Orbit Post Sitemap
Anthropic IPO new progress, the most worth watching is not how exaggerated the valuation is, but how much truth the prospectus dares to reveal
The primary market can talk about model capabilities, enterprise clients, and future potential. The public market is not so easy to please; it will scrutinize revenue structure, computing power costs, loss rhythm, client concentration, and will also watch every cloud vendor cooperation deal to see if it is revenue or just exchanged for binding
What I most look forward to in AI companies going public is this: finally being able to see how heavy the business behind the “smart model” really is. Training is expensive, inference is also expensive, talent is even more expensive. If every round of capability improvement requires burning more money, the valuation story cannot rely solely on the four words “huge future”
There is already enough faith in AI, what is now lacking is the accounts
#Anthropic：IPO新进展，招股书拟9月公开 #Employment data released intensively, Wash's policy stance under scrutiny
This week is the real drama.
Starting today, employment data will be released for four consecutive days—Wednesday JOLTS job openings, Thursday ADP and initial jobless claims, Friday August nonfarm payrolls. These four reports will directly determine whether there will be a rate hike in September.
Last month's nonfarm payrolls did something rare—it turned negative directly.
On one side is "weaker employment," on the other is "inflation is not over yet," two forces are clashing.
So the core question this week is just one—how weak is the employment data, and can it bring Wash's hawkish face back a bit.
For the crypto circle, the script this week is very clear. If Friday's nonfarm payrolls continue to weaken, or even show consecutive negative growth, rate hike expectations will be smashed down again, and Bitcoin has a chance to rise again around the 80,000 level. If nonfarm suddenly rebounds and employment remains strong, Wash will be more confident, the probability of rate hikes will continue to rise, US Treasury yields will rebound, and risk assets will take a hit in the short term.
My own view is that the data this week is unlikely to be one-sided—employment is cooling down, which is a fact, but not enough to make the Federal Reserve pivot. The best market script is "weak employment but not collapsing," allowing rate cut expectations to gradually build, rather than suddenly triggering recession panic.
What do you think?
$BTC $ETH The reason for the simultaneous drop in all coins has been found.
As expected, Trump is behind it again.
This time it's not the Middle East, but a sudden increase in digital taxes on Europe, causing a global plunge in risk assets. However, Bitcoin held firm, quickly bouncing back from 78,000 down to 76,500, indicating real money is buying at the bottom. Such a level of policy bearishness didn't create a deep pit; the chip structure is more solid than expected.
Ethereum is much weaker, breaking 2,500 directly, with 2,400 hanging by a thread. The valuation propped up by ETF funds last year is now being squeezed out bit by bit. Without new stories, faith can't support the price.
I'm still bullish on the crypto space; what’s really worth watching are the upcoming crypto bills and stablecoin regulations. Once the compliance gates open and incremental funds enter, the market will turn around immediately; the current struggles are just the prelude.
The three storage giants were mistakenly sold off along with the market; Hynix, SanDisk, and Micron just rebounded but were knocked down again. But the underlying logic of AI storage hasn't changed; HBM is still in short supply. The short-term valuation cuts don't change the long-term tight supply and demand. This sharp drop is just a pullback to pick up buyers, though it's a bit harsh.
SPCX remains resilient this time, holding steady around 141. The sector positioning is good; the valuation is high but justified. I'll be closely watching around 155; a breakout with volume will be a buy, and 200 is not a dream.
#BTC high-level oscillation, enhanced linkage with gold
#Employment data densely released, Walsh's policy stance tested Family, the reason for the simultaneous drop of all coins has been found.
As expected, Trump is at it again.
This time it's not the Middle East, but a sudden increase in digital taxes on Europe, causing a global plunge in risk assets. But Bitcoin held firm; it dropped from 78000 to 76500 and quickly bounced back, indicating real money is buying at the bottom. Such a level of policy negative news didn't create a deep pit, showing the chip structure is more solid than expected.
Ethereum is much weaker, losing 2500 directly, and 2400 is also precarious. The valuation propped up by ETF funds last year is now being squeezed out bit by bit. Without new stories, faith can't support the price.
I'm still bullish on the crypto space; what’s really worth watching are the upcoming crypto bills and stablecoin regulations landing. Once the compliance gates open and incremental funds enter, the market will turn around directly; what we're seeing now is just the prelude.
The three storage giants were mistakenly hit along with the market; Hynix, SanDisk, and Micron just rebounded and were knocked down again. But the underlying logic of AI storage hasn't changed, HBM is still in short supply, and short-term valuation cuts don't change the long-term supply-demand tightness. This sharp drop is just a pullback to pick up buyers, just a bit harsh.
SPCX remains firm this time, holding steady around 141. Good positioning in the sector, valuation is high but justified. I'll focus on around 155; if it breaks out with volume, I'll add, 200 is not a dream.
#BTC high-level oscillation, enhanced linkage with gold
#Employment data densely released, Walsh's policy stance tested 8.31 Second Bitcoin Dodan fulfilled as scheduled
$ETH pre-market strategy: scale in long positions between 2380-2400, stop loss at 2350, target range 2460-2520
During the session, when the price dropped to 2402, a real-time entry alert was given, precisely hitting the upper edge of the support range; current price reached 2444, prompting to reduce position by half, locking in 40+ points profit, with the remaining position held to play for the target
The range forecast was precise and on point, entry and profit-taking tracked throughout, pullbacks are always buying opportunities, trading rhythm is always one step ahead
#就业数据密集公布，沃什政策立场受检验 2026.8.31: $BTC ~ Stablecoins ~ U.S. Treasury bonds, oil ~ Iran ~ U.S. dollar! Bitcoin corresponds to the stablecoin supply; when stablecoins increase, U.S. Treasury bonds are purchased. Why pick a fight with Iran again today? It's still about the U.S. dollar. Once there is turmoil in the Strait of Hormuz, oil-importing countries will massively stockpile oil, which requires more U.S. dollars. Therefore, the U.S. will not let Iran be at peace, but will periodically create some tension with a rhythm, followed by sensitive reactions in oil prices. Similarly, behind stablecoins are U.S. Treasury bonds, so the virtual currency market capitalization needs to grow. When panic increases, $BTC rises. Record the planet! Record real trading!Triple Logic Behind August's Surge
August's surge was no accident but the result of a resonance of three forces:
1. Return of the "Dollar Devaluation Trade": The U.S. Treasury expanded long-term bond repurchase operations, lowering long bond yields and weakening the dollar, driving funds into hard assets like gold and Bitcoin for hedging.
2. The Largest Short Squeeze in History: Around August 19, roughly billions of dollars in short positions across the market were forcibly liquidated, triggering a chain reaction of "short covering → price push → more liquidations."
3. Record ETF Inflows: ETFs saw cumulative inflows exceeding $3 billion in August, signaling a strong return of institutional buying. $BTC $ETH $SOL #财报观察员：博通与戴尔接棒，AI回报再受检验 Fundamental Research Report $WLD / Worldcoin (AI/Computing Power) $3.20
Essentially: Worldcoin ($WLD) overall score 60/100, rating narrative outweighs execution. Breaking down the three layers, the company team has cash reserves, the protocol network shows paid usage traces, and token value capture has been realized.
Worldcoin (token $WLD), AI/computing power sector. Main focus: Sam Altman identity + AI. Comparable to FET, TAO. Traditional computing power rental giants are AWS, CoreWeave, charging by GPU hour; A100 monthly rent is $12,000–$25,000, expensive and high threshold. On-chain solutions fragment computing power for bidding; suppliers require no centralized approval; idle GPUs become available supply. Customer unit price $50–$500/month, settlement in USDC or fiat. Narrative-driven sector, usage drops 60–80% in bear markets. Positioned as an end-to-end vertical platform. Product implementation: protocol layer officially running, on-chain dashboard shows protocol fees accumulating, paid usage traces exist. Latest version not found; 60 valid commits in last 90 days.
