🚨Ladies and gentlemens, horsemen,Corporate drones, clergymen, and dedicated lovers: we have the perfect storm ... the trifecta. Oil is breaking out, yields are breaking out, and the inherently bearish rising wedge that has been depicted on the S&P 500 is not the only bearish technical flashing warning signs, but it is the most relevant for those conservatives looking for that perfect triple confirmation. There is a saviour: the white knight. It is the deleveraging event that is needed. Without central bank intervention, we need a mauling ... a mauling that will rip the heart out of the market, ’08 style. The excess, the recklessness, the dysfunction is so distorted that even the seasoned traders are having difficulty positioning in the current market. The next couple of weeks are critical, and something capricious is brewing. Stay vigilant, perhaps a little curious, but most of all, stay informed. Yours truly, The Great Martis.

The government's bank account holds about $1 trillion in cash right now. Starting September 9th, they’re going to use that money to buy back their own debt. $4 billion dollars at a time. When the money hits the market, it can push stock prices up in the short term. But over the long haul, your money slowly lose value. Oil gets more expensive. Food prices rise. Rent goes up. The cost of living climbs across the board. A similar setup played out in 2008. People holding cash watched their savings shrink over time. The S&P 500 fell 55% during the crash. 401(k) accounts dropped anywhere from 30% to over 50%. The difference is, this time around, the numbers behind it are even bigger.

WHERE is the “BOND MARKET CRISIS”? US High Yield OAS: 260 bps. 2008: ~2,000 bps 2020: ~1,100 bps Today, credit spreads are tight - and falling. If this is a “bond crisis,” why is the credit market showing ZERO signs of panic? Again... chit-chat! Maybe the real crisis is the narrative?

Fed Governor Barr said, “If inflation appears not to be moderating sufficiently, then I think we should act decisively to raise rates.” Inflation hasn't moderated for years. What's he waiting for? Plus, a 25 basis point hike is not decisive. Neither is growing the balance sheet.

The yield on the 10-year Treasury is now 4.78%, the highest since 2007. The high from 2006 was 5.15%. Once that's taken out, the next target is the 1999 high of 6.44%, then the 1994 high of 8.03%. In 1994 the national debt was well under $5 trillion. Now it's over $40 trillion!

*This* is why China is outcompeting the West. In China, private capital is subordinate to the state. This allows major investment and credit allocation decisions to be aligned with long-term national development objectives. In many Western capitalist economies, the state is subordinate to private capital. Stock buybacks and the maximisation of shareholder value are hampering long-term strategic investment. The West needs to stop blaming China for its own lack of strategic industrial policy and its over-financialised economies.

The 10-year Treasury yield is 4.75%, the highest since Jan. 2025. When the yield rises above 4.77%, it will be the highest since 2007. However, in 2007 Treasuries were still in a bull market, with yields headed lower. Now they're in a bear market, with yields headed much higher.

More bad news for $CRWV and $NBIS Anthropic signs another $35 billion Neo-cloud data with Lambda. This will be the second $NVDA backed Neo-Cloud preparing to IPO that Anthropic made a deal with after $45 billion deal with N-Scale on Friday. Neo-Clouds are Nvidia backed SPVs used to raise capital for Nvidia to buy its own chips. A new one is born everyday.

The US will be the last to fall, Japan and Europe are in much worse condition.

Retail investors are piling into Nvidia after the company's record quarter. Retail investors purchased ~$250 million in Nvidia, $NVDA, shares on Wednesday, their 3rd-largest daily purchase since mid-May. This extended the buying streak to 15 consecutive trading sessions. Over this time, retail investors acquired +$2.5 billion in $NVDA shares. By comparison, the largest retail daily purchase was recorded in late February, at +$1.0 billion. Over the last 12 months, retail investors have purchased +$30.0 billion in Nvidia stock, the most among all Magnificent 7 companies. For retail investors, Nvidia remains at the center of the AI trade.

Yes, August and September are 'supposed' to be bad, but this year hasn't been like other midterm years. Remember, Q2 in a midterm year was also supposed to be bad (it is the worst quarter out of 4-yr prez cycle), but instead we had the greatest Q2 midterm year return ever. Will Aug/Sept continue this trend? I think so.

I personally don't think the AI bull market experiences real headwinds, until Anthropic and OpenAI both go public. Hard to go wrong buying memory, neoclouds, photonics, and robotics at strong HTF support levels... with easy invalidation if it breaks. The people underwriting these deals have every incentive to keep the AI narrative alive. Trillion dollar IPOs don't price into weak markets. Every hyperscaler capex commitment, every neocloud LOI, every memory shortage headline, every $NVDA partnership is directly tied to the engine that makes these IPOs work. Just gotta continue paying attention to Startup support and resistance levels. 🫡

Believe it. The S&P 500 has now gone 900+ days without a 20% bear market drawdown. If the AI era rhymes with the dot-com boom, we may not even be a third of the way through this streak. In the 1990s, the index went 3,102 days without a 20% drawdown.

