Orbit Post Sitemap
For those who haven't entered $UNITREE yet, don't rush. This damn thing doesn't favor either longs or shorts.
Long positions get trapped, shorts get charged high fees, foreign capital dares not move, and the market maker controls the price with a slow drain. Although I am firmly bearish, I can't endure these fees. The manipulator has been sideways at a high level for half a month. At this daily rate, the principal will be almost depleted in less than a month. Now, the only things left to fight the manipulator are principal and patience.
This kind of manipulation is rare in the community, and such disgusting tactics are uncommon. If it had exploded upward, I would have no complaints. I just hate losing to fees because I chose the wrong direction 😭🤬🤬🤬
#CircleQ3营收超预期，USDC链上交易量增长49%
#IREN矿企转型AI云服务，GPU交付进展待揭晓 #توقعات Sony Earnings Report: How Did the Gaming and Music Sectors Perform? I expect Sony to excel in its profits this quarter; the gaming sector and the weak yen will help it, and I am with yes How should we talk about $OKB? Actually, it has the lowest market cap among platform tokens. BNB's market cap is close to hundreds of billions of dollars, HYPE has over twenty billion dollars in market cap, while OKB, as the platform token of the second largest centralized exchange, only has a market cap of just over two billion dollars! Ultimately, it's because the deflationary mechanism wasn't handled well. BNB directly burns tokens quarterly, HYPE uses over 80% of its revenue to buy back and burn tokens, but OKB is only treated as the GAS token for X Lyer. When the ecosystem was incomplete, few people participated, so the total supply of OKB is 21 million, and it has hardly decreased at all! I think if OKX put more effort into consuming OKB, OKB would have already hit new highs and wouldn't have been stuck at this two billion market cap for so long! Of course, a low market cap also means great potential. In the future, when OKX's ecosystem develops or OKB starts buybacks, its value will definitely soar. Overall, OKB is still worth going long on!The U.S. midterm elections on November 5 reshaped the power structure in Washington, with the Democratic Party regaining control of both chambers of Congress, and the Republican Party losing full control over the White House and Congress. On Wednesday, market risk appetite rebounded, with the S&P 500 rising 1.6% and the Dow Jones up 3%. Trump expressed willingness to accept the election results and cooperate with the Democrats.
The Senate confirmation power shifted to the Democrats, who are expected to limit the White House's ability to alter the Federal Reserve Board's structure through new appointments; the Senate Banking Committee led by Elizabeth Warren may strengthen oversight of Chairman Warsh and monetary policy transparency. ADP data showed private sector job additions of 86,000 in October, exceeding the expected 69,000, with wage growth at 3.2%, and wages for job changers up 4.8% year-over-year. JOLTS job openings increased by 333,000 to 7.523 million, while layoffs fell to 1.641 million, indicating labor market resilience.
The next key market focus is the November 6 nonfarm payrolls report (expected increase of 65,000, unemployment rate 4.2%); if employment remains stable, the Federal Reserve may continue to wait and see, but if it weakens significantly, the market may raise expectations for dovish policy. The Reserve Bank of Australia raised rates to 4.85% this week, reflecting ongoing global inflationary pressures. Overall, the election reduces some political uncertainty but does not eliminate the contradictions between monetary policy and economic fundamentals.In-depth analysis of the non-farm payrolls: Will the December rate hike suspense heat up? Which trading direction should be taken?
On November 6, the US October non-farm payroll report is about to be released.
This is not only a test of employment data but may also become a key variable determining the Federal Reserve's monetary policy direction in December.
Currently, the core contradiction facing the market is: Is the labor market strong enough that the Federal Reserve must continue to raise interest rates?
According to data released by ADP on November 4, the US private sector added 86,000 jobs in October, higher than the market expectation of 69,000, showing that the labor market still has some resilience.
This non-farm payroll report focuses on three key indicators:
• New non-farm payrolls (NFP): Growth exceeding expectations will reinforce employment resilience and increase pressure to raise rates.
• Unemployment rate: If it continues to decline, it means the labor market may still be tight.
• Average hourly earnings: A rebound in wage growth will exacerbate inflation stickiness and may become an important reason for the Federal Reserve to continue tightening policy.
Market logic:
• Strong non-farm payrolls + rising wages: December rate hike expectations heat up, US Treasury yields and the dollar may strengthen, growth stocks under pressure.
• Weak non-farm payrolls + rising unemployment: The necessity of rate hikes decreases, possibly easing market tightening expectations.
• Strong employment but cooling wages: Policy signals may diverge, and it cannot be simply regarded as positive for rate hikes.
Conclusion: What truly determines the market direction is not the new non-farm payroll number itself, but whether employment resilience and wage inflation both exceed expectations simultaneously.
#非农在即，就业数据能否强化12月加息预期？ $UNI rose 18% in a week, while $LTC is busy building bricks for nursing homes. The absurdity of this crypto circle is really a bit ridiculous.
$UNI protocol fees have started burning, and DEXs continue to nibble away at CEX market share. Even Robinhood and Kraken's public chains are using it as underlying liquidity. They are already eyeing the traditional finance meal ticket.
Looking at $LTC, the project team used $62 million to buy a physical nursing home. The secondary market is still struggling near the head and shoulders neckline. Token holders are waiting for the ecosystem to take off, but the project team seems to be arranging retirement life in advance.
One is busy turning protocol revenue into token value, the other is busy expanding the elderly care business. Who is the capital really rewarding? The market is already starting to show its stance.
#UNI
$UNI
👇I see this long-short ratio like watching a hotpot restaurant queue: the bulls' table hasn't cleared, just not as crowded as yesterday, and the bears finally don't need to bring their own folding stools 😂
OKEx contract data shows the $BTC long-short ratio dropped from 2.17 24 hours ago to 1.51 now, meaning for every 1 short account there are 1.51 long accounts. I think this indicates a cooling off in bullish crowding, not a solid signal for bears. The ratio is still above 1, so there are still more long accounts; plus it counts accounts, not positions—many small positions and one large position mean voter count and capital weight could be totally different.
My judgment is: sentiment changes have reference value, but using it directly to guess direction is overdoing it. Do you stand by “bullish cooling is healthier,” or “bullish confidence is loosening”? Has this table just removed a few chairs, or is it about to break up?Jeff Yan repeatedly emphasized in an exclusive interview with Bloomberg: Hyperliquid is not an exchange; it is the infrastructure layer that supports exchanges. It sounds like PR talk, but it actually explains all the product trade-offs.
