting for EU-wide approval.
$SPCX around 145, the IPO halo has faded, the merger is the real variable.
IPO priced at 135 in June, raised 75 billion, valuation 1.77 trillion, the largest IPO in history. But the post-listing performance is honest: on October 6 it was still at 172, by October 22 it had dropped to 144.93, a 9% single-day drop. Why? The IPO premium is retreating.
$SNDK around 1365, contracts provide a floor, but cyclical nature is unavoidable.
I've talked about SanDisk's logic before: 93.9 billion in long-term contracts lock in the base position. But look at the latest earnings data: Q4 revenue 8.97 billion, a year-over-year surge of 372%, net profit 6.9 billion, data center business growth 437%. The problem is, two-thirds of this growth comes from price increases, not pure shipment volume. Once NAND prices fall back, the valuationAfter a night's sleep, my account lost three points—negative.
$QQQ dropped 2.86% last night, marking the largest single-day decline for the Nasdaq since June. Strangely, the drop wasn't due to poor earnings but because the earnings were too good. $META's revenue rose 26%, beating expectations, yet its stock price fell 7%. The issue lies in the other half: it simultaneously raised capital expenditures, turning free cash flow negative by 2.1 billion. The market is pricing in "burning money" for the first time—not because you're earning less, but because you're borrowing too much.
What really ignited things was Jensen Huang's statement: demand for computing power is very strong, but capital is scarce. Sellers of shovels say money isn't enough, and buyers of shovels realize this AI wave isn't driven by profits but by financing.
On the same day, Apple rose 1.23%, hitting a record high because it doesn't burn money, it only collects money.
Now, neither cutting nor adding. The position isn't high, but the direction is unclear—before the October 28 interest rate decision, this kind of valuation kill shows no mercy.
#MetaQ3营收超预期，法律费用与AI支出承压 #10月FOMC临近，美联储暂停加息还有哪些变数？
"Federal Reserve Pause Changes, Bitcoin Plunges $3,000"
The Federal Reserve is set to hold its policy meeting next week, and the previously widespread expectation of a pause in rate hikes has suddenly loosened in recent days.
September inflation data unexpectedly rebounded, and Fed officials refused to commit to any rate cut path before the blackout period.
Risk-off positions retreated, with Bitcoin dropping more than 3% in a single day, plunging nearly $3,000 from the $84,000 high.
In an extreme scenario, if year-end rate hike expectations revive, the $80,000 long position defense line will face a liquidation test.
Today’s market is consolidating with low volume around $81,300, all waiting for the rate decision on the 28th to drop the shoe. $BTC 10.23 Chen Jie Morning Analysis
On Friday, focus on the final US consumer confidence, preliminary US services PMI, and UK and Canada retail data. Official reserves remain +23.02 tons, institutional holdings sharply decreased by 11.97 tons in a single day, main force chips concentratedly exiting. Opened today at 4029, early session was blocked at 4032 then continued downward, intraday lost the 4020 level and touched 4019, currently around 4021. Previous bottom defense line completely broken, all cycle moving averages diverging downward, current bears hold absolute advantage.
Daily and 4-hour: Daily candle closed below lower band at 4047, middle band moved down to 4136, large-scale downside space points to previous low at 3943; 4-hour broke below lower band at 4031, middle band at 4104 forming resistance.
1-hour and 15-minute: 1-hour middle band moved down to 4058, lower band touched 4005; 15-minute running near lower band at 4020, middle band at 4026, upper band at 4033. After sharp short-term drop deviating from moving averages, it is not advisable to chase long directly at the low of 4021, better to wait for a rebound to break the daily lower band and then support below the 1-hour middle band for a safer entry.
Trading Strategy
Around 4040-4050 to short, target 4010-3980-3950, stop loss at 4065 #October FOMC approaching, what variables remain for the Federal Reserve to pause rate hikes? $XAU $GRVT dog whales have already fled, after buying back those 1.2 million tokens from the trash bin for months, they are desperate to promote everywhere, spending money to hire KOLs to continuously post and lure more buyers. The dog whales opened over 200,000 orders at 0.2, used this news to pump the price to 0.3, then all ran away. Over 70 million tokens will be unlocked at the end of November.
This is the most disgusting and poorest whale I have ever seen. The tricks come one after another, from airdrops, to staking events, to listing on Korean exchanges. Every time they pump the price with good news and immediately dump. Just look at the previous candlestick charts and you'll know.$xMETA fell 7% last night.