User side: address MAU undisclosed, DAU undisclosed, 24h trading volume $80.00M, TVL not found. Wallet addresses do not equal natural person monthly active users; large addresses concentrated holdings overestimate real user count. Revenue side: user fees undisclosed; supplier income about 80–90% of user fees (to LPs and nodes); protocol treasury income $2.00M; token holder buyback and burn annualized no burn mechanism. 24h trading volume is business turnover, not revenue. Company profit does not equal protocol profit; protocol profit does not equal token holder profit. Code side: 60 valid commits in 90 days, 25 active contributors, latest version not found. GitHub is A-level evidence for direct verification. Investment background: company equity financing checked via PitchBook/Crunchbase (A-level); token private and public sales checked via whitepaper, release schedule, and on-chain unlock contracts (A-level); market makers and ecosystem funding are B-level, not representing long-term VC holdings; technical integration checked via API/SDK evidence (B-level); strategic partnerships and logo walls are D-level. NVIDIA GPU usage does not equal NVIDIA investment; exchange listing does not equal exchange strategic investment.
Token side: total supply 1,300,000,000; circulating 950,000,000 (73.1%); FDV $4.20B; next unlock 2026-Q4 (adds +3.50% to circulation); annualized burn/buyback no clear mechanism. Must buy tokens to use product? Partially, medium value capture (staking/discount/governance). Compared with peers (uniform criteria, no cross-sector comparison): Circulating market cap: Worldcoin $3.00B, FET undisclosed, TAO undisclosed. FDV: Worldcoin $4.20B, FET undisclosed, TAO undisclosed. Annual revenue: Worldcoin $2.00M, FET undisclosed, TAO undisclosed. Monthly active addresses or users: Worldcoin undisclosed, FET undisclosed, TAO undisclosed. Figures based on public data snapshots; missing data supplemented by official or industry sources. Valuation: circulating market cap $3.00B, FDV $4.20B, P/S 1500.0x, FDV divided by revenue 2100.0x. Pessimistic view discounts $3.00B by 50–70%, neutral range oscillates, optimistic expects revenue doubling, burn implementation, enterprise clients entering, FDV P/S aligns with top projects.
Summary: fundamentals solid (score 60/100). Token value capture realized (buyback/burn/gas). Circulating market cap relatively expensive compared to fundamentals, overextended expectations, FDV moderate. Potential risks: short-term large unlock sell-off, protocol income long-term zero, token demand relying solely on incentives (if incentives stop, usage collapses). Next focus metrics: weekly protocol fees, burn amount, active address retention, TVL/loan balance, GitHub version releases. Information source public, logic self-developed, not investment advice. Data deviation over 30% requires reassessment.
Fundamentals analyzed, market direction is another matter.
#FundamentalResearchReport #Crypto #Research #OKXOrbitETH has climbed back above $2,400, but don’t rush to call it a catch-up rally just yet. My view is that this level feels more like a crossroads for direction rather than a fully charged state ready to take off at any moment.
Why do I say that? Looking only at the USD price, ETH surged from below 1,900 this week, a pretty fierce rally. But if you look at it from another angle—its ratio against Bitcoin—it’s not so optimistic. ETH/BTC has been steadily declining since early 2025, weakening for over a year until it bottomed out in June this year. In the past two months, it has rebounded over 20% and even formed a golden cross. Sounds good, right? The problem is that historically, this ratio’s golden crosses have often failed—sometimes it rose 30% afterward, sometimes it crashed immediately. This signal is at best a reference, not a decree.
So what’s the current situation? The USD price is running fast, the ratio is just starting to rise, indicating some capital is testing the waters—ETFs continue to see net inflows, and staking yields are becoming attractive again amid expectations of falling interest rates—but big money hasn’t truly shifted from Bitcoin yet. To confirm a buildup, we need to see ETH/BTC volume-backed stabilization, not just a one-legged USD price surge.
In short: Around $2,400 for ETH, the catch-up rally story has already been told by some, but the direction’s exam paper hasn’t been handed in yet. Watch the ratio, not just the price.🔥 Marvell just beat expectations — but Broadcom still looks like the cleaner AI winner.
$MRVL delivered a strong quarter, with Q2 revenue hitting $2.739B, up 37% YoY, while Data Center revenue jumped 46%. Even better, Q3 revenue was guided to around $3.15B ±5%.
But there’s a catch. 👀
Non-GAAP gross margin came in at 58.9%, with next quarter guided down to 57.5%–58.5%. The custom AI ramp is clearly gaining traction, but the growth is also bringing some margin pressure.
#DailyOrbit Solana stablecoins surge to $16.4 billion, can SOL really benefit from this money?
Let's first pour some cold water: more stablecoins do not equal a rise in SOL. The $16.4 billion figure sounds intimidating, but if the money just sits in wallets without moving, it has no impact on SOL's price. What really matters is where this money flows.
If stablecoins just cross chains and sit idle, they are merely passersby, and Solana only earns a bit of settlement fees. But if the funds flow into lending markets, DEX liquidity pools, payment scenarios, or even become the settlement layer for RWA assets, then the nature completely changes—every swap, every collateralization, every liquidation consumes $SOL as fees, and on-chain activity directly boosts staking demand and validator income. This is the full chain of liquidity transmission to the token price.
On the positive side, Solana stablecoins already account for 10% of the global share, and RWA has a scale of $2.8 billion, indicating the ecosystem is indeed moving towards "usage" rather than just hoarding tokens. Plus, another $500 million USDC was minted in June, so incremental funds are still entering.
So the conclusion is simple: the positive outlook is real, but realization takes time. Monitoring changes in DeFi locked value and DEX trading volume is far more useful than just watching stablecoin totals. Money moves, SOL has a chance; money doesn't move, no matter how impressive the numbers look, it's just a paper prosperity.basically 0 leverage built up on latest BTC rally, price driven by etf buyers and short squeezes
wouldnt be surprised if it just happens again this week
#LaborMarketTestsWalsh
#BTCGoldCorrelation
#BroadcomDellAIResults HYPE whales start cashing out, over $20 million worth of spot continuously sold
The HYPE spot whale starting with 0x0a84 began continuous position reduction today. From 9:49 to 13:49 Beijing time, it sold a total of 11,113 HYPE, with a transaction amount of about $893,200 and an average transaction price of about $80.37.
Within nearly half an hour, this address sold another 1,588.95 HYPE, with a transaction amount of about $128,400. The latest sale occurred at 13:49, and since then, there has been a sell order of 26.14 HYPE spot at $81.055, indicating the position reduction has not completely stopped. The total realized profit from the above completed sales is about $429,700, with fees around $221.
As of the time of writing, this address still holds 247,555.22 HYPE, valued at about $20,059,000, with the remaining spot size about 22 times the amount sold in this 4-hour round.Four large addresses will receive $44.7 million worth of HYPE spot this week
As of press time, based on large stakers holding at least 100,000 tokens cumulatively, about 2.5658 million HYPE are in the large stake withdrawal queue over the next 6 days, valued at approximately $208 million at current prices.
Among them, protocol contracts, Trade.xyz, and Hyperliquid Labs-related addresses still account for the majority of the volume, but there are also 4 large external addresses: totaling about 551,100 HYPE, valued at approximately $44.69 million, which will successively complete the 7-day waiting period and re-enter spot accounts:
0x023a "Auros Global": about 101,000 HYPE, expected to arrive on the evening of August 31, valued at about $8.19 million;
0x251f: about 135,100 HYPE, expected to arrive on the evening of September 3, valued at about $10.96 million;
0xc288: about 115,000 HYPE, expected to arrive on September 4, valued at about $9.33 million;
0xe867 "Million HYPE Staking Whale": about 200,000 HYPE, expected to arrive on September 6, valued at about $16.22 million.
