Master WU

“You only find out who is swimming naked when the tide goes out” -- Warren Buffett “You only find out who is the real stock trading master in a bear market.” -- Master Panda Wu

MPW Bamboo Scroll #232 Posted: (1) later next week, we are going to witness something rarely happen but with some significant consequences. (2) it is a process and we are all in for it--let's brew some good coffee and just wait and see. (3) more details thekobeissiletter.com/panda

I just spoke to Mitch McConnell on the phone. He told me he’s dead.

Max out U.S.A.'s credit-worthiness. The payment will be due in the future.

MPW Mid-Week Update Posted: (1) a large W-C may have started, based on my primary count of SPX; (2) Also worth noting is that Korea's erratic stock index KOSPI has formed a complex H&S pattern, which is still working on the right shoulder. (3) check at thekobeissiletter.com/panda

JUST IN: Trump says Belgium is 2 weeks away from developing a nuclear bomb.

$MU=912 now, 62$ from my target of 850.

Couple of love taps only, best deal you ever seen still on schedule.

🚨US job market revisions are crazy: US non-farm payrolls for May and April were revised down by a combined -74,000, in the latest June report released on Friday. Since January 2025, non-farm employment has been revised in 15 of the last 17 months. During this period, job numbers were adjusted down by a net -778,000. Since January 2023, 31 of the last 41 months have been revised down. What is happening here?

MPW Bamboo Scroll #231 Posted: (1) quite an amazing time we are living in now, with so many historic events unfolding at the same time. (2) watching World Cup in China rekindled my momeory of watching it the first time, in 1982. (3) check updates here thekobeissiletter.com/panda

MPW Mid-Week Update Posted: (1) a more complex W-B prolongs the lingering time between 7500 and 7400 for SPX. (2) That said, the appointment has been delayed, but the meeting place remains the same spot. (3) check here for update at thekobeissiletter.com/panda

$MU targets 850: (1) ever after the much-anticipated & super-"good" earnings report, $MU has returned to reality, now testing DMA-20 at 1050. (2) a proper retrace to $850 is actually healthy & bullish for it mid-term, to reset those ugly TA indicators. (3) short since earning.

Trump made 3,642 trades in Q1 2026. That annualizes out to be 14,568 trades a year. Or just about 58 trades a day. Clearly someone in his camp is daytrading on his behalf but guhhh damn, 58 trades a day?! Lol For comparison sakes, Nancy Pelosi made roughly 12 trades in 2025.

Two Paths: (1) the bleeding was stopped last Friday with the a last second gap-fill needle, and the pattern has changed to a more complex path. (2) the light blue one is an expanding diagonal, finishing the 5-wave serie; whereas the purple points to a triangle-type consolidation

BREAKING: The Japanese Yen collapses to its weakest level versus the US Dollar since 1986.

Iranian negotiators still think they are going to get any money back from Forrest Trump - this whole thing makes you having the feeling you are watching the movie Dumb & Dumber

Tasnim--so far, it has been proven to be the most reliable source of truthful information during this entire conflict. For the rest, whenever you see a "source at the White House", that is a tellsign of a lie.

In the same month the United States treated its most powerful frontier AI models like a weapon and pulled it from the world, China answered by giving its rival model away for free, trained on its own chips, downloadable by anyone with a laptop. Two opposite theories of power collided in thirty days, and the cheaper one may have physics on its side. The trigger is one benchmark, narrower than the headlines suggest. An independent security firm tested China's new open model, GLM-5.2, against a mainstream version of Anthropic Claude on a single class of software bug. The Chinese model won, 39 percent to 32, at roughly one-sixth the cost per vulnerability found. It was never pitted against Anthropic's restricted cyber model Mythos, and it still trails the best American systems on the hardest reasoning. But on this task, the open underdog beat the closed leader, for pennies. How it was actually built by the Chinese is what should stop the room. This Chinese Ai model GLM-5.2 was reportedly trained on 100,000 Huawei chips, with zero American hardware in the loop. A frontier-class model, off the US supply chain entirely, free to download, modify, and run. Its maker's stock leapt 42 percent in a day and crossed a trillion Hong Kong dollars. Underneath sits the question everyone circles. American Ai labs accuse Chinese firms of distilling their models, training a cheap copy on an expensive original's answers. The technique is quite real, it works, and every major Ai lab uses a version of it. The dispute is whether the copying broke the rules, not whether copying works. And that is the irony of the wall America just built. The one thing you cannot stop a model from doing is being learned from. You can gate the weights and bar the borders, but a clever student can study the outputs and approach the original for a fraction of the price. Yup! America is betting intelligence is a treasure to be locked away. China is betting it is a flood that cannot be dammed. We are about to find out which one was right. Wild times ahead!

This is uneqivocally Bullish for the market, as "ceasefire" and "peace-deal" is imminent [based on the logic of the past, NTA]. In the meantime, "The third Blood" is coming....targeting sub-7200 on Monday.

The largest building boom in history is running on a loan that has not been taken out yet. OpenAI is spending hundreds of billions on chips and data centers against a single assumption … that the public markets will one day hand it the cash to pay for it all. This week that assumption slipped a year, and the entire AI complex shook, because everyone glimpsed the IOU holding up the floor. The chain is startlingly short. In June, SpaceX went public at a colossal valuation, then slid all the way back to its 150 dollar starting price within two weeks. That single wobble was enough to make OpenAI's bankers warn it could not command a 1 trillion dollar debut in this market. So OpenAI is now reported to be leaning toward pushing its IPO to 2027. The reason one delay rattled the world is the math underneath it. OpenAI lost roughly 38 billion dollars last year against 34 billion in total costs, much of it compute, and it plans to spend 600 billion on hardware by 2030, a target so large it requires hundreds of billions in capital it does not yet have. The IPO was the plan to raise it. The moment it slipped, analysts and traders flagged the obvious risk: a delay threatens the pace of the very infrastructure buildout the whole market is betting on. So look at what fell. Micron, AMD, Arm, SK Hynix, SoftBank, the entire chain that sells into AI, dropped together. Not because earnings were bad. Micron had just posted records. They fell because every one of them is ultimately selling into demand funded by an IPO that just got postponed. The most expensive buildout ever made is leaning on capital it has not raised, spending against a payday it has only promised. One space stock dipped, one IPO blinked, and for a moment the market saw the floor for what it is. A promise.

Retail day trading data is brutal: - 90/90/90 rule: 90% of new traders lose 90% of their capital in 90 days - Fewer than 1% remain predictably profitable over 5+ years - One study found only 0.88% of active day traders were consistently profitable - Brazilian futures study: only 3% made any profit - Only 1.1% earned more than minimum wage - ~72% of retail day traders lose money in a given year - 40% quit within the first month - Only 13% remain active after 3 years - Margin traders average negative ~5% returns - 95% of traders still come back after blowing up Why? - They cut winners too early - They hold losers too long - They overuse leverage - They pay spreads and slippage - They confuse luck with skill - They compete against firms with faster data, better tools, and billion-dollar infrastructure Retail day trading isn’t investing. It’s walking into a casino and calling yourself the house.