Traditional centralized exchanges pack three layers into one company: the platform manages accounts, maintains matching, and decides which markets to list. Hyperliquid separates these three layers: users control asset permissions with address signatures, the on-chain order book manages matching status, and third-party teams deploy their own terminals and applications at the market layer. The official defines HyperCore as the core execution environment for spot, perpetual order books, and margin status.
Why not use AMM? Anyone can add liquidity to AMM pools, but high-leverage perpetuals need to handle continuous quoting, depth, liquidations, funding rates, and extreme market bad debts, which pools cannot withstand. Order books clearly display quotes and depth, supporting limit, stop-loss, and market orders. The trade-off is that the underlying layer must handle matching order sequence, margin, positions, and liquidation in high-frequency orders.
So this is not about moving a CEX on-chain; it’s about breaking down a company’s backend work into verifiable protocol states plus composable application layers.
"Users truly own their assets" is only half true. Platforms can’t freeze withdrawals, but if your private key leaks, no one can recover your funds; if you sign the wrong authorization, you lose directly; if the front end is spoofed, on-chain execution still happens authentically. The value of $HYPE is not that there is no risk, but that the risk is borne by a different party.#Tether旗下XAUT黄金储备增长59%
The boss has something to say
I only caught the full meaning when I looked at this news together with this morning's gold price movement: Tether disclosed that as of the end of Q3, the physical gold backing XAUT increased to about 25.7 tons, a 59% increase from the beginning of the year, with a reserve value of about $3.44 billion, setting a historical record. At the same time, data from the World Gold Council showed that in Q3, global central banks net purchased 321 tons of gold, a 42% year-on-year increase, and gold ETFs had a net inflow of 211 tons; after a pullback, spot gold climbed back above $4000, even briefly surpassing $4100 this morning.
My interpretation is: The $3.44 billion backing XAUT is still just a fraction compared to the scale of USDT. The real signal is not from Tether, but from the combined buying of gold by "central banks, ETFs, and tokenization" capital flows. After the midterm elections, the dollar weakened, and gold, as a non-interest-bearing safe-haven asset, rose first, essentially discounting the dollar's credit; this same macro logic is also the underlying reason why BTC dared to surge above 80,000 with volume these past two days—gold and BTC are two columns on the same ledger, one slow, one fast.
BTC is currently priced around 80,400. I have already taken profits on my 30x long position today and am now watching the market with no position. My stance remains unchanged: do not chase highs above gold $4100, and similarly do not chase highs above BTC 81,000; wait for a pullback to stabilize around 79,000–79,500 before entering long again. The busier the market, the more you should wait for a pullback—make money from market sentiment, don’t stand guard for the market. $BTC $ETH $ZEC Order Book Strength Ranking
5-minute median slippage, estimated by order book
$APP cost differences mainly appear in large buy orders. Simulated slippage for 10,000 scale buy/sell is 0.16%/0.18%, and for 100,000 scale is 2.27%/1.54%. Small amounts on both sides are close, large amounts are significantly more expensive on one side.
$AKE large buy order costs decrease, sell order costs increase. Simulated slippage for 100,000 scale buy/sell changed from 1.94%/0.97% to 1.52%/1.10%. Depth changes are asymmetric, with different conditions on buy and sell sides.
$CAP large buy order costs increase, small buy orders remain almost unchanged. Simulated slippage for 100,000 scale buy increased from 0.83% to 1.07%. 10,000 buy remains about 0.15%. Slippage rises, order book is thinning.BTC broke 80,000, but this is not the horn of a bull market — it's the wail of shorts being liquidated.
The Democrats' sweep means "uncertainty landing" (short-term bullish) + "stricter legislation" (long-term bearish) for crypto, hedging both ways. Whether 80,000 holds depends on tonight's US stock market.Oil prices crashed, BTC laughed: a chain reaction triggered by an election
As soon as the election results came in, Jefferies played two cards: canceling Trump's tariffs and ending the Iran war. The market instantly exploded—crude oil plummeted over 8%, the US dollar and Treasury yields both fell, while gold surged against the trend.
On the surface, it looks like a political shock, but at its core, it's the collapse of inflation expectations. When oil prices crash, the chain "oil price rises Seeing listed companies continuously buying Bitcoin does give me a bit more confidence, but I don't plan to follow their pace. Companies can raise funds, while my dollar-cost averaging relies on leftover monthly spare money; these two sources of capital are quite different. Even though we're buying the same coin, it's hard to just copy their moves. However, I'm quite interested in listening to their shareholder Q&A sessions, especially after the market cools down. I wonder if anyone in the audience will directly ask: Are we still doing our original business?Democrats Sweep Congress: AI Data Center Construction Faces Regulatory Headwinds
In the 2026 U.S. midterm elections, the Democrats gained control of both the House and the Senate, meaning the AI data center industry will face greater policy uncertainty.
The Democrats' regulatory pressure mainly focuses on four areas:
1. Energy Constraints: Requiring data centers to bear the costs of new power generation, transmission, and grid upgrades themselves, and even requiring large data centers to build their own power supply.
2. Environmental Regulation: Strengthening scrutiny of water consumption, carbon emissions, and environmental impact, raising project approval thresholds.
3. Construction Restrictions: Some Democratic lawmakers have proposed suspending AI data center construction until new safety, energy, and environmental regulatory frameworks are established.
4. Tax Policy: Promoting the elimination of some tax incentives for data centers, increasing the actual cost of capital expenditures.
The core conflict lies in the fact that the AI industry needs to continuously expand computing infrastructure, while local communities are increasingly resentful of the electricity price hikes, resource consumption, and environmental burdens brought by data centers.
· Investment Impact: The Democrats sweeping Congress does not mean AI infrastructure will immediately halt, but regulatory scrutiny, capital expenditures, and project approval uncertainties will increase.
· Especially for data center projects that heavily rely on grid access, land approvals, and long construction cycles, there may be risks of delays or even relocation.
· Short-term negative impact on data center operators and some #美国中期选举：民主党赢得国会两院控制权 Crowding Index Observation: What Are the Rates Indicating?
This period's long-short crowding index shows two types of divergence.
$AKE price rose 1.23%, open interest increased 2.28%, but the 4-hour rate is -0.1038%, roughly -0.623% daily, with -0.112% realized over 24h, near the median in the past 7 days, about 236 minutes from settlement. Despite negative rates, price continues to rise and positions keep increasing, indicating shorts are not only under pressure but are also continuously drained by the rate, showing strong crowding.