The company plans to invest $135 billion to $145 billion in capital expenditures by 2026.
However, Q3 free cash flow was negative $2.1 billion.
AI demand is indeed very strong.
But now the market is starting to seriously calculate:
With so much money invested, how much can be earned back in the future?
A good industry does not mean all stocks are worth buying at a high price.🔍 Here's the suspicious point: The bill hasn't advanced, so why hasn't the expectation of institutional participation completely cooled down? According to The Block, Morgan Stanley analyst Nikolaos Panigirtzoglou and others stated in a Wednesday report that recent actions by U.S. regulators have partially offset the negative impact of the stalled Clarity Act; however, political and legal risk premiums will persist until the crypto market structure bill passes.
An interesting detail is that regulatory actions and legislation are not the same thing. The former can provide institutions with some short-term participation space but may not replace long-term, stable rules; the stalled bill process also makes it difficult for the market to confirm whether these arrangements can be sustained.
My judgment is that short-term sentiment is buffered, but long-term uncertainty remains. For institutions, regulatory signals may reduce some of the costs of waiting; for the entire industry, political and legal risks will still affect whether funds dare to make long-term plans. Don't rush to conclusions: this does not mean institutions have already entered en masse, nor does it mean the impact of the stalled bill has been fully absorbed.
This is my speculation: regulatory measures may be providing temporary support during the legislative gap, but the real watershed will be whether predictable market structure rules can be formed later. Next, I will focus on the specific implementation of regulatory measures and whether there are new developments in the bill process. Which clue do you think is more critical? Let's watch the follow-up together and see whose speculation is closer.Three price tags, three ways to follow up
$BEAT is quoted at 0.07081, down 4%. Seven cents sounds light, but the total supply of 1 billion tokens is right there—multiplying it out gives a fully diluted valuation of $70.81 million. This is not the circulating market cap, just the "if fully unlocked" price tag. A small unit price doesn’t mean a small market cap; moving the decimal point a few places creates a unit illusion. Whether it’s expensive or not depends on showing both the price and the corresponding quantity together.
$LINK is quoted at 12.29, up 0.5%. Chainlink reserves convert revenue into LINK tokens stored, not burned. There is indeed a connection between revenue and token purchases, but evaluating this connection requires looking at new issuance, execution frequency, and revenue sources. A single transfer doesn’t indicate strength or weakness; the accumulation mechanism needs continuous records for scoring.
$JUP is quoted at 0.3223, up 3.6%. The automatic payout path first advances fees, then adds charges in the exchange, so the user’s final received amount is reduced. The steps are hidden, but the cost doesn’t disappear. It’s a fact that there is one less barrier for new users, but when comparing different entry points, focusing on the final received amount is more reliable than reading button text.
Three targets, three readings: BEAT checks the supply table, LINK examines reserve flows, JUP calculates net received amount. The price tag is just the beginning; the quantities, frequencies, and net amounts that follow are the main content.
#MetaQ3营收超预期，法律费用与AI支出承压 #SpaceX's performance surges, can the merger with Tesla be advanced? #ETF买盘持续回流，BTC为何仍未走强？
Simply put, the fundamental reason is that ETF inflows are mostly long-term allocation funds, lacking futures leverage long follow-up; existing profit-taking pressure and institutional arbitrage hedging offset the buying; combined with the unresolved FOMC rate hike expectations, macro risks suppress, and heavy overhead positions, making it difficult for the price to strengthen.
$BTC $ETH #ETF buying pressure continues to flow back, why hasn't BTC strengthened yet? 🌤️ Three directions for Friday morning: Privacy coins in regulatory awkwardness, robots moving independently, gold waiting for next week
$ZEC around 1,187, privacy coins caught between two sides 🧡
On one hand, anonymous narratives are resurging; on the other, global regulations (MiCA, GENIUS Act) are imposing compliance requirements onto the blockchain. Zcash's shielded transaction feature remains unchanged, but the position of privacy coins within compliance frameworks remains ambiguous. The 1187 level was pushed up by sector sentiment; before regulatory direction is clear, it has high volatility but low certainty. Before next week's interest rate decision, this thematic coin is more sentiment-driven.
