Stocks Plunge Across the Board Today as Korea Hits Trading Halts
Samsung Just Released Its Q2 Earnings Guidance: Revenue of 171 Trillion KRW, Profit of 89.4 Trillion KRW—Great Data. AI Has Somehow Pulled the Memory Cycle Back Up.
If the data is this good, why did it fall?
Because the market isn’t trading “good or bad.” It’s trading whether it’s better than what people expected. The rally had gone too smoothly for a while—money had already priced in the “memory super-cycle” expectation. Samsung slammed the good numbers on the table, which instead turned into an “earnings are out—so sell” window.
So it’s not that the fundamentals got worse. It’s that expectations were fully priced in, and the market is crowded at high levels.
Now people are shouting “it’s topping out.” When the stock is up, everyone studies the supply chain—especially HBM capacity. When it drops, nobody cares.
What’s actually happening:
The supply chain hasn’t changed. HBM is still in short supply. SK hynix and Micron are both pouring hundreds of billions into expansion. Leopold’s fund is even preparing to subscribe for $7 billion of Micron ADRs.
This doesn’t look like the kind of action you’d expect at the top of a cycle.
Tonight’s key validation point: PENG’s earnings report. It can answer one question—when memory prices rise, is it just a chip-level issue, or has it already transmitted into the real orders for AI infrastructure?
If PENG is strong: then today’s drop is just high-level rotation and turnover—not the end of the main story.
If PENG is weak: then we need to rethink whether the downstream demand has been pulled forward and already over-consumed.
Also, Micron is set to list on the US stock market on the 10th. Timing a sell-off right before that to let large funds pick up shares cheaply also makes sense.
My view: the cycle hasn’t ended, but in the short term it’s definitely shaking people out. This round of leverage and chasing is going to feel miserable.
Fear makes people stop thinking and just want to run—but when the market drops is exactly when you should study.
Yesterday, Spain and Belgium both won—strong teams keep pushing forward
Spain beats Portugal. C罗’s final World Cup ends here—at 41, his stamina and breakaways can’t keep up. Spain’s midfield control is too strong; Portugal was pinned down and beaten all match.
Belgium 4-1 USA. The host is eliminated. After the match, the whole Belgium squad jumped into the Trump dance. In this World Cup, all three hosts were knocked out in the Round of 16—home advantage isn’t as effective at this level.
🔮 Today’s predictions:
1⃣ 🇦🇷 Argentina vs 🇪🇬 Egypt
No real suspense here. Messi’s last World Cup—Argentina’s overall form is on point. Egypt’s defense can hold up, but on the attacking end they don’t have anyone who can change the course of the match.
Prediction: Argentina win.
2⃣ 🇨🇭 Switzerland vs 🇨🇴 Colombia
This one is more complicated. Colombia played more aggressively in the group stage, with multiple attacking options. Switzerland is steadier and more conservative in style. But Switzerland has plenty of knockout-stage experience, so it’s not easy for them to collapse.
Leaning toward Colombia, but not highly certain.
These are the last two matches of the Round of 16. After that, into the quarterfinals—all matchups will be between strong teams, so strategies have to change.
@predictdotfun
Welcome to join us at predict and play World Cup together: https://predict.fun?ref=3AA0C
Just finished building an automated market-making system for a prediction market with xBubble. From writing the requirements to going live took less than 10 minutes ☑️ Lately I’ve been playing on Predict: placing prediction orders for both profit and to earn project points. Before this, I was manually judging the direction and placing bets.