$CAP shows a different kind of pressure: price fell 0.29%, open interest increased 0.67%, 4-hour rate +0.0757%, roughly +0.454% daily, +0.157% realized over 24h, relatively high in the past 7 days, about 236 minutes from settlement. The simultaneous rise in rate, weakening price, and increasing positions indicate that long position costs are accelerating accumulation.
$SOXL price dropped 1.49%, open interest basically unchanged, 8-hour rate +0.0519%, roughly +0.156% daily, +0.141% realized over 24h, relatively high in the past 7 days, about 476 minutes from settlement. Currently, it looks more like a weak consolidation under expensive rates, with positions not yet significantly loosening; if positions shift from flat to shrinking, it can be interpreted as a clearing signal. Overall, whether the rate is expensive or not is just a clue; the key to whether crowded trading can continue is if price and open interest keep moving in the same direction.
#索尼财报前瞻：游戏与音乐业务表现如何？
#非农在即，就业数据能否强化12月加息预期？ When I look at crypto prediction markets, sometimes before I even finish reading the settlement conditions, my eyes already follow the popular results, and I think to myself, surely so many people can't be wrong. Unlike just watching coin prices, here you have to see if the specific event can actually happen according to the conditions. The odds include judgments but may also be mixed with money following the trend, so you can't take them as everyone's individually researched opinions. I then go back to check the reasons over there, and after looking for a long time, even the wording is almost the same.💵 Is a stablecoin pegged to 1 USD necessarily safe?
Not necessarily.
When evaluating stablecoins, you can't just look at the price; you also need to pay attention to three key issues:
🔎 Reserves: What mechanism does it rely on to maintain value?
🔓 Redemption: In case of anomalies, can assets be redeemed according to the rules?
🛡️ Technology: Are there additional risks in smart contracts, wallets, and networks?
If you put stablecoins into DeFi or yield products, you also need to consider extra protocol and liquidity risks.
Price stability does not equal risk elimination.
What do you think should be the top priority when evaluating stablecoins?
#ONEA #Web3 #StablecoinsKey on-chain signal! A long-term whale dormant for a year decisively takes profits during this round of rebound 🐳
During this round, Bitcoin and Ethereum violently reversed in a V-shape and just recovered, and the dormant whale that had been silent for a whole year suddenly woke up and moved.
Within nearly 10 hours, this address was the first to redeem staked tokens from $AAVE, then concentrated on-chain selling followed, totaling $6.357 million worth of ETH + stETH sold, with a stable average transaction price of 2428.
A very key point: a long-term position untouched for a year won’t be operated casually.
Such dormant whales represent extremely patient long-term capital. Taking profits right as this rebound starts indicates that from the main players’ perspective, the current price level is already a prime profit-taking zone.
The short-term market looks strong, but the whale cashing out first also indirectly reminds us: there is real selling pressure above, so the rebound height should not be overly optimistic, and chasing highs requires caution. Just finished lunch feeling sleepy, but after checking last night's ETF data, I instantly woke up. The contract rally looks lively, but in reality, institutions have huge disagreements.
$BTC had net inflows of $9.63 million after four consecutive days of outflows; BlackRock fled with $26 million, relying entirely on Fidelity's $30 million to support the bottom.
$ETH shows apparent net inflows, but BlackRock made large sell-offs, a typical distribution at highs.
$SOL ETF funds are directly at zero, leaving only retail contracts to speculate.
Contracts reflect sentiment, spot funds are the trump card. Although the BTC ETF has stopped falling, the divergence remains; don't be blindly lured in by candlesticks.
What do you all think—can BTC hold above 80,000 this time? Long or short? #AAOI高速光模块放量，Q3财报能否超预期？ Shrunk from over ten million to over six million, it's not a total loss
The account dropped from over 10 million U to over 6 million U.
Positions still have an unrealized profit of over 300,000 U.
What do these numbers mean: the cost of three orders is all below the current price.
$SNDK cost 1223, current price 1377.
$SOXL cost 80.61, current price 108.
$SKHY cost 134.62, current price 157.12.
Where did the money go: unrealized profit is just a book figure; if not sold, it’s not realized.
When the position shrinks, the margin that can be supported shrinks accordingly.
If the price drops further, reducing positions is passive.
Most likely it’s not a wrong direction call, but the position size was too heavy.
The more consistent the earnings expectations, the easier it is to be pushed at this moment.
Stop losses set shallow have already been triggered.
#美股AI板块走势分化：云服务商走强、数据中心股承压 $SNDK $SOXL On November 1st, I posted about $AKE here, saying that the funding rate was continuously negative and the open interest was increasing against the trend. Today, all three numbers have reversed.
$AKE current price is 0.00247, down 88.6% in 9 days, only 1.7% away from the September 20th high of 0.1467.
① Leverage liquidation: OKX perpetual 24-hour volume is 76 million USD, open interest only 2.97 million USD, down 67% from 9.09 million on October 20th, turnover 25 times, all short-term chips.
② Paper liquidity: Top 20 order book levels total 62,000 USD, entire order book 630,000 USD, daily volume is 120 times that.
③ Selling pressure not finished: Total supply 100 billion tokens, only 22.8 billion circulating, 2.11 billion tokens unlocking on November 21st, about 9.3% of circulating supply. After the last two unlocks, prices dropped 22% and 49% respectively within a week.
Another signal: funding rate has been negative for four consecutive periods, latest -0.123%/8h, shorts are crowded; OKX only has perpetual contracts, no spot, 76 million perpetual volume is 4 times the global spot volume of 18.65 million, no spot anchor.
I don’t chase this kind of drop: the unlock calendar is not finished, the order book is paper-thin.
At this level, do you think anyone will still catch it? $AKEUnder the surgical light, $30 trillion worth of "qualified donors" lie on the operating table, but the hearts with actual blood perfusion number only $3.8 billion — this is not a shortage of capacity, it's treating the donor pool as cardiac output.
Morgan Stanley was among the first to access the US Depository Trust & Clearing Corporation's securities tokenization service, with 34 US stocks, ETFs, and government bonds gaining on-chain rights. Institutional transfers, on-chain settlement, and collateral management are all integrated. From my perspective, this is not a heart transplant; it's connecting an old heart to extracorporeal circulation: the circulation pathway is established, perfusion is maintained, but extracorporeal circulation itself causes inflammatory responses and coagulation disorders, and every minute of heparinization is a gamble.