$UNITREE around 76.81, robots moving as an independent theme 🤖
Yesterday, tech stocks weakened overall, but tokenized robot shares like Unitree jumped over 6 points, indicating it follows an independent thematic logic rather than tech sector beta. On-chain share pricing is anchored on primary financing and product delivery news. With low liquidity, news causes large swings, distinguishing thematic pulses from fundamental progress.
$XAUT around 4,131, gold is waiting for next week 🥇
XAUT is backed by physical gold bars, redeemable 24/7, with price anchored to spot gold. Before next Tuesday's interest rate decision, gold is also waiting; the opportunity cost of holding gold follows rate expectations. Around 4131 this morning, no movement—neither betting on a rate cut nor panic buying gold. Direction awaits the 10/28 statement.
This article is for market analysis only and does not constitute investment advice.
#OctoberFOMCApproaching, what else if the Fed pauses rate hikes?#Robinhood交易业务增长能否延续？
Robinhood's recent trading revenue has surged significantly, but such high growth rates are difficult to sustain long-term. The growth has ongoing potential, but volatility and regulatory risks are significant.
From the drivers' perspective, the company is shifting from a single-stock brokerage to a diversified trading ecosystem. In Q2 2026, trading-related revenue grew 44% year-over-year, with substantial increases in stock and options trading volumes. Event contract business grew tenfold year-over-year, becoming a new growth engine that offsets the decline in cryptocurrency trading revenue. Meanwhile, the platform continues to launch advanced trading tools, the AI research assistant Cortex, and community trading features to attract active retail investors, increasing per-user trading frequency and monetization levels. The rollout of financial products such as cash accounts and credit cards improves user retention and client asset deposits, providing a stable user base for the trading business. Overseas expansion and asset tokenization initiatives also open incremental opportunities beyond the U.S. domestic market.
However, constraints are also prominent. Robinhood's trading revenue heavily depends on market sentiment, and retail trading behavior is highly cyclical. Once the U.S. stock market enters a volatile or downward cycle, user trading willingness declines, and high-fee businesses like options and event contracts will see trading volumes shrink rapidly. New products like event contracts are driven by major events and represent phase-specific market trends rather than stable, perpetual growth;SNDK dropped from 1427.2, last night at 21:00 that candle looked most like a trap.
193,000 volume is 6.8 times the previous candle, pulling 1300.2 up to 1350.6 and closing at 1345, looking like a start.
At 22:00, a 131,000 volume candle smashed from 1356.5 back to 1296.7 and closed at 1304.9, with an upper shadow of 11.6, swallowing all gains and even reversing.
Five bearish candles with none recovered: 1281.3→1267.3→1237.3, at 04:00 with 56,000 volume pierced down to 1229.7.
1225.7 sideways for five candles, volume 11304→9357→10329, no one selling or buying.
MA20 at 1288.43 overhead by 48 points, can't even reach the midline.
If 1H candle closes with volume above 50,000 over 1250.4, pull back to 1245~1248 to go long, stop loss at 1237.00, target 1274.4, break to watch 1288, odds 2.9
If 1H candle closes with volume above 60,000 below 1225.7, rebound to 1233~1238 to go short, stop loss at 1247.00, target 1200, odds 3.1
Between 1225.7~1250.4 no movement, before 21:30 open is a vacuum.
$SNDK
(Not investment advice, trade at your own risk)#10月FOMC临近，美联储暂停加息还有哪些变数？ ✅ Baseline Expectations
CME FedWatch shows a 92.8% probability that the October FOMC will keep rates unchanged. The market consensus expects a pause in rate hikes this time, leaving the decision on whether to raise rates again to the December meeting after observing data.
Core contradiction: Inflation has not fully settled, but the earlier rise in U.S. Treasury yields has already tightened financial conditions spontaneously. The Fed wants to wait a bit longer for data. However, "pause" does not mean the rate hike cycle is over; there are still many uncertainties.
⚠️ Main uncertainties for this FOMC pause in rate hikes
1. Whether inflation rebound will exceed expectations (the biggest risk)
- September CPI rebounded beyond expectations, core CPI rose month-on-month, with rents, used cars, energy, and many commodity sub-items rising simultaneously, casting doubt on the sustainability of inflation decline
- Oil prices and geopolitical conflicts are the biggest external disturbances; if oil prices continue to surge, inflation expectations will rise again, making it difficult for the Fed to comfortably pause
- AI capital expenditure is pushing up electricity and equipment prices, becoming a new source of upward inflation pressure, already mentioned by several Fed officials
2. Financial conditions: Can the rise in Treasury yields continue to substitute for rate hikes?
This is the Fed's core logic currently: a sharp rise in long-term bond yields = tightening financial conditions, which can substitute for one rate hike.