The four addresses together account for about 21.5% of the large amount arriving in the next 6 days.
The remaining approximately 2.0147 million HYPE involve stakedHYPE, Kinetiq, Trade.xyz, and Hyperliquid Labs-related addresses, with a single-day peak of about 1.34 million HYPE expected on September 5 (large stakers account for 1.05 million HYPE).*Employment data is being released intensively, and Walsh's policy stance is under close watch*
Big news is coming, big news is coming, this round is going to be easy points 📊
*$PUMP current status*
Down *-21.5%* from the top
My position is currently floating with a profit of *+14 points*
Feels like this decline isn't over yet
Several bad signals in the market:
1. *Open Interest (OI) continues to shrink* = bulls are cutting losses and exiting
2. *Long-short ratio 1.31* = 31% more longs than shorts. Too many people on the long side, no need for the market makers to push up
3. To translate: retail investors are all long, the big players are just accumulating
*Why I think it will continue to fall*
1. *#LaborMarketTestsWalsh* data is piling up. As long as it's strong, Walsh will take a hard stance, and the market will directly price in a rate hike. Highly volatile altcoins like $PUMP will definitely get hit first
2. *#BTCGoldCorrelation* Big brother $BTC is only *$77,020*, down from *$81,400*. Without the main market stabilizing, altcoins don't stand a chance
3. *#BroadcomDellAIResults* All the money is going to wait for Broadcom and Dell AI earnings. Without foreign capital coming in, $PUMP can only fight internally
*My plan*
*Take profits first*
+14 points is good. Making money despite a 21% drop means risk control was done right
*Wait for 2 conditions before considering shorting:*
1. OI stops falling and rebounds + long-short ratio drops back below 1.1 CryptoQuant's CEO publicly stated that this round of the $BTC bear market has ended because the current trend is very similar to that of November 2023. Their main indicator for judging whether Bitcoin is in a bull or bear market is how far the current price is from the 365-day moving average.
But personally, I think it's not so early, not that early yet. I believe BTC needs to break above 83,000 and even close multiple daily candles or one or two weekly candles above that level to confirm the end of this bear market, with 57,000 being the low point of this wave.
Everyone must always remember: the end of a bear market only means that lower lows are unlikely to appear, but it does not mean there won't be pullbacks. During the one and a half to two-year uptrend in the bull market cycle, there will still be many 25 to 35% pullbacks.After $NVDA released its earnings, I added more $MU shares. This time, Nvidia’s results gave me a signal that I think is more interesting than simply chasing $NVDA ’s headline numbers. Everyone is focused on another revenue beat and the continued expansion of AI infrastructure. But one detail caught my attention: $NVDA ’s Q2 gross margin was 75%, while Q3 guidance fell to 74%, partly due to rising memory costs. For Nvidia, that’s a cost pressure. For $MU, it could be an opportunity. 👀 The logiChina's leading AI chip companies saw nearly 2000% in revenue, and high-end computing power is in short supply across the entire chain
Recent financial reports from Chinese AI chip leaders such as Biren Technology, Cambricon, and Moore Threads show significant revenue growth, with Biren Technology's revenue increasing nearly 20-fold year-on-year. Industry research shows that by 2026, domestic AI chip demand will be about 4 million units, with actual deliveries around 3 million, resulting in a million-yuan gap. High-end computing power is in short supply across the entire chain, from chips to intelligent computing centers and system integration.
In August this year, leading domestic AI chip companies released a series of semi-annual reports, with collective explosive performance. Last Friday, Biren Technology released its financial report, with revenue growing nearly 20-fold; Companies like Cambricon and Moore Threads also recorded significant revenue growth. The core driver of the surge in performance is the shortage of high-end computing power. Currently, the supply side faces a full-chain tightness: not only are resources at downstream intelligent computing centers tight, but orders in the midstream system integration segment are also being intensively released, with some companies posting orders three years ahead. According to industry research data, the demand scale for domestic AI chips in 2026 is about 4 million units, with actual deliveries around 3 million units, resulting in a capacity gap in the millions. Industry insiders generally believe that domestic chips are in a strong upward cycle for high-end AI chips, and recent financial reports from multiple companies confirm this trend. This round of shortages not only reflects the acceleration of domestic computing power infrastructure construction but also reflects the sustained strong global demand for AI computing power.
Market Impact:
Direct benefits: AI chips/GPUs
- NVDA: Global demand for AI computing power remains tight, with NVIDIA as a high-end GPI believe the key for $SOL this time is not short-term sentiment, but a change in the supply logic. After the governance vote passed the dual deflation proposal, the future supply of SOL will decrease, effectively reducing inflationary pressure by one notch. What’s even more remarkable is that this is happening while the ecosystem is still hot: demand is paying attention, and the supply side is tightening again. The market’s pricing of SOL is likely to no longer just follow thematic rotations but will re-evaluate its scarcity. I wouldn’t interpret this as an instant positive, but if the momentum continues, SOL has a chance to enter a new round of value reassessment. My judgment leans positive; the core depends on whether the expectation of reduced supply can sustain trading, rather than just a one-day hype.#嘉信理财拟新增SOL、AVAX与LINK
Why is the most noteworthy thing in the market recently the divergence between Wall Street and retail investors?
On one side, Wall Street continues to allocate BTC through ETFs and other channels, while on the other side, many retail investors are still watching cautiously.
Retail investors are not blind to the rise; they just don't dare to believe in this rally. After experiencing several sharp rises and falls, many have developed a conditioned reflex: afraid to chase when prices rise, afraid of further drops when prices fall, and some still believe BTC is just a game for large funds, with the final profits coming from those who buy in last.
But Wall Street's logic is different.
For institutions, BTC is evolving from a pure speculative asset into an alternative asset that can be included in asset allocation. Institutions may not be concerned with price fluctuations over a few days but rather with BTC's position in global asset allocation over the coming years.
Of course, Wall Street doesn't only buy and never sell. Concentrated selling by institutions can also cause the market to drop quickly, making ordinary investors more likely to be the last to hold the bag.
So now, I tend to view BTC as being in a mid-term recovery phase after institutional funds have flowed back in.
However, the area around $80,000 is no longer the lowest risk position.
Opportunities still exist, but it's more suitable to control position sizes and participate in batches rather than chasing heavily out of fear of missing out.
True wealth opportunities have never been about betting on a single surge but about staying rational when trends emerge, first learning to survive, and then waiting for your own chance.*Yesterday, Brother Ma's account still had $8.2 million, today only $5.1 million left*
Last night $BTC and $ETH both crashed, dragging the whole market down 📉
Brother Ma's heavy positions in *$HYPE* and *$PUMP* couldn't hold and all hit stop-loss.
But his reaction was like this:
*Just got stopped out, immediately reversed*
Directly dumped *$10.2 million* to open a long $BTC position. That's ruthless.
*Why did this happen*
1. *#LaborMarketTestsWalsh* data was stronger than expected, raising rate hike expectations again. Risk assets were the first to be sold off.
2. *#BTCGoldCorrelation* $BTC dropped from *$81,350* to *$77,000*, and $ETH couldn't hold either. Leveraged altcoins $HYPE and $PUMP naturally were the hardest hit.
3. *#BroadcomDellAIResults* Money was waiting for Broadcom and Dell AI earnings reports, so liquidity was tight. Brother Ma's altcoins were taken by the market makers.
*Why does he dare to go long $10M BTC directly*
1. *Position* $77,000 is previous platform support. If broken, it goes to $75,850, risk is controllable.
2. *Sentiment* On 8/28 ETF single day *net outflow $204.5M*, breaking the 8/9 record of *$2.64B* net inflow. Institutions often rebound after dumping. $BTC BITCOIN
Last night $BTC tried to bounce back up again, but the local high was descending, signaling a weakening buying momentum and market overheating after active growth.