The most fatal preoperative misjudgment is mistaking anatomical potential for physiological perfusion. That $30 trillion is the donor list lying in cold perfusion solution, not beating myocardium. What truly determines prognosis is the on-chain blood flow of $3.8 billion, vascular resistance, and whether the anastomosis leaks. The angiographic results of the 34 securities only confirm the pathway is unobstructed, far from guaranteeing transplant success.
The linkage of S&P semiconductor leveraged targets is more like adding positive inotropic drugs to already fragile myocardium — blood pressure rises, oxygen consumption increases, and the threshold for arrhythmia lowers. The moment collateral is revalued on-chain is equivalent to withdrawing part of the preload at the instant the heart restarts; the waveform on the monitor looks good until ventricular fibrillation occurs.
My judgment: this is not a new heart, but a left ventricular assist device. It does not cure heart failure; it only maintains blood flow, and the three challenges of pump thrombosis, driveline infection, and right heart failure are unavoidable.
The real risk never lies in the nominal $30 trillion donor volume, but in acute rejection during the first hour post-transplant — when the custodian's monitoring electrodes are not yet attached, the on-chain collateral has already undergone a round of revaluation. #morganstanleyjoinsdtccTalking about $BTC, I believe the ultimate price and long-term value of Bitcoin. First, we have witnessed Bitcoin reaching $126,000 per coin, with a total market cap of about $2.5 trillion at that time. Silver is currently at $3 trillion, but silver's peak was $7 trillion. Gold is incomparable, as it has been the world currency for thousands of years and is held as reserves by central banks worldwide. Even if Bitcoin reaches fifty years old, compared to the understanding in 2000, it is still only between 0 to 3 years old. The goal and consensus among whales for Bitcoin has always been to replace silver, which has industrial attributes, as the special currency number two. Therefore, considering a $7 trillion peak and a long-term value of $1 to $2 trillion, Bitcoin's peak price should reach $350,000 per coin. The current Bitcoin price is like the international silver price at $28 per ounce. There will come a day when a price of over $100,000 per coin will be considered a bottom, and seeing Bitcoin at $80,000 per coin will be like looking at silver prices three years ago.【Forking a chain for $6.8 million?】
On November 5, Binance and Polymarket jointly submitted a governance proposal: to implement a hard fork on the Polygon network to recover approximately 70.49 million POL tokens (about $6.8 million) stolen assets — including 4.65 million tokens related to the Binance case ($4.51 million), and 23.86 million tokens related to a previous third-party supplier breach at Polymarket. Details: Polygon block producers have actually frozen transfers from these 5 attack addresses, but freezing can only "lock" the tokens, not "return" them — only a hard fork can transfer the tokens back to the recovery wallet. The proposal does not change the total supply of POL, requires governance voting approval, and can only be executed after testing on the testnet; there is no mainnet date yet.】Why it matters: This is the first time the "code is law vs. human intervention" debate has reached the L1 governance level — the ghost of Ethereum's 2016 DAO fork has returned. The fact that Polygon validator nodes can freeze transfers already answers the question of "how decentralized this chain really is." If the proposal passes, it will become a template for asset recovery in the industry; if rejected, it will be a constitutional-level statement.As soon as the planning authority tightened the wind, those projects still lying on the blueprints waiting to start construction saw their valuations collapse halfway—what happened on November 4th was not capital picking stocks, but the review agencies picking structures.
Microsoft touched a historic high, Amazon followed suit, and to me, these are two supertall buildings that have long been structurally topped out and fully inspected: the main load-bearing system is crystal clear, the mechanical and electrical shafts are reserved, the seismic rating is maxed out. Now if you add two more floors of machine rooms on top, a load calculation check will pass easily, and the marginal cost is almost just interior finishing. What they’re selling is not a future rendering, but shear walls already poured.
Meanwhile, the Philadelphia Semiconductor Index is going down, and operators like IREN and Hut 8, who rely on new construction projects, are being held down—that’s a different story—they are subcontractors whose general contractors haven’t even arrived on site, with only a third of the foundation piles driven. No matter how beautiful the blueprints are, once planning conditions change, with lowered floor area ratio and height restrictions, the rebar can only rust outside the site. The supply ceiling for new computing power isn’t suppressed by market sentiment, but by approval red lines and construction permits.
The real variable after the midterm elections is that everyone starts redoing geological surveys: can this data center site still be approved, how high can it be built, and can the integrated pipelines pass? This revaluation immediately splits assets into two categories. One is existing infrastructure—already connected to the grid, passed fire inspections, with redundant load-bearing systems. If policies change, at most they add one less floor, but the main structure remains. The other is chip suppliers and data center operators relying on new projects—their cash flow is fully staked on the next construction permit. If approvals slow and schedules drag, financing costs start eating profits. No matter how strong the prefabricated component factories are, if the general contractor doesn’t start excavation, orders just become inventory.
This is the oldest lesson in the industry: white papers are design drawings, design drawings are worthless; what’s valuable is how deep the foundation is dug, whether the load-bearing walls cut corners, and how many redundancies the structure has. What the market is doing now is a belated rebar inspection—peeling back those skyscrapers propped up only by renderings, layer by layer, to check the reinforcement ratio inside. Policy uncertainty is not a negative; it’s just an X-ray revealing which buildings have cast-in-place concrete and which are just dry-hung stone veneers.
What truly determines the direction of this round of differentiation is never the emotional swings of news headlines, but whether those cloud giants can continue to build upward on existing plots, and after new computing power supply is blocked, whose existing load-bearing systems can still handle the next wave of loads. Where the review stamp can’t be affixed, no matter how beautiful the tower is, it’s just a model.
In this business, what can withstand planning changes is never the renderings, but the shear walls already poured to the top. #aistocksdivergeShort Squeeze Night: After $1 Billion Liquidation, Don't Mistake Forced Liquidations for New Buying
In the past 24 hours, the entire network saw $1.004 billion liquidated, with 135,000 people liquidated. Unlike previous times dominated by long positions, this time shorts liquidated $571 million, longs $434 million, with shorts liquidating $137 million more.