- Uncertainty: If long-term bond yields quickly fall back and financing conditions loosen again, hawks will advocate for a rate hike in October#SK海力士Q3财报在即，HBM4能否提振盈利？
Next week's earnings reports are coming fast, here is my ranking 👀
Amazon (10-29): AWS has accelerated for five consecutive quarters to 37%, operating margin 39.4%, the strongest fundamental momentum among the four, my top pick.
Microsoft (10-28): Azure +40% last quarter, exceeding expectations for 8 consecutive quarters, AI annualized revenue $37B, the most stable.
SanDisk (10-29): NAND price up 235%, exceeding expectations for four consecutive quarters, but company guidance is already very aggressive, smallest expectation gap.
Google (10-28): Cloud +82% is impressive, but full-year capex nearly $200 billion, free cash flow turned negative, the most uncertain.
Ranking: Amazon > Microsoft > SanDisk > Google. Who do you favor most? 🔮 $BTC has dropped back to around 81,000, right on the average cost line of ETF buyers and coin-holding companies 👀 It fell about 3.7% in 24 hours, with mainstream coins dropping between 2% and 5%. On Wednesday night, it just surged to 84,800, but within a few hours, it gave all that back.
Why the drop? It seems more like external risks hitting all at once. After Meta and Tesla earnings reports came out, the market started worrying again about AI burning too much cash; the Nasdaq fell 2.6%, and the Philadelphia Semiconductor Index dropped 6.7%. With tensions rising in the Middle East, Brent crude oil climbed above $112; Asian stock markets were also bleak, with the CSI 300 entering a bear market. On the crypto side, the long leverage that had pushed prices up the day before was heavily liquidated, with about $470 million in 24-hour contract liquidations.
This cost line is roughly between 80,000 and 81,500; the dip on October 15 was also caught here. ETF money is actually still flowing in, with a net inflow of about $1.07 billion over three days from the 19th to the 21st.
Next, it depends on whether this area can hold: if it holds, it could be another "cost line defense"; if the daily close is below 80,000, the on-chain data point around 77,600 will probably come into discussion.As of October 23, 2026, the U.S. stock market is undergoing a highly structural sell-off: on the surface, it appears that tech stocks are dragging down the market, but at a deeper level, it reflects a market repricing of whether AI capital expenditures can be converted into free cash flow. This, combined with soaring oil prices and rising expectations of a Fed rate hike in December, has triggered a concentrated risk release through the resonance of these three factors. However, funds have not fully withdrawn but have instead undergone intense rotation among sectors. #SK海力士Q3财报在即，HBM4能否提振盈利？ As of 8:12, the total liquidations across the network in the past 24 hours reached $329 million, with long positions accounting for $308 million or 93.7%, and short positions only $20.72 million, indicating a flip in the market.
ETH dropped 3.8% to 2382, with liquidations totaling $88.86 million, longs making up 90%. Binance funding rate is -0.0015%, OKX -0.0091%, turning negative as a passive effect after the decline, thinning the short counterparty.
BTC lost 83,000, currently at 81,246, down 3.6%. Liquidations reached $138.9 million, longs accounting for 97%, shorts almost unscathed. The "shorts cleaned out too thoroughly" safety cushion has been removed, longs become the fuel, and the liquidation density of longs below 81,000 is thicker than shorts above.
SOL's positive news failed to hold. Alpenglow was activated on October 21, with final latency compressed to 150 milliseconds, but SOL failed to hold 108, dropping 3.56%, liquidations at $10.32 million, longs accounting for 94%. Early runners retreated, leveraged longs became the main liquidation force, "selling the fact."
Greed index at 64, the excess hasn't been squeezed out. BTC long-short ratio is 0.82, ETH spot ETF net outflow of $103 million, funds haven't returned, 64 seems more like emotional inertia rather than a signal of new capital.