🫱It is also worth noting that there is already an inflow of coins to exchanges (in particular, Binance balances have updated the local highs of the year). This highlights the readiness of large players to take profits at current prices.
❗️I also want to report separately that Warsh, in his speech at Jackson Hole, took a "hawkish" stance. Instead of the expected hints at easing, he made it clear that the regulator is ready for further pressure. At the same time, most often the price of BTC starts to decline after such speeches. This was the case from 2021 to 2025, and 2026 may not be an exception.
In conclusion, I am preparing for a correction, at least to 75K. But ideally, I would like to see a deeper decline, at least into the 70K range. Recently, FB has experienced a significant drop, and the halving is approaching.
FB will halve at block 2,100,000. As of noon on August 31, the block height was 2,077,545, with 22,455 blocks remaining. It is expected to trigger between September 8 and 9 Beijing time, but the final trigger depends on reaching block height 2.1 million.
After the halving, the Fractal chain block reward will decrease from 25 FB to 6.25 FB, with another 6.25 FB allocated by the Bitcoin mainnet. In other words, the Fractal chain output decreases by 75%, but the overall new FB issuance only halves; do not confuse the two.
UniSat's buyback plan is divided into two parts: the first batch was 500,000 FB purchased in February this year, which has been completed and used for index staking; the second batch will start after the halving, investing $200,000 monthly for five consecutive months, totaling $1 million. The purchased FB will be held as a long-term reserve and locked for at least 5 years.
Currently, the second batch of funds has not started buying; the specific date, price, and lock-up address have not been announced. The fixed monthly amount is the investment, but the actual purchase quantity depends on the market price.
Whether this plan can truly create positive effects depends on three points: whether purchases are made as planned, the actual purchase quantity, and whether the lock-up address is disclosed. Before the funds actually enter the market, the announcement can only be considered an expectation, not a realized positive impact. $BTC A $31 billion investment was made, yet SanDisk's stock price dropped 20% in two weeks — the money wasn't wasted, it just hasn't been spent yet.
Watching the candlestick drop from 1828 back to around 1400, feeling itchy to buy but afraid of catching a falling knife. I've experienced both feelings.
📍 First, admit the dilemma
Is 1400 a golden pit or a falling knife? Both camps have valid points.
Bottom-fishers say: The long-term contract locked in $93.9 billion, covering about 50% of fiscal year 2027 capacity and about two-thirds of fiscal year 2028 capacity. The new factory won't start mass production until 2029, so supply remains tight.
The wait-and-see camp says: The price surged over 500% within the year, profit-taking is concentrated, the candlestick is still grinding downward, and RSI has dropped to 22.
Both sides have their cards; neither can convince the other.
📍 Debunking the most common misunderstanding
"Capacity expansion = negative" is a superficial misunderstanding.
The $31 billion isn't a one-time spend; it's phased over six years until 2032. The new factory in North Shanghai aims for mass production in fiscal 2029, with ramp-up delayed by a year. The current supply shock might be close to zero.
UBS explained this clearly: The capacity expansion announcement is more like a postcard to cloud providers — telling them: I have capacity; if you want to lock in supply, prices need to be negotiated. SanDisk + Kioxia merged market share is 35-37%, making this postcard weighty.
📍 Three real risk zones
Capacity expansion itself may not be a risk; the pace of disproof is. Three alarms to watch closely.
⚠️ Alarm one: Long-term contract fulfillment. $93.9 billion is signed, not earned. If any cloud provider delays delivery or renegotiates, valuation logic could collapse by half.
⚠️ Alarm two: BiCS8/9 ramp-up. Stacking above 218 layers is an implicit premise to absorb new capacity. If yields fall short of expectations, capital returns could be directly discounted.
⚠️ Alarm three: Whether AI Capex slows down. AI server demand in 2024-2025 drives about 30% global NAND bit shipment growth. If this curve turns downward, 1400 might not be the bottom.
📍 Four risk mitigation actions
🎯 Build positions in batches: start below 1450, buy more the lower it goes.
🎯 Position limit: no more than the amount you're willing to lose 20% on.
🎯 Stop loss: exit if 1400 breaks, wait for the next 4-hour candlestick before deciding.
🎯 Data to watch: next quarter NAND spot price, SanDisk data center business month-on-month, AI major players' capital expenditure guidance.
Don't chase above 1550; build positions in batches below 1450 but keep position size within your 20% loss tolerance; exit if 1400 breaks, wait for the next 4-hour candlestick before deciding.
$SNDK
#闪迪铠侠拟投310亿美元，NAND供需重估 $BTC*Interest rate hike expectations rise, $BTC takes the first hit*
The previous high was directly hammered.
It dropped straight from *$81,200* to *$77,050*, a waterfall decline. But this position is not a place to "buy blindly" yet. The reason is just one:
*1. ETF funds peaked and then fell*
This is the most painful.
On *August 9th*, net inflow was *$2.63B*, a record high, but on *August 28th*, there was a *net outflow of $203.1M*. This completely reversed the bullish sentiment from the 9th. Today, ETFs are leading the sell-off 💵
*The market also exposed 3 problems*
2. *#LaborMarketTestsWalsh* data is tough, and the market is recalculating the Fed. As long as rate hike expectations remain, crypto will be sold first.
3. *#BTCGoldCorrelation* Gold at *$2,578* is steady as ever, but $BTC is leading the decline. This shows it's not a safe-haven move, but a risk-off move.
4. *#BroadcomDellAIResults* Money is waiting for Broadcom and Dell earnings. Until AI is clearly defined, no one cares about crypto.
*Where is the key level now*
Current price is *$77,050*, the $77.5K support has been lost.
*Below looks at $75,900*. If broken, it will head to $74K.
*Above $78,900* if not reclaimed, the bearish trend is not over.
*What I will do*
*Hold back, do not catch the falling knife* In simple terms, this incident involved hackers manipulating the price of the low-liquidity TONIC token, using the inflated token as collateral to siphon assets from the Tectonic lending protocol. The project team couldn't directly pause block production on the entire Cronos chain, which blocked most of the stolen funds but also froze all on-chain operations for regular users.
My personal view is that if such attacks on DeFi happen again, the priority should be improving risk control mechanisms.
While oracles can be exploited through price manipulation, the root cause is that the protocol allows tokens with such poor liquidity to be used as large collateral. Knowing that this token has a small market cap and is easily manipulated, yet no collateral limits or liquidity checks were set, gave hackers an opportunity. Simply modifying the oracle without restricting collateral permissions for low-liquidity assets will lead to similar attacks with other tokens in the future.
Even on decentralized chains, in emergencies, the entire chain can be halted. Once halted, your DeFi positions and transfers become immobile, locking funds inside, which is a real trust risk.
In the future, when participating in DeFi, I will be more cautious with lending protocols that use small tokens as collateral. The smaller the token's market cap, the easier it is to manipulate the price. Don't assume that being on-chain means absolute safety.
Although most of the stolen assets were blocked this time, ordinary users were also locked out. This incident reminds us that DeFi security is not just about contract audits; risk control rules for collateral are equally important. #Tectonic遭操纵，Cronos暂停出块 Cronos pausing block production after an attacker manipulated low-liquidity TONIC pricing puts two defenses under the microscope: collateral controls before an incident and emergency intervention after one.
Researchers estimate about $75M was affected and roughly $6M bridged out, while Tectonic has not confirmed either losses or cause. My read: the more durable test is whether oracle design and risk limits can contain thin-market collateral without relying on a chain halt. Moonwell and Avici make that scrutiny broader than one protocol. Not advice, just analysis.
#CronosHaltsAfterAttackAt the beginning of June, UNI dropped to $2.31.
The market was full of wails. The group chat was all "Uniswap is done," "Governance tokens are trash," "This round of DeFi is dead."