BTC was the hardest hit, with $408 million liquidated, 75.6% of which were shorts; ETH liquidated $248 million, also dominated by shorts. Within 12 hours, $621 million liquidated, with shorts at $425 million, concentrated in the stampede after the election results were announced. In the following 4 hours, longs and shorts shifted to $17.17 million vs. $14.83 million, entering a tug-of-war. The largest single liquidation came from HTX's ETH, a $37.99 million large short forcibly liquidated.
The essence of this round of market action is a "short squeeze": bad news is fully priced in, shorts panic liquidate, pushing prices up. But the liquidations are forced liquidations, not new long entries. Short squeezes often rise fast and fall fast; after shorts clear out, if no buying follows, a pullback can happen anytime. The 130,000 liquidated also shows leverage remains crowded. Don't chase highs before tomorrow night's nonfarm payrolls.
$BTC $ETH
#非农在即，就业数据能否强化12月加息预期？ BTC holds at 80,000, SLX talks about access, Micron still depends on cash
$BTC rose 5.5% in the past 24 hours, slower than some smaller coins, but it remains an important benchmark for the market. The volatility of large coins and small coins differs; simply ranking by gains can overlook the risks each bears.
I use it to judge whether the market rebound will continue and then assess changes in various projects. How far the rise can spread across assets is more interesting than lining up all coins by the same gain ranking.
$SLX’s utility goes beyond governance. After converting to stSLX, you can gain access to certain products, governance weight, and some credit qualifications. Demand must grow from these use cases.
So I ask whether using these rights is truly necessary. Writing rich rules is one thing; users’ willingness to lock tokens long-term to obtain qualifications is what really matters.
$MU’s corresponding Micron contract is pulling back, but after adjustments in Q4 of fiscal 2026, free cash flow reaches $33.2 billion. Cash has already been received, yet the price can still fall.
This reminds me to separate operations from expectations. The market buys how much profit will be made later; you can’t judge short-term drops by cash flow already realized in one quarter. Growth sustainability is the real question for the latter stage.
ADA is approaching 0.20 again, up 7.2% in the past 24 hours, but the high range of 0.2015 has not yet been surpassed. The round number and previous high are close, making every small fluctuation seem dramatic.
I prefer to see if the price can stay after sufficient trading around that area. Touching it, pulling back, and truly moving upward are three completely different patterns.📊 Paper Performance: Far Exceeding Expectations, Full-Year Guidance Raised
Sony's Q2 (Fiscal Year 2026) operating profit surged 45% to ¥622.3 billion, far surpassing market expectations. There are two core engines driving growth: Games & Network Services, and Film & Television Business.
• Games & Network Services: Operating profit ¥220.1 billion, up 82.9% year-on-year, the largest profit engine.
• Film & Television Business: Operating profit ¥92.6 billion, up more than fivefold year-on-year, mainly driven by the box office boom of "Spider-Man: A New Day."
• Full-Year Guidance: The company raised its full-year operating profit forecast by about 9% to ¥1.87 trillion, with full-year dividend expectations also up 40% year-on-year.
⚠️ Hidden Concerns to Watch Closely
Although the data looks good, chasing the stock just because "operating profit exceeded expectations" carries significant risks.
• Cost pressures loom: The AI boom has pushed up memory chip prices, putting cost pressure on Sony's gaming business; the entire consumer electronics industry is troubled by this, with Apple also warning of similar risks this week. However, Sony stated it has locked in the memory supply needed for this year.
• PS5 sales clearly declined: PS5 sales for the quarter were 2.6 million units, far below last year's 4 million units. Profit growth in the gaming business is more from software and network services, while hardware is weakening.
•#非农在即，就业数据能否强化12月加息预期？ On the radar echo map of the duty room, a strong echo of 65dBZ is continuously accumulating over the Golden Plate. The data just reported by Tether exploded like a weather balloon: XAUT's physical gold reserves surged to about 25.7 tons by the end of Q3, a sharp increase of 59% compared to the end of 2025, with a valuation of approximately $3.44 billion. This is not a drizzle; this is a high-pressure ridge stably maintained over precious metals.
Looking at the circulation background. The World Gold Council simultaneously observed that central banks worldwide net purchased 321 tons of gold in Q3, a 42% year-on-year increase, and ETFs added another 211 tons. Cold air and warm moist airflows are simultaneously delivered, with water vapor flux maxed out. Spot gold reclaimed the $4000 mark on November 4; the previous pullback resembled a shortwave trough passing, with rapid reconstruction of isobars after the trough and before the ridge. Neither the central bank nor ETF currents have withdrawn; tokenized gold is equivalent to adding a batch of automatic stations to the existing observation network—sampling frequency is higher and coverage denser. The question is whether it can also absorb the off-site precipitation under observation.
On the US stock side, the linkage of the tokenized target XBMNR is worth monitoring. It behaves like the near-surface wind field during a frontal passage, reacting half a beat faster than the upper air but with an unstable direction. If the main gold ridge continues to hold, it will rise with the momentum; if an upstream cutoff low forms, it will be the first to turn around. Currently, the dual currents formed by gold ETFs and central bank purchases remain, but the inflow through the tokenized channel depends on whether the pressure gradient is steep enough.
There is no sign of clearing in the upper air situation yet; the gold stratification is still thickening. Whether the tokenized station network can catch this wave of moisture depends on the next round of sounding. This is not a sunny day; this is a high-pressure ridge redefining the entire circulation pattern. #xautgoldreservesup59%The rebound is not a reversal. Is it now a phase of chasing gains, consolidation, or accumulation? Let me set the tone first: this looks more like an emotional recovery phase after an oversell, not a trend confirmation phase. BTC returned to $80,693, up 4.66%, ETH rose to $2,429, up 7.18%, and SOL returned to $95.98, up 4.18%. The gain leaderboard looks lively, but what I’m watching is who’s leading the rally—ETH is clearly ahead, which indicates that capital is willing to move a bit further out on the risk curve, but it’s not yet a full-scale offensive 🌙. What’s really worth dissecting is the event layer. Circle’s Q3 revenue exceeded expectations, and USDC on-chain transaction volume rose 49%. Many people just brushed this off as an ordinary earnings report. I understand its importance in two points: first, the on-chain usage intensity of stablecoins is recovering, indicating that settlement and real demand haven’t gone dormant; second, it gives a breath of life to the "compliant stablecoin narrative," which is exactly the story institutions want to hear. Another point is that OKX launched USDC trading pairs for xIREN and xASML, with tokenized stocks continuing to push into trading scenarios. Putting these two things together, the market is quietly repricing the "on-chain dollar + tokenized assets" line, but the price hasn’t fully reflected it yet. I see the transmission path like this: stablecoin data warming up first benefits payment and infrastructure narratives, then assets like ETH that carry settlement activities, and finally the sentiment spreads to altcoins. So ETH leading the rally is no coincidence.First, the current situation: the market has priced in more than an 80% chance of a 25 basis point hike in December (CME about 81%). The ADP small nonfarm payrolls came in above expectations again, with the private sector steady at 86,000 for two consecutive months, the service sector holding up, and wages accelerating with job switching. Under this combination, as long as the official nonfarm payrolls are "not too bad," the rate hike expectations will be hard to drop.