The judgment has changed. Shorts were just cleared last night, chasing longs is risky; now longs have just been liquidated, chasing shorts isn't cheap either. No new catalyst for SOL's positive news, ETH funding rates turned negative, short crowding increased. The knife has switched sides, the new side is also risky—chasing shorts bets on continued collapse, bottom fishing bets on a V-shaped rebound, both are zero-sum games. First, see if 81,000 can hold, then decide which side to take.The crypto market fell from last night to this morning, not because of issues within the crypto space itself. 📉
On October 22, US stocks saw the Nasdaq close down 2.6%, the S&P 500 down 1.2%, with money flowing into defensive sectors like healthcare and consumer staples, and US Treasury yields dropping sharply. On the same day, Brent crude oil rose above $112, and tensions in the Middle East escalated again. $BTC followed back down to around 81,000, with $ETH and $SOL experiencing similar declines.
LMAX strategist Joel Kruger put it plainly: the period when crypto outperformed the stock market is over; now it’s moving more closely with stocks.
My view: when watching crypto prices lately, you have to first look at the Nasdaq and oil prices. Tech earnings falling short of expectations and rising oil prices lead to risk assets being sold off together, making it hard for crypto to move independently. Conversely, once US stocks stabilize, crypto prices will rebound faster.
When you watch the market now, do you also keep an eye on US stocks?
This is my personal opinion and does not constitute investment advice.08:16, the morning rebound is a bit weak. BTC pulled back a bit from the overnight low and is now hovering around 81,000, but the rebound volume hasn't kept up. ETH followed to 2,380 but couldn't hold above 2,400, rising slower than it fell. This is a typical downward consolidation, not a reversal.
My response is very realistic: if BTC's rebound doesn't surpass 81,800, it's still a bearish structure. The 82,000–82,200 range is actually a spot to look for shorts; for ETH, the corresponding level is 2,420—if it can't break above that, don't expect 2,500. There's only one scenario where I'd change my stance: if BTC rallies with volume to reclaim 82,800, that would mean the bears have run out of fuel, and then we can talk about a reversal.
Support remains at 80,000 and 78,000 for BTC, and 2,300 and 2,200 for ETH. Declines don't happen all at once, and rebounds don't reverse in a day. Just follow the price.93.5% voted for "going solo," $STRK dropped 32% in a week
Starknet just concluded a vote: whether to explore breaking away from Ethereum and creating an independent, quantum-resistant Layer 1 chain. The result was 93.52% in favor, with 3,236 votes, an overwhelming approval.
In the first few days of the vote, $STRK rose by double digits. After the result came out, the coin price dropped about 32% in a week, currently around $0.053 (Beijing time 8:13 this morning), clearly underperforming the market.
Where’s the gap? Looking closely at the terms, this vote is non-binding: what passed was only to "establish a working group to conduct research," with a draft whitepaper due in November; the token remains $STRK, cross-chain bridges continue as usual, and a real decision to go solo would require another decisive vote.
In short, the market bought into the word "independent," but what materialized was just a research report. Expectations were high, but there was zero concrete proof, so the price gave back the overestimated portion. Until the whitepaper comes out in November, this is just a headline.#10月FOMC临近，美联储暂停加息还有哪些变数？
Glassnode data from 10.15–10.21 clearly breaks down this round of recovery.
BTC's rise is supported by capital: active buying increased by 4.9% week-on-week, selling only by 1.5%, with buying pressure stronger than selling; on-chain realized losses shrank by 32%, reducing capitulation. Short liquidations reached 158 million, a 52% week-on-week surge, with short covering providing additional buying momentum.
ETH's rebound quality is much weaker: active buying for the whole week dropped by 6.1%, selling by 9.7%, with the rise driven by reduced selling pressure rather than new buying; on-chain realized losses actually increased by 14%, with ongoing capitulation.
However, there has been improvement in the last three days: ETH active buying exceeded selling, profit and loss basically balanced, coin-margined futures positions decreased by 3.5%, short liquidations increased by 58%, and short covering added some confidence to the market.
But risks remain: BTC's buying advantage is not large, ETH's weekly buying is still shrinking, and the rebound foundation is weak. Once selling pressure reemerges, it can easily reverse at resistance levels.
The rebound is real, but the foundation is thin. $BTC $ETH $ZEC Uniswap quietly switches on hooks that route trades to rival pools
$UNI
Uniswap Labs has turned on aggregator hooks, a feature that lets Uniswap v4 pools pull liquidity from outside platforms instead of trading only against what sits in their own pool. The hooks are live on Ethereum, Base and Robinhood Chain, according to Crypto Briefing, which cited The Block. There was no formal launch event or press release the developer documentation is what confirms the feature.