I posted a tweet: "UNI is worth watching."
In the comments, some said I was crazy, others said I was foolishly stuck.
Today, UNI broke through $5.4. In three months, it doubled.
Not bragging. It's time to review—what I was thinking at the time.
Judgment One: The fee switch is a "gray rhino," everyone saw it but no one cared.
From 2020 to the end of 2025, Uniswap’s protocol generates hundreds of millions to over a billion dollars in fees annually, but UNI holders get not a penny.
UNI was only used for voting. Zero cash flow.
After five years of debate over the fee switch, in December 2025 it finally passed with 99.9% support—one-time burning of 100 million UNI and activation of the protocol fee switch.
On the day the news came out, UNI rose 50%, then followed the market down steadily, dropping to $2.3 in June.
The market thought this was just a typical "good news priced in" scenario.
But this is not a typical scenario. This is a qualitative change at the tokenomics level.
From zero cash flow to real protocol revenue—this kind of change is called a "qualitative change."
Judgment Two: Robinhood Chain is a "catalyst," bigger than anyone imagined.
On July 1, Robinhood Chain mainnet launched.
Uniswap’s v2, v3, v4, and UniswapX were the first public main AMMs on this chain on day one.
My judgment was simple: Robinhood has 24 to 28 million active accounts. These users previously could only buy crypto, now through Uniswap they can trade stock tokens 24/7.
What happened?
In the first week after launch, Uniswap’s trading volume exceeded $250 million. In six weeks, cumulative volume surpassed $1 billion.
On August 29, single-day stock token trading volume hit $130 million—a 10x increase in one month.
In the past 24 hours, Uniswap’s revenue was $4.29 million, accounting for nearly half of Robinhood Chain’s fee share.
Traditional finance traffic is flowing into DeFi through Robinhood Chain. Uniswap is the only faucet.
Judgment Three: The market completely ignored the value of "cash flow."
UNI was fully unlocked by 2024. But the market kept pricing it as a "governance token"—used for voting, worthless.
After the fee switch opened, protocol revenue flows into the TokenJar contract. Arbitrageurs wanting to extract assets must first burn an equivalent amount of UNI.
This is called Firepit.
It’s not the company buying back tokens—it’s an on-chain auction of protocol revenue; whoever offers UNI takes the assets.
As of August 31, about 110 million UNI have been burned, totaling $630 million in value. Since August, daily burns exceed $400,000, with Robinhood Chain contributing nearly half.
The more active the trading → the more fees → the more burns → the scarcer UNI → the higher the price.
This is the deflationary flywheel.
UNI’s valuation model must shift from "governance token" to "yield-generating deflationary asset."
Of course, I didn’t predict it would surge to $5.4 so quickly.
The direction was right; the rest is up to time.
The most important thing in investing is recognizing "qualitative change."
UNI changed from a "voting token" to a "deflationary yield asset"—that is qualitative change.
While most people are still watching the candlesticks, a few are watching fundamentals.
Find the qualitative change, then go heavy, be patient, and wait.
$BTC $ETH $UNI US military suddenly launches night raid on Iran! BTC plunges across the board, crude oil soars, is the market's worst fear coming true?
This past weekend, the Middle East situation suddenly escalated again.
The US Central Command confirmed that the US military attacked two Iranian rocket launchers in the Strait of Hormuz, marking the first publicly acknowledged US strike on Iran since late July. Iran then responded by firing missiles at US military bases.
The market reacted almost immediately: BTC led the crypto market down, while crude oil quickly rose.
What really needs caution is not the attack itself, but the Strait of Hormuz. A large volume of global crude oil shipments pass through here, and if the conflict escalates further, rising oil prices could push inflation expectations higher again, thereby affecting the market's judgment on the Federal Reserve's interest rate path.
The most concerning aspect of such sudden events is when news runs ahead of the market. I will directly use ave to monitor BTC and on-chain capital flows, while also observing if there is abnormal volume in the market; if the situation continues to develop, I will focus on whether funds on ave are bottom-fishing or continuing to withdraw from risk assets.
If the US-Iran conflict escalates → crude oil continues to rise → inflation expectations heat up → rate cut expectations are repriced, the real impact may extend beyond just the crypto space.
The most critical question now is:
Is this a one-time military action, or the start of a new round of US-Iran conflict?
#BTC高位震荡，与黄金联动增强 #美伊军事对抗升级，原油供应风险升温 $BTC $ETH Federal Reserve Chair Wash Jackson Hole hawkish, September rate hike probability jumps to 60%, BTC once retraced to 76,000, indicating continued high sensitivity to liquidity expectations;
$TRUMP
The rebound includes short squeeze elements, futures enthusiasm and spot follow-through have improved but have not completely decoupled leverage;
- 77,000 support, 79,400–80,800 resistance zone remain effective before the U.S. employment data on September 4, with a high probability of short-term consolidation to digest profit-taking.
Conclusion: BTC's recent advantage essentially reflects the fourfold realization of "USD credit hedge + ETF institutionalization + halving supply clearance + regulatory confirmation," with mid-to-long-term allocation logic stronger than in 2021; however, the 80,000 level is a macro data and leverage dual-sensitive zone, advantage does not mean one-sided without pullbacks, phased buying rather than chasing highs better fits the current structure.
#就业数据密集公布，沃什政策立场受检验 #BTC高位震荡，与黄金联动增强 #闪迪铠侠拟投310亿美元，NAND供需重估 UNI at $5.4, should you chase or not?
On August 31, UNI broke through $5.4.
It has risen over 100% in 3 months. The 24-hour trading volume ranks among the top in the entire market. FOMO has already started.
But I want to ask you a question: if you rush in now, are you investing or taking the bag?
Don’t rush to answer. After looking at these three sets of data, then decide.
First, the bull logic—why can it still rise?
First, the incremental growth brought by Robinhood Chain has just begun.
The mainnet launched in July, and in less than two months, TVL has exceeded $700 million. Uniswap’s daily trading volume of stock tokens on Robinhood Chain hit a new high of $130 million—10 times growth in the past month. In the past 24 hours, Uniswap earned $4.29 million in fee income on Robinhood Chain, accounting for nearly half of the entire chain.
Do you think this is the end? Robinhood Chain’s active users have surpassed 20 million—these people haven’t all come in yet.
Second, the deflationary flywheel has started turning.
UNI was criticized for five years as a “zero cash flow” token—the protocol earns hundreds of millions to over a billion dollars a year, but token holders get nothing.
In December 2025, the fee switch will finally be enabled. As of August 31, 110 million UNI have been burned, worth $630 million. Since August, the daily burn has exceeded $400,000.
UNI has transformed from “voting-only air” to a “deflationary asset with cash flow.”
Third, institutions are buying, whales are accumulating.
Whale withdrawals of UNI on Binance hit a five-year high. Three whale wallets moved 146,000 UNI from Upbit hot wallets to cold wallets. Standard Chartered Bank set a $100 target price for 2030.
Smart money is accumulating, retail is hesitating.
Now the bear logic—why should you be cautious?
First, whales are already taking profits.
On August 6, a whale sold 429,000 UNI for 929.3 ETH, profiting $320,000. The average price was $4.09.
Others were running at $4, and you want to rush in at $5.4?
Second, UNI is still down for the year.
$5.4 looks high, but UNI was higher at the start of the year. It’s still down about 20% year-to-date. From the all-time high of $44.97, it’s down 88%.
A 100% rise in 3 months is a rebound; only rising back to $44 would be a reversal.
Third, this rise is not driven by trading volume.
Uniswap’s overall trading volume declined in July and August. The incremental growth on Robinhood Chain supports UNI’s price, but the protocol’s fundamentals have not improved accordingly.
The short-term rise is too large; RSI has entered the overbought zone. The risk of chasing highs is for you to weigh.