My judgment: it is very likely to strengthen rather than overturn.
As long as new employment is significantly higher than the previous value, the unemployment rate does not exceed 4.3%, and hourly wages increase by more than 0.2% month-over-month, the market will turn the "December rate hike" from an expectation into a more certain pricing. The dollar will be slightly stronger in the short term, gold and rate-sensitive assets will come under pressure, and risk asset volatility will increase.
What you really need to be cautious about are two extremes:
A surprise miss (new jobs far below expectations + rising unemployment) → the probability of a rate hike could quickly fall below 60%, creating greater short-term trading opportunities.
A strong surprise (new jobs 100,000+ and wage stickiness) → not only locking in December but also pushing up the pricing for another hike in 2027.
For trading, don’t just look at the numbers tonight; watch the "deviation relative to expectations" and subsequent adjustments. The first 30 minutes after the data release are often driven by sentiment; the real pricing comes afterward. Being prepared for both sides is more important than betting on a direction. #非农在即，就业数据能否强化12月加息预期？ On the evening of November 6 (U.S. will release the October nonfarm payroll report). The market consensus expects an increase of about 65,000 jobs (previous value only 29,000), unemployment rate to remain at 4.2%, and average hourly earnings to rise about 0.3% month-over-month.
The preceding ADP "small nonfarm" already exceeded expectations with an increase of 86,000, mainly contributed by the service sector, and the wage growth for job switchers reached the fastest pace in over two years. The latest CME FedWatch shows the probability of a 25bp rate hike in December has risen to about 81.5%, with the probability of no change at only 18.5%.
The conclusion is very clear: if the official nonfarm payrolls significantly exceed expectations (especially if the increase breaks 100,000, hourly earnings remain sticky without decline, and unemployment rate stays stable or slightly decreases), it will further strengthen the narrative of labor market resilience, giving the Federal Reserve more room to focus on the inflation target. The expectation of a rate hike in December will move from "high probability" to "almost certain," and may even raise the pricing for additional hikes in 2027.
Conversely, if a September-style weakness reappears (increase below 40,000 and unemployment rate rises to 4.3%+), it may quickly lower the probability of a rate hike, putting pressure on short-term interest rates and the dollar. The current baseline scenario still leans toward strength, but the key is not "whether there is growth," but "whether the growth is sufficient to support further tightening."
Before the data release, the market is already pricing in a "resilient employment + sticky inflation" combination. What can truly change the path is a result that deviates significantly from expectations. Pay more attention to hourly earnings and unemployment rate, which are more important than simply looking at the number of new jobs. #非农在即，就业数据能否强化12月加息预期？ Data Perspective: Why is the Market Betting on "Beating Expectations"?
• Profit Far Exceeds Expectations: Adjusted loss per share is only $0.51, while the market expected a loss of $2.15, a significant "beat." This improvement in profitability is the most direct source of confidence for bettors.
• Revenue Barely Holds: Total revenue is $1.02 billion, down 45% year-over-year, but still much higher than the market expectation of $860 million.
• Strong Cost Control: The company has significantly cut expenses through layoffs and pipeline reductions. R&D expenses dropped to $801 million, and sales and administrative expenses fell to $268 million. The CFO stated that operating costs were reduced by 34% year-over-year, and the cash reserves give confidence in reaching the breakeven target by 2028.
⚠️ Concerns: Cracks in the High-Growth Logic
• The COVID-19 Vaccine Base Has Collapsed: The revenue beat is due to "saving money," not "selling more." The flagship COVID-19 vaccine sales were $971 million, down 46% year-over-year. The blame for the decline in vaccination rates is hard to shift.
• RSV Vaccine Faces a Waterloo: As the second growth curve, the RSV vaccine mResVax performed poorly, with Q3 sales of only $2 million, down 80% year-over-year, far below the market expectation of $30 million. Facing pressure from GSK and Pfizer, this business has fizzled out before it could take off. 1. Currently, SanDisk's open interest (OI) is extremely high; the more it falls, the more positions are added, and the OI hasn't dropped much, with a significant rise, nearly 70% up since November 2nd, and the funding rate is climbing positively.
2. According to the thoughts in the major holders' comment section on the semiconductor index, many are still buying on the rise, entering a momentum mindset, and the high negative funding rate of SOXS indicates that those going long on SOX are still the majority.
3. After ETH's OI surged significantly at night, the funding rate climbed positively; the spike and drop in OI happened during the upward wave, indicating heavy short covering and many shorts getting wiped out. Essentially, no new long positions from outside have entered. Now it depends on whether the spot market will sell or not. #美国中期选举：民主党赢得国会两院控制权 After $BTC surged, it entered a high-level consolidation, which easily equates to a short-term sideways stabilization and directly implies that the bullish trend has been established.
Personal market observation: The current price is about $80,425, with a 24-hour fluctuation range of $78,006.3 to $81,200. This rebound wave, starting from the low point of $75,734.5, is quite significant, but multiple attempts to break through the $81,200 level have failed to sustain, indicating persistent resistance above.
Before the price effectively stands above this previous high, the risk-reward ratio for chasing highs is not favorable. Repeatedly going long within the consolidation range, each small pullback accumulates and erodes risk capacity, so risk control boundaries must be clearly defined in advance.
My judgment: Only if there is a volume-backed steady hold above $81,200, with solid buying support on pullbacks, will the upward space for the rebound further open; if pressure continues and the price falls below internal support of the range, this strong rebound needs to be reassessed. Without a clear breakout signal, it is not suitable to actively add positions. Will you wait for breakout confirmation before entering, or rely on support to try light positions? This is just for opinion exchange.