#WillFedPauseInOctober There's another detail worth noting today: the world's largest gold ETF reduced its holdings by nearly twelve tons in one day, and silver even plunged over 5% intraday yesterday.
Many people's first reaction is—how can safe-haven assets fall? This is precisely the key to this market cycle.
When gold, silver, tech stocks, and $BTC all drop together, the market isn't selling based on views about any particular asset; it's indiscriminately deleveraging and converting back to cash. This isn't risk aversion; it's deleveraging. During risk aversion, gold prices rise; during deleveraging, even gold gets sold to meet margin calls.
It's important to distinguish these two types of declines: in the former, you can shelter in gold; in the latter, you can only shelter in cash and stay out of the market. We are in the latter now. So my largest position these days is called "no position."$BTW don't short, 98% controlled supply, this control is no worse than RAVETonight, keep an eye on a number that hasn't trended: Japan's core CPI for September is 2.1%, exceeding the central bank's target, the first time since the end of last year.
What does this mean? The probability of the Bank of Japan raising interest rates in December has risen significantly. Many people wonder what Japan's rate hike has to do with $BTC? It matters a lot.
One of the hidden fuels for global risk assets in recent years has been the yen carry trade—borrowing yen at almost zero cost to buy US stocks, crypto, and everything else. Once Japan truly enters a rate hike cycle, this tap will gradually tighten.
This isn't a bearish factor that will crash the market tomorrow; it's the kind of subtle undercurrent you don't notice day-to-day but realize in hindsight that liquidity levels have been steadily dropping. In a deleveraging environment, this fundamental change in liquidity is far more valuable than a breaking news flash from the Middle East.The stablecoin rewards channel has finally been loosened by regulators.
The U.S. Office of the Comptroller of the Currency (OCC) released the final rule for the "GENIUS Act," with two core points: relaxing restrictions on stablecoin rewards, allowing incentives linked to holding and using stablecoins to be offered to users within a compliant framework; and providing issuers with greater flexibility in reserve management.
The relaxation of reward criteria effectively pushes the previously contentious "earning interest on held coins" issue toward compliance, increasing stablecoins' appeal for capital and helping issuers retain on-chain funds through yield incentives; more flexible reserve management reduces operational friction and costs for compliant issuers, benefiting leading issuers like USDC and USDT as well as the entire payment stablecoin ecosystem.
There is no direct price catalyst at the token level. The real signal lies in how the boundaries of reward forms, reserve asset composition, and disclosure requirements will be further detailed, and whether channels for banks and non-bank institutions to enter stablecoin issuance will open accordingly. This is a regulatory framework implementation rather than a one-day catalyst; it should not be viewed as a short-term buying opportunity but rather tracked as a mid-term variable for stablecoin institutionalization and industry demand expansion.
Source: BlockBeats
Are you more concerned about the boundaries of reward forms or the disclosure requirements for reserve asset composition? The storage narrative cracked open another gap today, but hardly anyone looked up.
SK Hynix just announced it is abandoning the plan for its flash memory subsidiary Solidigm to go public in the U.S., and has also suspended the pre-IPO fundraising process. The official reason is that Korean regulators are worried about "dual listing" and diluting minority shareholders. But if you trace it back: Goldman Sachs and Morgan Stanley were already set as underwriters, and the deal was pulled at the last moment.
When a sector is hottest, companies scramble to spin off and cash out through IPOs. Now, stepping on the brakes voluntarily shows that management itself isn’t confident about this valuation window.
Retail investors have been shouting "storage bottomed out" for four days straight, but I only watch how the company votes with its feet. Talking bullish is one thing; putting real money on the line by going public is another.#10月FOMC临近，美联储暂停加息还有哪些变数？
🚨 The October FOMC is approaching! A pause in rate hikes is almost a consensus, but the real risk for BTC may lie in December!
📅 The Federal Reserve will hold its policy meeting on October 27–28.
Currently, the federal funds rate is 3.75%–4.00%.
📊 Three data points reveal the current dilemma:
🔴 US September CPI rose 3.7% year-over-year, well above the 2% inflation target.
🔴 September PPI rose 6.3% year-over-year, indicating persistent cost pressures on businesses.
🟢 A Reuters survey shows that out of 100 economists, 94 expect a pause in rate hikes in October.