On X I saw @ZhanweiC open-sourced a set of automated market-making strategies for prediction markets. I reviewed it and thought the idea was great. Since I’m also playing on Predict, I just took it and deployed it. But when I looked into the current approach, it required writing scripts to deploy servers and manage a watchdog process. I’m in trading—I don’t know how to code, so I got stuck. I remembered using xBubble before for a video, and they also have a programming feature. I opened the chat interface, described the rules in natural language, and it worked: “Help me build an automated market-making system for a prediction market. Only trade the next two matches. Cover win/draw/lose and the spread handicap (big/small goals). If the order book depth for Buy 1 is greater than 500, place a Buy 1; if not, place a Buy 2. Check the order book depth every minute and automatically adjust. Stop buying 6 hours before the start, only sell/exit; 6 minutes before kickoff, cancel all orders. If I have positions, automatically place sell orders.” Very quickly it generated everything and deployed it for me. I didn’t need to set up an environment, configure a domain, or manage the server. Cursor solves “how to write code.” xBubble solves “how to get the business running.” It’s not just code—it’s an entire business service from idea to deployment. All those middle steps like server/domain, payment gateway and API configuration are handled too. For someone like me who has a strategy but can’t deploy it, you can basically copy and go live. This approach isn’t only for market-making. Prediction sites, arbitrage tools, and sports event data dashboards—any time-sensitive requirement has the same core problem. The video shows the whole deployment process from zero. If you’re also interested in prediction markets, you can take a look. @dappOS_com @xBubble_ai
Brazil got eliminated by Norway—I bet on Brazil winning and lost big, it hurts.
Brazil’s five-star team reaches the Round of 16 and goes home. They controlled the ball 34% the whole match, the lowest in World Cup history. After the game, Neymar cried right away. Haaland alone tore through Brazil’s defense. This World Cup really has a lot of upsets.
England won 3-2 against Mexico, though. A five-goal thriller—my heart couldn’t take it. Mexico went 4 for 4 in the group stage, won all games and kept clean sheets; yet in the Round of 16, they went home. That’s football.
📊 Current record: direction 5 for 4, hit rate 80% 💸 But the loss on the Brazil game nearly made me give back everything I’d earned so far. The strategy of backing strong teams just doesn’t work well when you run into monsters like Haaland.
───
🔮 Today’s prediction:
1⃣ 🇵🇹 Portugal vs 🇪🇸 Spain
This match is truly a clash of top teams. It might be Cristiano Ronaldo’s last World Cup.
Neither side is an easy opponent. Spain’s overall coordination is smoother and their control is stronger. Portugal relies on individual talent. But Ronaldo is 41—his condition is a question mark.
I lean toward Spain, but I won’t go heavy on it. In a strong-vs-strong matchup, I’ll only take a light position to share my view—I’m not putting my whole stake on it.
2⃣ 🇺🇸 USA vs 🇧🇪 Belgium
The home advantage is clear for the USA. Belgium’s golden generation is getting old. De Bruyne has already played his last match. I’m leaning toward the USA in this one.
⚠️ Yesterday’s lesson: don’t blindly trust strong teams. In knockout rounds, teams with a superstar are the biggest variable.
Yesterday I placed bets on France to win and on Morocco to win on Predict, and both came in.
So far, the strong-team strategy in the knockout stage has been doing quite well. In the Round of 16, when a strong team faces a weaker one, a draw isn’t allowed—there must be a winner. The strong team’s depth and bench strength get amplified in these single-elimination, life-or-death matches.
Today there are two games:
• 🇧🇷 Brazil vs 🇳🇴 Norway • 🏴 England vs 🇲🇽 Mexico
Brazil are clearly stronger—keep backing Brazil.
England vs Mexico — this one is a bit tricky. Mexico went 4-0 in their group stage with zero conceded goals—their form really is great. England were down to Congo until the 75th minute in the Round of 32 before they turned it around. They played very struggle-ridden football. For this match, I’m leaning toward not betting, or possibly a small bet on Mexico.
What’s the safer way to bet:
Only bet on matches where the skill gap is obvious. Games like Brazil vs Norway, where the certainty is high. For matches like England vs Mexico, which are more of a 50-50, either don’t touch them or keep the stake light.
The further you go in the knockout rounds, the more frequent strong vs strong matchups you’ll get. The strong-team strategy works best in the Round of 16; by the time you reach the quarterfinals, you’ll need to switch your approach.
The core of the strong-team strategy is: bet only on matches with high certainty. If there’s any doubt, don’t bet.