My view:
UNI’s valuation logic has indeed changed. From a zero cash flow governance token to a protocol asset with a deflationary mechanism.
This is a fundamental change.
But even the best assets have corrections. Is UNI at $5.4 expensive?
In three days, yes. The short-term rise is too steep and could correct at any time.
In three months, reasonable. Standard Chartered’s year-end target is $6.5, so there is room.
In three years, possibly cheap. If the deflationary flywheel keeps running, if Robinhood Chain continues to expand, if the tokenized asset wave really comes—the $5.4 price could be the starting point.
Your time horizon determines your answer.
$BTC $ETH $UNI Correlation indicators have changed: BTC is gradually aligning with gold, while ETH remains tied to tech risk assets
Recently, the 90-day correlation data has shown significant changes: the correlation between BTC and the Nasdaq tech index has noticeably declined, while its correlation with gold has increased; meanwhile, ETH's correlation with Nasdaq remains high, showing no signs of decoupling.
This signal indicates that asset positioning within institutions is diverging:
Some institutions are beginning to treat $BTC as a debt hedging tool, not fully following the ups and downs of US tech stocks;
$ETH is still defined as a growth-oriented risk asset, with the sentiment in the tech market directly affecting ETH's capital flows.
However, there is an important misconception here: a shift in positioning does not mean interest rates can be ignored in the short term.
Even inflation-resistant assets like gold and BTC will face downward pressure and decline when US Treasury yields rise sharply.
The long-term logic is that US debt expansion benefits BTC; the short-term logic is dominated by Federal Reserve interest rates, with short-term logic taking precedence over long-term narratives Jackson Hole released a hawkish statement prioritizing inflation suppression, and the market quickly priced in the change, with the September rate hike expectation soaring from 35% to nearly 58%. Risk assets collectively came under pressure, and sentiment in the crypto space cooled down.
But do not directly assume that the rate hike will definitely happen, nor should you take a pullback to 75,000 or 72,000 as a safe bottom-fishing zone.
The August CPI (to be announced on September 11) is indeed a critical turning point: if inflation remains high, rate hike expectations will continue to rise, putting pressure on the crypto market; if inflation drops significantly, rate hike expectations will cool, giving risk assets some breathing room.
However, macro expectations themselves will fluctuate repeatedly. Even if the CPI data looks good, subsequent official statements and non-farm payroll data can still change market pricing. Macro is not a one-time decisive event.
From the market data:
BTC current price is 77,500, with 110 million liquidations in 24 hours; ETH at 2,400, with heavy long liquidations; $SOL long liquidations reached 24.61 million. This round of sell-off mainly cleansed the chasing long positions.
A large number of long positions being swept does not mean that a drop to 75,000 or 72,000 will definitely see buying support. In a high-level oscillating market, support levels can be broken at any time, and bottom-fishing during a downtrend can easily catch you halfway down the slope.
A pullback does not equal a bottom-fishing opportunity. It is necessary to observe the strength of support and capital inflow signals during the decline, rather than pre-setting price points and blindly going long.
With macro headwinds looming, do not assume a pullback is a good entry opportunity. It is safer to wait and confirm signals.
Question: Is the pullback before the CPI release a buying opportunity or the start of a further decline?
⚠️This is just a viewpoint sharing and does not constitute investment advice
$BTC $ETH $SOL The scale of Stellar on-chain tokenized RWA has reached $3.996 billion with a 360% increase this year, reflecting institutional accumulation expansion, but the secondary market $XLM spot support has not yet followed, with the core contradiction being that on-chain asset accumulation has not converted into net inflow of spot funds.
Market facts show a clear lag between on-chain asset expansion and spot liquidity performance. The liquidity driving forces rank as follows: spot net inflow growth rate, on-chain token consumption activity, and the absolute increment of external RWA scale.
The trigger condition for the upward scenario is the rise in on-chain activity activating token consumption, thereby driving a surge in secondary market spot liquidity. At this time, it is necessary to observe whether spot buying can support the valuation premium transmission to $XLM; if external asset growth slows and spot volume shrinks, the upward scenario immediately fails.
The trigger condition for the downward scenario is that the 360% annual increase cannot continue, and the on-chain volume continuously fails to convert into spot buying. At this time, it is necessary to observe whether spot capital flow maintains net outflow; volume accumulation will lose defensive value due to lack of liquidity support, and if spot buying unexpectedly surges, the downward scenario fails.
When the scale of on-chain assets and spot liquidity are completely disconnected, the valuation transmission chain will be cut off. The core basis for judging failure is whether spot support can achieve synchronous rise during capital circulation.
In the next 7 days, focus on observing changes in $XLM spot net inflow, as well as the synchronization rate between on-chain trading activity and new RWA volume.
#Solana通胀缩减提案获投票通过 #黄金ETF大额吸金，避险资金如何重配The 12:30 market is the most misleading: you think you're trading in the right direction, but you're actually at the execution venue. BTC is still tugging around 80,000, ETH and a batch of high-volatility stocks are falling in sync, and the hottest sentiment in the community is still two words: chase or not. Many contract traders focus all their attention on the direction at this point, watching the 1-minute K, the breakout line, and others calling out trades. But what really makes a trade uncomfortable is often not the wrong direction, but opening a position in a position that doesn't suit the trade. The same trading pair may be completely different on different venues. The first difference is depth. During sharp rises and falls, if the order book is a bit thin, chasing the market price will cost you an extra layer of hidden cost; When a stop-loss is triggered, slippage can magnify losses you could have accepted. You see the same candlestick, but the actual transaction experiences are two different price levels. The second difference is the funding rate and fees. Many people only look at the fee at the moment of opening a position, but ignore whether the funding rate will continue to reverse after holding for several hours. In short-term trading, this might just be a few jumps; Once you switch from short-term trading to taking positions, it starts to eat up the margin for error like a timer. The third difference is the mark price and forced liquidation rules. The most counterintuitive part of contracts is that you are not necessarily pierced by the final transaction price, but by the rules. For the same fluctuation, some places just sweep stop losses, while others may directly wipe out the margin safety cushion. When reviewing, simply saying "wrong direction" actually overlooks a crucial layer:In the past month, Uniswap's daily trading volume of stock tokens on Robinhood Chain has increased tenfold.
UNI rose from $2.31 in June to $5.40, a 100% increase in three months.
But this is not a victory for a DEX.
This is a milestone for DeFi being recognized by traditional finance as a "global settlement layer."
Robinhood's dilemma
On July 1st, Robinhood Chain mainnet launched.
Stock tokens for NVIDIA, Tesla, Apple, and SPY were listed for 24-hour trading.
Sounds great, right?
But Robinhood faces two problems:
First, the cost of market making themselves is too high. They have to fund the liquidity for stock tokens out of their own pocket.
Second, the SEC is watching. Traditional brokers listing tokenized securities risk crossing regulatory red lines.
What to do?
Robinhood made a decision: connect non-US retail and stock tokens to a public AMM, not keep them in their own RFQ.
They chose Uniswap.
Uniswap's opportunity
What about Uniswap?
The protocol generates hundreds of millions to over a billion dollars in fees annually, yet the UNI token has been criticized as "zero cash flow"—used only for voting, with no dividends.
In December 2025, the fee switch was finally approved.
Uniswap takes a small cut of the fees, about 6%. This small cut doesn't go to a bank account but into a contract jar called TokenJar.
Who wants to take money from the jar? They must first burn an equivalent value of UNI.
This mechanism is called Firepit.
In short: the official "company buyback" is transformed into "on-chain auction protocol revenue."
Arbitrage bots monitor the net asset value in the jar, burn UNI to extract fees, then sell on the secondary market to complete arbitrage.
The more active the trading, the more UNI is burned, making UNI scarcer.