Here’s the question: Inflation remains high, so why not continue raising rates?
Because monetary policy has a transmission lag. The effects of previous hikes have not been fully realized, and continued tightening could increase the risk of economic downturn.
But a pause does not mean a shift to easing!
⚠️ The September FOMC minutes show that most officials believe rate hikes may still be needed before year-end.
🎯 What does this mean for BTC?
If there is a pause in October but the Fed emphasizes hikes in December, the dollar and US Treasury yields may get support, keeping BTC under pressure.
If inflation falls and employment weakens, expectations for future tightening may truly cool down.
💡 My judgment:
Whether October pauses has already been priced in by the market; the real determinant of the next market move is how rate expectations change in December.
📌 Key focus: policy statement, chair’s press conference, two-year Treasury yields, and the dollar index.
💬 Do you think the Fed will hike rates again in December?
$BTC $ETH
#FOMC #美联储 #BTC #比特币 #宏观经济At the early morning when 180,000 people were liquidated, the Nasdaq was also bleeding. This is not a coincidence; it's the same rope tightening.
Bitcoin dipped to $81,000, Ethereum fell below 2,400, and over 90% of long positions were liquidated within 24 hours. On the same day, the Nasdaq recorded its largest single-day drop since June, the Philadelphia Semiconductor Index plunged 6.7%, Nvidia dropped over 4%, and CoreWeave fell more than 10%.
On the surface, it's concerns about AI financing and crypto leverage liquidation; underneath, it's the same thing: the cost of money. JPMorgan Chase and Goldman Sachs led a $5 billion AI infrastructure loan, of which $3.7 billion in GPU chip loans were refused underwriting guarantees by both banks, pushing yields above 11%. Even Wall Street is unwilling to fully take on AI debt risk. The 10-year real yield remains high at 2.74%, and both high-valuation AI and liquidity-supported crypto are being pulled up simultaneously by the same discount rate rope.
But first, distinguish: stopping the fall and reversing are two different things; right now, it's only the first. Bitcoin hit $81,000 three times, losses are shrinking, whales are accumulating, USDT reserves are rising, but spot buying hasn't caught up, and Coinbase premium remains negative—just because it can't be sold doesn't mean it will rise.
What truly determines the reversal is the data; before that, any rebound is just a breather. Watch the FOMC on October 27–28; then look at the PCE on October 29, the non-farm payrolls on November 6, and the CPI on November 10. Any clear cooling could dismantle the December rate hike expectations. 20261023
1. Bottom fishing or portfolio adjustment?
1. BlackRock bought $665 million in BTC ETFs and sold $95.4 million in ETH ETFs on the same day.
2. On that day, BTC ETFs had a net inflow of $526.7 million, ETH had a net outflow of $103.3 million, a difference of $630 million. Excluding BlackRock's ETF, the other 9 BTC ETFs combined had a net outflow of $138.5 million.
3. Next, watch three things:
(1) Whether the daily outflow of ETHA (the spot ETH ETF issued by BlackRock) can be suppressed below $50 million (a level not seen since the end of September)
(2) The settlement dates of several crypto-related ETF products
(3) The Federal Reserve meeting on October 28; inflation has already been ignited, what decisions will be announced next
4. On the same day, BlackRock also filed for a new fund PBIT—a BTC ETF with "downside protection." While increasing BTC holdings, it sold ETH and simultaneously prepared a product for "what if the price drops too much."
5. This behavior seems more like a single product, single-day action; building positions, adding positions, or hedging and closing positions could all look like this. The real criterion is whether it can continue.
2. Is Ethereum going to be completely reduced to a settlement channel?
1. JPMorgan Chase announced on October 20 the contract address of JPMD on the Ethereum mainnet. Upon inquiry: 0 holders, 0 supply, 0 transfers.
2. JPMD is a bank deposit token,Under the surgical lights, when a heart is put on an extracorporeal circulation budget of $195 billion to $205 billion, the real question is never how fast it beats, but how much autonomous coronary perfusion pressure remains.
Alphabet has raised its 2026 capital expenditure guidance to this range, equivalent to connecting a high-flow pump to the systemic circulation. Artificial intelligence infrastructure is the motor of this pump, and Google Cloud is the distant organ it must support. Clinically, the greater the pump flow, the seemingly better the peripheral perfusion, but the myocardium's own oxygen consumption and microvascular damage also accumulate simultaneously. The market only watches the rhythm of cloud revenue growth, but I focus on the ejection fraction—that is, profit margin—and coronary blood flow—that is, free cash flow. If these two indicators decline early post-operation, no matter how good the systolic pressure looks, it is sustained only by vasoactive drugs and machines.