I recently bought a cat to keep me company. I also bought some vitamin D to make up for the lack of sun exposure. I got a few books to see how other people live.
I've always thought of myself as pretty easygoing, but lately it seems like I've been having some mental health issues.
I feel very troubled and strange. It's like my whole self is trapped in nothingness—my body feels weak. Even though I'm taking medication for anxiety, those are different. All along, I haven't really been leaning toward depression; it's only been physical symptoms. I honestly find it hard to believe that I'm depressed.
Weekly active users have exceeded 120,000. This week added 70,000—up from just 20,000 last week. In one week, it’s tripled.
TVL hit an all-time high at 29M. The single-match trading volume for Argentina vs. Cape Verde reached 10 million USD for the first time.
From the group stage until now, the line has kept getting thicker. Liquidity is clearly much better now—this is the result of users pouring in.
The World Cup is a catalyst, but prediction markets already have demand. People watch games and want to guess. If there’s a judgment, people want to place bets—no need to educate users.
The later the knockout stage, the fewer the matches, but the higher the attention. Single-match trading volume will only keep rising.
Also, on Predict, buy a single bet of 50 USDT or more to get 5 Binance Alpha points. The promotion runs until July 6.
Binance’s 9th anniversary merchandise has arrived!
A huge box—an outerwear jacket, a backpack, phone charms, scarves, pins, and more. The giveaways have come one after another: for the Dragon Boat Festival, then for DJI Pocket 4, and Binance’s gifts never stopped. My mom said the company is really generous.
The outerwear jacket is genuinely great—perfect for my recent plans to get outdoors. The backpack is good too.
I’ve already hung the phone charm on my phone. I might use the scarf to match and style my outfits.
Every time I open a Binance package, it feels like a relationship with a platform—somehow like a connection with a person. There’s accumulation, there’s memory, there’s warmth.
I’m one of the earliest seed users. It’s been nine whole years now.
Many projects showed up and disappeared during these nine years—exchanges especially. They’ve changed hands again and again, with whole waves of people.
But Binance has always been there.
There’s so much I’ve done on Binance. My first time buying BTC was on Binance. My first time trading with contracts was on Binance. My first liquidation was on Binance too. My first big profit came there as well. And the first crypto-related gift I received was also sent by Binance.
It’s almost the carrier of my entire crypto history.
From 2017 to now—I went from someone who knew nothing to a person who runs nodes, then starts writing content, and now helps answer questions for people around me. Through this whole journey, Binance has been there.
Hoping that the next nine years, I’ll still be doing this—and Binance will be doing this too.
Thanks to @binancezh for another year—@heyibinance @yayabinance @sisibinance
Seeing Brother Niao’s solid “long resistance” makes me think about early 2025: when the leader pledged ETH to borrow U to buy coins, then gradually staked on-chain—only to eventually not be able to withstand the pressure and started manually taking losses one after another. The direction was wrong. It looks like even a small amount of leverage can be terrifying—worse than making short-term high-multiple bets. Here’s some leverage—so strong it can literally lift the Earth and make it explode.
So I was already free—this is the conclusion I drew from chatting with AI.
I don’t have much money, but I have complete freedom to manage my own time.
I skipped the long cycle of “trading time for money, then using money to buy back time” and went straight to the final destination—absolute control of time.
And it even told me a story.
A wealthy man went on vacation to the beach and saw a fisherman sunbathing. The wealthy man urged the fisherman to work hard, buy a big boat, open a company, and make a lot of money. The fisherman asked, “After I make a lot of money, what then?” The wealthy man said, “Then you’ll be able to lie on the beach like me and sunbathe comfortably.” The fisherman smiled: “But I’m doing that right now, aren’t I?”
So it turns out I’ve been doing nothing every day on the beach, sipping coffee, because I’ve already reached life’s ultimate goal—it’s just too boring 🥱
🏄Deepcoin this draw, Argentina co-branded merch two sets Congratulations 🎉 to the two of you—please check your DMs @Baoer_BTC @Nvwueth
Thanks to @deepcoin_cn for sending over a complete set of World Cup co-branded merch: jerseys, caps, a ball, and a backpack. The quality is better than I imagined—everything you could ask for. My niece was absolutely in love with the ball and couldn’t put it down.