Peter Thiel once said: “Amazon is the most ferocious company in the U.S. at this point. It's probably the company that you don’t want to be competing against.” Coincidentally, he just started a 30% position. $AMZN

The typical response I see to this image is a massive mistake. It goes something like this: “This is cherry picked data. If you simply dollar-cost averaged during this period, you still would have generated positive returns!” This is absolutely true, and something investors should understand. However, it misses the entire point of the chart. Dollar-cost averaging assumes you are still earning income and regularly adding money to your portfolio. But what if you retired in 2000? Instead of buying more shares as prices fell, you were selling shares to fund your lifestyle. When withdrawals collide with a prolonged bear market, you must sell more shares at depressed prices. Those shares are permanently removed from your portfolio and cannot participate in the eventual recovery. The market may recover, but the investor may not. That is sequence-of-returns risk. And calling the starting date “cherry-picked” misses something else: real people invested real money in 2000. Nobody knew beforehand that they were investing near the beginning of a 13-year period of negative real returns. Every historical starting point looks obvious only in hindsight. The lesson isn’t that you should avoid the stock market. It’s that average long-term returns do not arrive in a straight line and accumulation stage investors and retirees face fundamentally different risks. Cash reserves, dependable income, a flexible withdrawal rate, and growing dividends can all reduce the need to sell assets during prolonged downturns. Dollar-cost averaging can help you take advantage of a lost decade. But if that lost decade begins after your final paycheck, living off growing dividends prevents you from selling shares at depressed prices, allowing your income to rise while your principal remains invested for the eventual recovery.

People keep asking where the ROI is on $META’s massive AI capex. This is the answer. Bernstein says $META is on track to surpass $GOOG Search in ad revenue before the end of 2026. In Q2, Meta captured nearly HALF of every incremental digital ad dollar. The ROI is already showing up.

In 1957, a professional dancer turned $36,000 into $2.25 million in just 18 months. While touring the world. Getting stock quotes days late by telegram. His system was built around one simple idea: - Stocks climb in boxes - Price moves between a floor and ceiling - When it breaks the ceiling, buy - As the stock moves higher, new boxes form - Raise your stop as the trend progresses Nicolas Darvas wasn’t blindly buying every breakout either. He looked for strong, active companies with improving fundamentals - then let price action tell him which ones the market was actually rewarding. Fundamentals helped identify the opportunity. Price confirmed it. Risk management kept him in the game. Darvas couldn’t watch the market all day. Being thousands of miles away from Wall Street forced him to ignore most of the daily noise and focus on what actually mattered: Price. Volume. Earnings. Trend. Being far from the noise wasn’t his handicap. It was his edge.

What I was explaining this weekend... electric utilities in the 21st century are NOT the safety stocks they were last century. Two biggest losers today are Edison $EIX and PG&E $PCG. Before you say $VST, $TLN, $CEG, etc. have minimal to no transmission line wildfire liabilities, read my post from yesterday.

Bank of America says $NVDA is 34-50% undervalued. $350 target. The discount exists because of what makes people uneasy. Nvidia sells chips to OpenAI, has funded parts of the AI ecosystem around it, and now backstops up to $105 billion of residual value on an OpenAI data center lease. Sells the chips. Helps finance the build. Stands behind the lease if it blows up. Arya’s counter is that the backstop isn’t cash going out. It only triggers if OpenAI defaults and a relet or sale still leaves a shortfall. It’s also back-loaded to 2028. By then BofA sees Nvidia generating roughly $470 billion in free cash flow over two years, enough, in his view, to absorb the risk. The stock is around 16-18x forward earnings. Cheapest in about 7 years. His fix isn’t a product, it’s buybacks. Nvidia returns about 50% of free cash flow while peers do 75-100%. Almost every analyst already rates it a buy.

Quite a few interesting notes out on $META today - let's summarize the bullet points. META is capturing nearly half of every incremental digital ad dollar and is projected to overtake $GOOGL search in ad revenue by late 2026 or 2027, driven by AI systems. Recall Advantage+ at a $75B+ run-rate, GEM, Lattice etc that can drive both impressions & pricing. Now that teen litigation overhang is clear, it opens the runway for new product launches, similar to how GOOGL rerated late 2025. They have native consumer AI agent "Hatch" across IG/WA, premium subscription tiers, the frontier "Watermelon" model, custom MTIA silicon, and Meta Connect announcements. At roughly 16x NTM EPS, the stock trades ca 25% below its historical average and does not price in $10+ in potential incremental EPS from new AI, subscription, agentic ad tooling, and enterprise compute call options.