A perfect match
On the first day of Robinhood Chain's launch, Uniswap v2, v3, v4, and UniswapX were all available simultaneously.
The result?
Uniswap captured nearly 99% of the stock token liquidity on Robinhood Chain.
Cumulative trading volume exceeded $1.5 billion.
Active users surpassed 20 million.
Fees in the past 24 hours reached $4.29 million, nearly half of Robinhood Chain's total.
Single-day UNI burn value once hit $590,000—the highest in history.
Robinhood Chain's TVL surpassed $1 billion within 7 weeks of launch, the fastest-growing blockchain ever.
What's the most ironic?
In 2020-2021, Uniswap was the top star of DeFi.
In this cycle, it fell from grace, with little discussion and a declining token price.
Now, a traditional financial giant has pulled it back.
Robinhood needs to avoid regulation and reduce costs.
Uniswap needs real external revenue to support its deflationary model.
One needs a settlement layer, the other needs traffic. Neither loses out.
The deep integration of TradeFi and DeFi is just beginning.
This is not a victory for a DEX—this is the first "marriage" of DeFi officially recognized by traditional finance.
UNI's surge this round may just be the first chapter of this big story.
$BTC $ETH $UNI Guys, this data is amazing. The FIL long-short ratio has soared again, now at 421 times. Long borrowed 2.3896 million FIL, short borrowed 56.71 FIL. Every short is attacking 421 longs. You read that right, 421 long positions can't beat one short position. 💀 Same script, same outcome. When the price dropped to 0.61 on August 18, the long-short ratio was also over 2,000 times, and the bulls were exposed everywhere, with the price plunging to 0.61. Then the price rebounded to 0.86, and a bunch of people rushed in to long positions, thinking, "This time it must be different." And then? It fell from 0.86 back to 0.68, and the bulls were buried in again. Today's long-short ratio is 421 times, which is a bit lower than the previous 2000 times, but compared to last Friday, it's hardly a 'clearance'—the long borrowing volume still has 2.39 million FIL, and over $600 million bulls are eagerly waiting to break even. These positions are always a mountain weighing down FIL's head; every rebound means someone wants to exit, someone wants to sell. The more long positions you take, the harder it is to rise. 😤 The hardest part isn't losing money, but that every time you think you've hit the bottom, it drops a bit more. 0.70 sounds cheap, right? It fell to 0.68. 0.68 is low enough, right? Then it went back to 0.667. You never know where the bottom is; bottom-fishers are always swinging between "surprise" and "shock." 💡 What can be done? First, don't go against the data. The long-short ratio is 421 times, and the short side only borrows 5671 FIL, so this structure is still lacking$BTC is still in control, but $ETH is starting to pull focus 👀
If Bitcoin chills out and ETH runs, that’s when money rotates and the next leg kicks off 🔄
Watching the $BTC/$ETH pair closely.
Rotation could be coming sooner than people think.
#LaborMarketTestsWalsh #BroadcomDellAIResults
#BTCGoldCorrelation UNI doubled in three months, many only saw the price rise but didn’t understand why it rose
On August 31, UNI broke through $5.4.
Starting from $2.31 in June, it rose over 100% in three months.
Many rushed in chasing the rise, but when asked—"Why is UNI rising?" no one could explain clearly.
Today, I will break this down.
First, a question: Why did UNI keep falling before?
UNI tokens will be fully unlocked in 2024. But the Uniswap pools generate hundreds of millions to over a billion dollars in fees annually—
which have nothing to do with UNI holders.
All fees go to liquidity providers (LPs). UNI can only be used for voting.
A protocol handling trillions of dollars in transactions annually, yet its token has zero cash flow.
This is the "fee switch" debate that has lasted five years—the project team wants to enable it, the community resists, fearing it would affect liquidity.
In December 2025, it finally changed.
A vote passed: a one-time burn of 100 million UNI from the treasury, and the protocol fee switch was turned on.
How does it work?
Uniswap takes a small cut of fees from each transaction. On Robinhood Chain, about 6% is taken.
This money does not go into Uniswap Labs’ bank account.
It goes into a contract jar called TokenJar.
The jar holds ETH, stablecoins, stock tokens—whatever pool the fees come from is stored accordingly. Only in, no out.
The only way to "open the jar": burn an equivalent value of UNI through the Firepit contract.
Arbitrage bots monitor the net asset value in the jar 24/7—
when they find $1 million in assets inside, they immediately burn equivalent UNI, withdraw the fee assets, sell them on secondary markets, completing risk-free arbitrage.
In plain language:
The official "company buyback" has become "on-chain auction of protocol revenue."
It’s not the project team spending money to buy tokens to pump the price—it’s arbitrageurs forced to burn UNI to "grab" the money in the jar.
How scary is this flywheel?
Robinhood Chain mainnet launched in July, and Uniswap was the first public AMM on day one.
On August 30, Uniswap’s daily trading volume of stock tokens on Robinhood Chain reached $130 million, a 10x increase in one month.
In the past 24 hours, revenue was $4.29 million, nearly half of Robinhood Chain’s fee share.
Uniswap accounts for about 99% of stock token liquidity on Robinhood Chain.
Higher trading volume → more protocol fees → more TokenJar assets → more UNI burned → reduced circulating supply → higher token price.
Once this flywheel spins, it is self-reinforcing.
Data doesn’t lie:
Since August, the daily average UNI burn value exceeded $400,000.
On August 21, a single-day burn of 150,000 UNI worth $590,000 set a new record.
By August 31, about 110 million UNI had been burned, totaling $630 million in value.
Annualized burn volume has reached 31 million UNI, worth about $113 million, and is still growing.
What’s the most ruthless part?
Why doesn’t Robinhood use its own RFQ (Request for Quote) system to handle stock tokens?
To avoid SEC regulation and save market-making costs.
They connect non-US retail users and stock tokens directly to the public AMM—Uniswap.
A DEX has become the settlement layer for the world’s largest retail brokerage’s stock tokens.
TradeFi and DeFi are deeply integrating.
Other tokens rely on project teams spending money on buybacks. UNI relies on arbitrageurs "grabbing money" style burning.
This mechanism is unique in the crypto market.
But there is a risk you should know:
Trading volume is cyclical. DeFi activity follows market sentiment.
The flywheel spins fast in a bull market, but burn speed slows in a bear market.
But one thing won’t change—
UNI has transformed from a "voting token" into a "money printer."
A protocol handling trillions of dollars in transactions annually finally has real cash flow backing its token.
The market is simply repricing it.
$BTC $ETH $UNI #就业数据密集公布，沃什政策立场受检验
This Monday, a bunch of U.S. employment data came out all at once: JOLTS, ADP, initial jobless claims, and finally the nonfarm payrolls. Previously, Walsh took a tough stance in his Jackson Hole speech, putting inflation first. Now we have to see if the employment data can support his position.
Personally, I won’t bet in advance on whether there will be a rate hike in September. You can’t decide the final policy based on a single data point; you have to look at the whole set of employment data to see if it’s overall strong or weak.
For my trading, the nonfarm payrolls are the data I value most. The earlier ADP and JOLTS are only references and often deviate significantly from the final nonfarm numbers, which can mislead judgment. When the nonfarm data comes out, it really stirs the dollar and directly drives volatility in Bitcoin and the entire crypto market.
My approach won’t change the original big trading strategy, but I will proactively reduce position sizes. It’s common for the market to jump around and spike before and after data releases, so I won’t take heavy positions to gamble on the outcome. I will firmly set stop losses on my current holdings and won’t hold through losses caused by sudden news. After the data settles and market sentiment is digested, once the real direction is clear, I will consider whether to open new positions.