#CircleQ3营收超预期，USDC链上交易量增长49% $TRUMP
At 2 AM, I wrote about short positions trapped between 1.22 and 1.25; looking back, I was only half right. People are trapped, the price didn’t rise to free them, but it also didn’t continue to crash.
At 8 AM, that 1H candle dumped 158 million contracts and $19.9 million down to 1.238, then for over seven hours it hovered between 1.244 and 1.292, with volume dropping from 1.2 billion contracts to 31.67 million. The daily low of 1.218 was still left by that 1 AM candle and hasn’t been broken since.
In 24 hours, it opened at 1.508, touched 1.524, probed 1.218, current price 1.258, down 16.6%, with perpetual contract volume at $336 million.
With midterm election results out, analysts are saying it’s good for stocks and bonds. Political coins don’t buy that. Their perpetual open interest is only $17.75 million, daily volume is nineteen times the open interest, all short-term traders going back and forth, no one holding real positions.
The funding rate is negative 0.022%, more than twice the benchmark, shorts pay longs every eight hours, so holding short positions isn’t easy either.
The story is told, positions are still here, and this small gap in the middle is why it’s currently moving sideways.
#非农在即，就业数据能否强化12月加息预期？ Seeing the election results become the main theme for asset pricing again gives me a sense of déjà vu from the last cycle: once the macro narrative heats up, the community starts looking for new sectors to connect with it. According to this observation from MSX, with the Democrats taking both the House and Senate, the Dow Jones rose 3.03% on the first trading day after the election, with healthcare and clean energy strengthening, while oil and defense sectors came under pressure.
What’s different this time is that US stock RWA has brought sector rotation into on-chain discussions. I think this is more like a test of the ability to research sectors, rather than evidence of a demand explosion. The divergence in sector performance means that simply talking about “US stocks on-chain” is no longer enough; we need to explain the policy expectations for specific assets. Also, the rise in US stocks itself does not prove that on-chain users and liquidity are increasing in sync.
In the context of the cycle, I prefer to see this as RWA shifting from focusing on ease of entry to competing on asset understanding. Moving forward, I will focus on real participation rather than rushing to count the traditional market’s excitement as on-chain prosperity. U.S. stocks surged, but the Philadelphia Semiconductor Index plummeted?
1. Macro overview: No more war, the market celebrates wildly
The Democratic Party opposes war with Iran, and there is opposition within the Republican Party as well. Now that the Democrats have swept both chambers, Trump's authority is weakened, and Iran is very unlikely to continue fighting. With the war threat reduced, oil prices fall, and the Federal Reserve doesn't need to rush to raise interest rates, which is positive for the entire stock market.
2. Industry phenomenon: Data center construction halted, tech giants soar, chip stocks crash
Why is it good for the giants: The Democratic Party opposes building new data centers (due to public opposition, fearing it will raise local electricity prices and lower housing prices). If construction stops in the future, existing data centers become scarce assets, directly increasing their value! Companies like Amazon, Google, Microsoft, and Meta that own data centers soar.
Why chips plummet: No new data centers are being built, so chip companies can't sell their products, causing the semiconductor index to crash.
3. Pharmaceutical sector: Government pays the bill
The Democratic platform is to restore Obamacare (the government pays for healthcare), causing pharmaceutical companies to surge.
The Democrats have always supported new energy; last night, U.S. new energy stocks performed well, but funds are temporarily busy buying the "Seven Sisters."
Lower war expectations support the market, halting data center construction benefits existing giants but hurts chips, and healthcare stimulus favors pharmaceutical mergers and acquisitions. $xSOXL #AAOI高速光模块放量，Q3财报能否超预期？ $ETH
9.16 billion vs. 8.39 billion. Today, ETH perpetual contract trading volume surpassed BTC, a scene rarely seen this year.
In 24 hours, ETH rose from 2302 to 2451, current price 2421.6, up 5.2%; BTC only 2.9% in the same period. At 09:00, a 1H candle released 1.33 million contracts, $323 million, then for the next six hours, the price squeezed within 2406 to 2436, a 30-point range, with volume dropping to 120 to 190 million.
Where did the money come from? Yesterday, Ethereum spot ETF net inflow was 53.02 million, after five consecutive days of net outflow. Also, one address apparently built a $14.76 million position in three days, now floating profit is 477,000, implying a cost around 2346.
No signs from the perpetual side. Fee rate 0.0044%, only 43% of the benchmark 0.01%; open interest $1.54 billion, about the same as this morning. ±1% order book buy side 92.58 million vs. sell side 67.12 million, buying is strong but not frenzied.
My interpretation: This wave is driven by spot and ETF money lifting, leverage is still waiting outside. This move is unlikely to crash suddenly, nor should we expect acceleration—real acceleration requires fee rates to return above 0.01%.
#非农在即，就业数据能否强化12月加息预期？ There was an on-chain movement today, the numbers aren't large but very clean: a new address transferred out 6,277 ETH from Binance hot wallet in two transactions, totaling about 14.76 million USD. The first transaction was 3,000 ETH, about 6.8 million USD, and the second was 3,277 ETH, about 7.96 million USD.
First, the nature of this. A hot wallet is the address an exchange uses for daily withdrawals; coins leaving here means this portion of ETH has left exchange custody and is no longer sitting on the order book waiting to be sold. This is a mechanistic conclusion, no guesswork needed.
What’s worth pondering is the recipient. The address is new, with no history or recharge traces, and it received all 6,277 ETH in just two transactions. Retail withdrawals usually happen in one transaction and in smaller amounts; this operation looks more like setting up a self-custody wallet.
But can this be taken as a bullish signal? No. The most common misinterpretation on-chain is "withdrawal equals bullish." Withdrawal only proves the coins are off the exchange; it doesn’t prove the holder intends to hold long-term—they can deposit back anytime, which takes just a minute.
14.76 million USD is not significant in ETH’s daily trading volume; a single transaction can’t support trend judgment. Its value lies in sample significance: if later a batch of similar new addresses withdraw coins, that would indicate institutions are setting up wallets in bulk, and that would be worth portfolio adjustment.
One question remains: when we say "exchange balance decreases," how much is real exit, and how much is just coins moving to another wallet to sit? On-chain, these two look exactly the same.$ETH This afternoon's surge upward made my heart skip a beat. The 5-minute candlestick just twice pierced the upper boundary of the 2432 range. However, both times it failed to hold, closing back inside the consolidation range. The chart showed two UTADs, followed by a bearish engulfing candle.