The main purpose is to capture launches like RAVE LAB in a timely manner. Sometimes, we follow the trends by tracking movement. Using a tight stop-loss strategy with small positions, recently the anomaly/momentum trades have been going pretty well. The general logic behind the movements is that AI watches the price of the copycat, funding, OI, and on-chain activity. Before, we used to let my friends and some clients use it, and everyone’s feedback has been pretty good. We’re preparing to open a group, and release the signals for free to the public for a period of time to test. For the first batch, we won’t let in too many people. If you want to join, DM me and include the note “signal”. You can DM a friend’s ID or join the boss’s chat room.
帮主聊天室
乔帮主退休月球收租
·
--
The Yocoin system for vibing with friends has iterated through several versions It has very high accuracy—basically all market movements are caught Who says tokens are wasted in a bear market and have no use? Anyone want to try it?
Developed a counterfeit “anomaly” AI signal. The main purpose is to catch the launch of coins like RAVE LAB in time.
Sometimes, by following the anomalies, we try to ride the trend. Using a small-position, tight-stop-loss strategy to chase anomaly trends has been doing pretty well lately.
The general logic behind the anomaly is: AI watches the price of the counterfeit, the flows (funds) and OI, as well as on-chain anomalies.
Previously, we only provided it to friends around us and some clients, and the feedback has been good.
We’re getting ready to create a group and release the signals for free to publicly test for a period of time.
For the first batch, we won’t let too many people in. If you want to join, DM me and include the note “signal”.
Fitness check-ins—lost 4 jin! But for now the food is a bit too monotonous, and I can’t eat much. My strength for anaerobic training isn’t enough. Countdown: 2 months and 20 days!!!
River S6 is back on track again. From July 1 to September 30
I’ve held this project for more than a year. Every season they’ve been distributing airdrops. Even in the sixth season, they’re still doing it—this kind of consistency is rare.
Key points of S6:
1. New product direction — River is going to extend into the AI track, though the specifics haven’t been announced yet. 2. RIVER staking — Stake to earn points. Points can be converted into staked RIVER, which gives voting power. Voting power lets you earn points again every week. It’s a positive flywheel. 3. satUSD is still the main foundation — Deploy cross-chain, provide liquidity, and participate in DeFi protocols to earn points. 4. Cooperative Campaign — Throughout the whole season, there will be activities with different partners, with multiplier bonuses. 5. River4FUN — Posting content is still the same: earn points.
The admin has been participating since S1. Here, I also racked up an A7. The most reassuring thing about this project is that it keeps moving forward steadily according to schedule, orderly and methodical. After six seasons, it’s still actively building.
It’s fine to join if you want—there’s no high barrier, but what matters most is坚持 (consistency).
After the World Cup knockout stage gets underway, there are more and more on-chain activities.
Recently I’ve been playing @NeoSoulAI’s Fantasy World Cup—super✍️ simple: make a single on-chain interaction to mint an NFT for a knockout-stage team. The further the team goes, the more points you earn. The champion team’s NFT can have up to 15k points.
You can mint on both BSC and 0G. If you’ve already interacted on 0G and then do another on BSC, your rewards stack. Also during the World Cup, all on-chain interaction points get +20%.
The entry threshold is very low—you can mint just by doing one on-chain interaction.
My plan is to pick two or three teams from different tiers to diversify.
For a popular team like Brazil, pick one. For a dark horse like Morocco, pick one. If the dark horse ends up going far, the value is the highest.
In the knockout rounds, every round is filtering teams. For each match your chosen team wins, your points roll over once. The later the rounds, the more valuable it is. Now is a perfect time to join—right when the 32-team round just started, before too many rounds have been played.
Join in and mint along with your World Cup participation as a little bonus. If you end up picking the champion, getting the points for free is very tasty! ! !