The surge in capital expenditure is a typical myocardial hypertrophy response: short-term compensation and increased workload, but long-term fibrosis and diastolic dysfunction. The depreciation curve is that fibrotic scar; it does not show up on the ECG that day but continuously suppresses compliance in the following quarters' cash flow. If cloud growth cannot simultaneously raise cardiac output, the oxygen required per unit of revenue will increase, and marginal efficiency will deteriorate.
For linked targets like $xINTC, it is more like a pressure sensor attached to the extracorporeal circulation circuit. The moment the parent company's financial report is released is like an intraoperative transesophageal ultrasound: any regurgitation in cloud growth, guidance scope, or depreciation pace instantly changes the downstream circuit's pressure gradient. No matter how detailed the preoperative assessment is, what truly determines whether the patient can be weaned off extracorporeal circulation is the autonomous rhythm in the few minutes after the aorta is opened.
I do not perform emotional defibrillation; I only look at bleeding volume and perfusion pressure. If this coronary artery of cash flow continues to narrow during the capital expenditure expansion period, even the strongest myocardium will eventually enter low cardiac output. #alphabetq3cloudandai#SK海力士Q3财报在即，HBM4能否提振盈利？
$SKHY $MU
The storage upcycle is very likely to continue until 2027-2028,
but the stock price has already priced in most of the expectations.
After the earnings report,
if there is a combination of "performance exceeding expectations but guidance being conservative",
there might be short-term profit-taking on the good news;
Conversely, if HBM4 volume and 2027 pricing exceed expectations, that will be the real new catalyst.
The performance is no surprise; the suspense lies in the guidance.
Keep an eye on the HBM4 shipment pace and next year's pricing,
these two are the decisive factors for the stock price direction.BTC continued its weakness today (10/23), currently trading around $81,500, down 2.5%–3.7% in 24 hours, with a low of $81,331, basically giving back the rebound from Wednesday this week. Technically, the $80,000–$81,250 range is the core defense line, having absorbed selling pressure multiple times in October; resistance above is at $84,000, further at $87,570, the annual opening price. Glassnode views this as a box range oscillation between $80,000 and $87,000. On the funding side, on 10/22, the total network liquidation was about $221 million, with long positions accounting for 93.7%. BTC liquidation was about $77 million, with the largest single Binance long position at $11.78 million, with a liquidation price around $83,526, indicating crowded longs. From 10/19 to 21, spot ETF net inflows were $1.07 billion accompanied by a short squeeze, yet $84,000 was not held, indicating macro selling pressure outweighs crypto buying. On the macro front, BTC still behaves like a high-beta Nasdaq, with the Nasdaq down 2.47% synchronously; CPI and PPI exceeding expectations suppress risk appetite, and the FOMC on 10/27–28 is the biggest variable. In the short term, $80,000 is a psychological and chip threshold; a daily close below it could open space below $75,000; holding it would continue the box range, but breaking through $84,000–$87,000 before the FOMC is difficult.Bitcoin is a digital commodity and also a decentralized alternative to fiat currency. It is often referred to as "digital gold," and in our view, it should be valued according to this logic. However, most other crypto assets can be valued based on cash flow. According to Grayscale, rigorous fundamental valuation methods will play a decisive role in the next phase of digital asset investment. On-chain applications usually have clear and identifiable business models, generating fee income through transactions, lending, and other financial activities. After covering supply-side costs, the residual income can be quantified and compared horizontally across different protocols. Similar to traditional enterprises, these fundamental indicators form the basis for digital asset valuation. Crypto protocols have various mechanisms to return value to token holders: buybacks can reduce circulating supply, token burns permanently reduce total supply, and staking dividends directly reward token holders. Meanwhile, new token issuance and token unlocking offset these returns through equity dilution effects. Therefore, understanding the relationship between capital returns and token supply is crucial for assessing the economic value each token can capture. Grayscale Research evaluates token holders' economic returns using two metrics: Tokenholder Yield and Adjusted Earnings Yield. Tokenholder Yield reflects the distribution of returns to token holders after deducting supply inflation. Tokenholder Yield = (