During the second upward push, the trading volume reached 1.5 times the average volume of the range. Despite the volume increase, the price couldn't be pushed higher, indicating heavy selling pressure above. Bulls were blocked at the top of the small-scale box, and the buying momentum started to weaken. The price surged up then dropped back down; this kind of movement is unfavorable for the bulls in the short term.
Looking at the 4-hour chart for position context, the current price is right near the 2440 4-hour resistance level. This barrier has been tested 4 times before, each time quickly pulling back after touching it. The 4-hour EMA lines are intertwined sideways, and the larger trend has not yet formed a clear directional move.
I am slightly bearish in the short term. The price is hitting major resistance overhead, and short-term UTAD rejections keep appearing. If buyers can't push up with strong volume, the market is likely to pull back and test support.
$BTC $SOL #美国中期选举：民主党赢得国会两院控制权 $BTC spot net inflow +$136.6 million, but contract net outflow of $79.88 million — this is not a shortage of funds, but funds choosing direction. The most worth watching in the mid-term capital structure is this kind of mismatch: "real money entering, leverage actively pulling back." First, look at $BTC: spot buying is positive, contract positions are negative, indicating incremental funds prefer physical holding, while leverage side is cooling down. This structure usually means chips are transferring to low-leverage accounts, with support below the price, but short-term explosive power is suppressed. Stability is its keyword and also the reason it temporarily cannot break out into a one-sided market. $ETH is the most active and also the one to watch out for among the three. Spot +$88 million, contracts +$156 million, both sides of funds are expanding simultaneously, with leverage accumulation clearly faster than spot. The flexible side is also fragile: once the price fails to break upward, the accumulated long leverage will become reverse fuel, and the risk of a bull trap is brewing here. The direction choice of $ETH depends on whether spot can continue to keep pace with contracts. $SOL tells a different story: spot +$11.25 million, contracts -$12.01 million. Spot is willing to buy, leverage is watching, a divergence between the two. This "spot leads, contracts lag" combination often appears on the eve of a trend reversal — not yet started, but waiting for confirmation. The focus for $SOL is not current strength or weakness, but when leverage will replenish. Putting the three together, the conclusion is clear: the market is not short of money, it lacks consensus direction. $B BTC pulled from 75,600 to 81,100, rising over 4% in 24 hours, with the total market cap of the entire network returning to 2.7 trillion. News headlines are all about "rebound," "recovery," and "election market." But if you open the capital flow, you'll see another world— US spot BTC ETFs have had net outflows for three consecutive days, totaling 1.157 billion USD. The price is rising, but institutions are exiting. This is not a bull market. This is someone using the corpses of shorts to pave a rBitcoin cycle indicators show that the current pullback may not be over yet. Market structure points to a cooling cycle but has not yet reached the extreme conditions typical of major turning points. The 30-day moving average of the Bitcoin cycle extreme index has dropped to 15.8%, within the "watch/mild stress" range. Bitcoin is currently trading near $80000, down 46% from its previous all-time high. Although the retracement is significant, this index remains far below levels seen in active or extreme phases. The cycle extreme oscillation indicator also supports a cautious view, with its 30-day moving average near the bottom of the historical range and well below the 1.0 threshold, showing no clear upward signal at present. The indicators suggest that Bitcoin is in a pullback and sluggish phase, rather than a confirmed cycle bottom. #交易之声：你的经验值得被听到 $BTC On the night the midterm elections concluded, the U.S. stock market showed a rare split: the Dow Jones rose 3.03%, marking its largest single-day gain since April last year, while the Philadelphia Semiconductor Index fell 3.13%, dropping back to a six-month low. The situation is actually straightforward: Texas accounts for about 27% of the country's planned data center capacity, and the new governor intends to suspend approvals. With data center construction slowing, those selling the shovels get hit first, while cash-rich giants save on capital expenditures. The same amount of money is a cost to one side and income to the other.Is $CAP being manipulated by the same operator? It's so strange $SOXS is one ticker, with long and short positions separated. When the long position falls, it doesn't mean the short position rises; when the short position falls, it doesn't mean the long position rises. The reason for separating them is that most people will go to the long side, so the short side rises.
Analogy 2/1
Now it has become 4/1. I don't understand, why is that.Apple's Q4 revenue and profit both exceeded expectations, but the market is really pricing in the "next quarter's gross margin guidance under pressure." The company gave a gross margin forecast of 45%-46% for the December quarter, a significant decline from this quarter's 47.9%, and also below analysts' consensus expectations. The CFO clearly stated in the conference call that memory prices are at an "unprecedented" level, and previously low-priced inventory is being consumed, with cost pressureI see this as somewhat positive, but not enough to support chasing the rally. On November 4th Eastern Time, Bitcoin spot ETFs had a net inflow of $74.5677 million, with Fidelity's FBTC leading with a net inflow of $237 million.
The net inflow of a single product far exceeds the total net amount, indicating that other products combined had outflows, so this cannot be taken as a sign of a comprehensive demand recovery. My judgment is that if the inflows continue and spread, a bullish stance has a winning chance; if you only bet on single-day data, the odds may not be favorable. I will wait for broader capital validation.Last night, the US stock market was quite lively, but semiconductors were singled out and took a hit 😮💨
The S&P 500 rose 1.6%, but Micron $MU, SanDisk $SNDK, and Marvell all fell more than 7%, and Nvidia was not spared either.
I think this round of decline should be viewed in the context of the US midterm elections:
The Democrats won the Texas governor's race, and the new governor has explicitly advocated suspending new data center approvals until stricter regulatory rules are introduced.
This is a major shift:
In the past, the market priced AI chips with the implicit premise that tech giants would continue to invest huge capital expenditures and data centers could keep expanding.
Now this premise is being challenged.
The power consumption, water usage, and impact on local residents' lives from data centers are becoming increasingly sensitive political issues.
I tend to believe that this sell-off is more about the market reassessing the expansion speed of AI infrastructure; there is not yet sufficient evidence to show that actual computing power demand has significantly deteriorated.
After all, projects already contracted and under construction will not disappear overnight, but future incremental investments do face greater policy uncertainty.
Going forward, I will pay more attention to tech giants' capital expenditure guidance and the actual approval policies implemented by various states.
The AI story is still ongoing; it's just that the market is starting to seriously calculate how fast this story can continue to develop.
#美国中期选举：民主党赢得国会两院控制权