In the crypto space, if employment data is very strong and rate hike expectations rise, both Bitcoin and altcoins will come under pressure and weaken; if employment cools noticeably, there will be a short-term rebound, but that doesn’t mean the trend reverses immediately—it still needs to be considered together with inflation. Two things overlapped at the end of August:
1. Jackson Hole was hawkish, and the market quickly raised the probability of a September rate hike from about 35% to around 56%, causing BTC to drop from 81,455 to 76,845 that day.
2. The US spot Bitcoin ETF ended a continuous 9-day net inflow (totaling about $3.04 billion), turning into a net outflow of about $202 million on August 28.
Technically, August has already pushed the 200-day moving average back up, which is the strongest bullish argument in this wave. But the 78,670 level is exactly the death zone of the May rebound.
So now it's not about "whether it can still rise," but: Is there a second wave of real spot buying to consume the 81,455 level? If not, watch 74,450 first. The hawks have come back again; this week's table is completely different from last week's.
At this time last week, the market was still celebrating the triple benefits of Treasury buybacks, ETF rush, and short squeeze of 3.1 billion, with $BTC surging to 80,906. A week later, Powell turned hawkish at Jackson Hole, the probability of a September rate hike soared to 55.7%, Brent crude oil prices rose 5.4% in one day, and the stagflation combo directly knocked risk assets back to reality. BTC is now at 78,029, with a 24h range of 77,380-78,135, +0.43%. It looks stable, but actually neither bulls nor bears dare to move.
My view: The nature of this correction has changed. Previously it was "taking a breather after a big rise," now there's a macro variable. Under rate hike expectations, the holding cost of zero-yield assets rises, and the sustainability of ETF inflows is in doubt. 78,000 is the lifeline this week. This is the resonance point of the previous breakout platform plus the 20-day moving average; if it holds, the narrative of 80,000 remains; if it breaks, look down to 74,000-75,000.
Summary of thoughts: During macro headwinds, don't guess the bottom or bottom fish; reduce position by half and wait and see; decide direction after 78,000 breaks. Good news can be late, but rate hikes won't be absent. $BTC
$ETH
$SOL
In a crypto bull market, most investors who are still actively trading essentially aim to outperform the spot price increase of $BTC. In past cycles, among friends and community members, many made aggressive moves, but few ultimately outperformed BTC.
Personally, I adopt a phased grid coin accumulation strategy. For example, currently, the expected extreme pullback level for BTC is above 70K. While holding spot, I place low-position long contract orders and Sell Put orders (a low-buy strategy) below the current price, controlling leverage within 50%.
If the orders are not filled, I collect the option premiums. If they are filled, there will always be opportunities to take profits at higher levels during the bull market.
Since the spot base position is always held, as long as there is no liquidation and BTC remains in a bull market, this strategy will inevitably outperform BTC. Additionally, profits gained from the grid strategy are continuously used to buy some further out-of-the-money Puts as protection against extreme market moves, which can bring unexpected "surprises" during lightning pullbacks in the bull market.
This week's planned order range: BTC 77K-71K, ETH 2.4K-2.2K, SOL 100-92. BTC's relative stability near $78,000 matters more than the quiet headline move. With ETH and SOL lagging over the past day, this looks like selective risk appetite rather than a broad crypto rebound.
I would keep a defensive bias while oil-sensitive US-Iran tensions and labor-market questions remain in focus. A firmer BTC-gold relationship may support the store-of-value case, #LaborMarketTestsWalsh #BTCGoldCorrelation #BroadcomDellAIResults At 3 a.m., four signals resonated: breakout of the neckline, MACD golden cross, 1.8x volume increase, positive news. A textbook-level pattern. I opened a position with more than 5x value. The next day it dropped 3%, the third day below the neckline, and the fourth day I cut to the lowest point. This wasn't the first time. Every time I lost money, my conclusion was the same: if my technique isn't good enough, I'll learn another indicator. After learning MACD, I learned Bollinger Bands, after learning Bollinger Bands theory, after paying tuition once, the account remained the same—slowly climbing, then a line dropped. Trading is gambling. It's not "a bit like," it's gambling. In casino gambling, expectations are inevitably negative because the rules are designed by the householder; In trading, you can make your expectations positive. Admitting this is the start of your profits. Once you accept "no certainty," you stop looking for a winning strategy and switch to calculating the truly useful thing: is this odds worth it? Then repeat it a hundred times. Why? Most people think the reason: the technique isn't good enough. Wrong, this is a downstream problem. Upstream there's only one thing: you're using deterministic thinking to play a probability game. "The signal is so standard, it should go up"—this sentence implies a hypothesis: analysis can turn uncertainty into certainty. Logically, this assumption doesn't hold. If there is a "win as soon as the signal comes out" method, it will be killed by its own validity: the more people use it, the more profits are eaten up the moment the signal is confirmed. So indicators are never used for prediction; their function is to give "bets" a repeatable, well-defined trigger barbasically 0 leverage built up on latest BTC rally, price driven by etf buyers and short squeezes
wouldnt be surprised if it just happens again this week#LaborMarketTestsWalsh#LaborMarketTestsWalsh #BTCGoldCorrelation #BroadcomDellAIResults I have read multiple Bitcoin “insurance policies” in the last 4 weeks from multiple jurisdictions to evaluate them for myself.
None of them are true Bitcoin insurance by my definition or my expectation for when I’m buying “insurance”.
Most of them fall back to relying on the tech stack if ANYTHING happens - so why am I paying for insurance?#LaborMarketTestsWalsh #BTCGoldCorrelation #BroadcomDellAIResults I'm actually not that pessimistic about the semiconductor trend this week. Wash's hawkish tone in his Jackson Hole speech isn't a bad thing; it's the dovish stance that's the real problem 🧐
Hawkishness will indeed suppress valuations in the short term, but for long-duration assets like semiconductors, the outcome might not be bad.
After the speech, the 2-year US Treasury yield rose from about 4.22% to 4.35%, and the probability #LaborMarketTestsWalsh #BTCGoldCorrelation #BroadcomDellAIResults *Tonight's $BTC move is not because of "war" — the market is sending us a bigger signal*
I reviewed tonight's price action.
At first glance, it looks like a risk rotation, with BTC hyped as "digital gold," so it surged 💎
But there's a bug that doesn't add up:
*Gold didn't move at all* 💵
*Data directly contradicts this*
1. *$BTC* rose from *$77,300* to *$78,200*, up *+1.2%* in 24 hours
2. *Gold* is still stuck at *$2,574*, no inflow of funds
3. *#LaborMarketTestsWalsh* employment data was flat, and Walsh didn't make any harsh remarks
If it were truly a risk-off move, gold and BTC should have taken off together. Now only BTC is flying solo.
*So what is the money betting on?*
I think it's these three points:
1. *#BTCGoldCorrelation risk-off narrative is broken* This move is not buying insurance, it's buying expectations
2. *#BroadcomDellAIResults* Before Broadcom and Dell's AI earnings, risk appetite warmed up. Money rushed first into high-volatility assets like BTC
3. *Short squeeze* $77.5K lingered too long, short stop-loss orders pushed the price up to *$78.2K*
*Core conclusion*
*Not a risk-off trade, it's a gamble*
Gold not following = no macro panic. BTC rallying = someone is front-running the next leg Stellar's on-chain tokenized RWA scale has surpassed $3.996 billion, with a year-to-date increase of 360%. The liquidity of institutional base holdings is rapidly expanding, but the on-chain volume has not yet directly translated into spot market support on the secondary market. If new assets can continuously activate token consumption and capital flow, valuation premiums may be transmitted to $XLM. Once the growth of external asset scale slows and fails to drive spot liquidity expansion, this type of volume accumulation will lose its supporting significance. The next focus will be on whether on-chain trading activity and net spot inflows can rise synchronously.
#马斯克回应大摩，3.5万亿美元营收或提前七年 #财政部拟用TGA回购，财政压力仍待化解 #Tectonic遭操纵，Cronos暂停出块