The advance decline line was green on a red day. The advance decline line is the running count of how many stocks are rising against how many are falling across the whole market. Green on a red day means more names went up than down while the index went the other way. Most people read that as damage being contained. It means the opposite, and this is the setup that catches them. What positive breadth is telling you Money is rotating instead of leaving. Somebody sold semiconductors and bought staples, somebody dumped a megacap and bought healthcare, and the index barely registers any of it because the buying offsets the selling. That sounds fine until you ask what those buyers ended up holding. Nobody wakes up excited to buy Target, and nobody has a thesis on toilet paper. They bought it because the money had to go somewhere and it

There is a market condition almost nobody watches for. You already know how a normal selloff behaves. Price drops, fear shows up, volatility spikes, and somewhere inside that spike the selling exhausts itself and the buyers come back. The dangerous version is when price drops and volatility does nothing at all. It has happened twice in the last fifteen years, and both times it got expensive before anybody noticed what they were looking at. What you are looking for You need two things happening at once. The market grinds lower over several sessions without crashing. Down a half percent, a bit more the next day, nothing that makes the news or gets anybody’s attention. And volatility sits flat or falls right alongside it. I call that vol down, market down, and it is one of the worst situations you can be in. Why it does so much damage You cannot

The speech gets released before the man says a word. The exact words sitting in the teleprompter get shot out fractions of a second before he opens his mouth. Bloomberg charges about a million dollars a year for algorithmic access to that feed. So when Kevin Warsh stepped up at Jackson Hole yesterday, the market had his speech before anybody in the room heard the first sentence. Which is why I did not watch it. Let me show you what happens in those thousandths of a second. There is a small number of machines that read the text and interpret it. They fire trades based on what they think it means. Then there is a much larger group of machines that do not read anything at all. They watch the first group and react to what those machines just did. The second group is far more plentiful, and it is

Three bearish trades in a row on the Schwab Network this morning. Marley Kayden noticed and said it’s been a while since I didn’t bring a bull sandwich, which is fair. But look at what’s in front of us.  End of August, VIX is down, the S&P feels barely awake, and we are sitting right on the cusp of a seasonally volatile stretch. It’s rare to get a quiet September. So strap in, and maybe go buy some volatility while it’s still this cheap. There’s one thread running through all three of them. Every one of these names has run too far, too fast, and I’m fading each of them… Click below, and you’ll get the symbols, thesis, and trade structure.  → Watch the replay To your success, Don Kaufman

You already know the rule everybody uses.  Down 10% is a correction, down 20% is a bear market, rally 20% off the low and congratulations, you’re in a new bull market. But here’s the thing few think about… A 40% loss needs a 67% gain to break even. And it’s why I want you to stop with the bull-bear BS. True story… Micron sold off 40% from its recent high. Bear market, by anybody’s definition. Then that little monkey rallied 37%. New bull market, according to the rule. Except then it sold back off about 12%, which drops it into correction territory again. So that’s a bear market, a bull market and a correction inside one stretch of chart.  What did you learn about the stock?  Anything at all? South Korea is worse. That crap got hammered. The whole market went into a full-fledged horrendous bear market, down 35%, then

There’s a reason markets get stuck in a range and then get violently pulled back into it, and it has nothing to do with technical analysis. I call it the volatility box. People hear the word range and immediately start drawing Fibonacci retracements all over it. I’m not going to sit here and prognosticate on a 618 or some crap like that. Here’s what happens. A market channels back and forth for a considerable period, weeks or months, and every single day inside that channel you’re accumulating more and more open interest at those strikes. Think of it like a little tiny snowball that starts to grow, and it grows, until eventually there’s one mother snowball sitting inside that pocket. Now, who’s on the other side of all that? Market makers, and here’s what most people miss about them. On every trade you do, whether you buy stock or sell

Broadcom is shopping for a hundred billion dollars. Not through a bond offering. Off balance sheet, through a special purpose vehicle, in a structure that would be the largest deal of its kind ever funded. Roughly $30 billion of junior debt, another $60 to $70 billion of senior secured paper that Broadcom guarantees a piece of. Blackstone and Apollo are in talks to participate. Same two firms that backstopped Nvidia’s $500 billion compute deal. Is it legal? Absolutely. But nobody does off balance sheet to the tune of $60 billion unless money is so stupid that it just needs the yield. And Broadcom’s credit default swaps exploded the moment it hit the tape. Let me break that down, because credit default swaps confuse the hell out of everybody. Forget the word swap. The two words that matter are credit and default, and have you ever heard those used together in

I was on Schwab Network this morning for the Big Three. Three trades, and two of them are bets against stocks that just went up. The first one is a sector that has been the biggest beneficiary of the rotation out of big tech. It popped again this morning on drug trial news and I’m using that pop to get short. The catch is I can’t time the pullback I’m looking for, and I said so on air. So I gave myself all the way out to December. A dollar of risk on a spread that can be worth five. Around a 20% probability, which means I lose on this one far more often than I win, and the math still works because of what it pays when it hits. The second one is a chip name that got cut in half from 452 and then bounced after earnings. I

Somebody in the room told me they got stopped out of a spread. I want to talk about that for a minute, because it drives me absolutely nuts. Did anybody ever ask you to put a stop order on a defined risk trade? A spread has your risk written into it already. You paid what you paid, and that’s the most the market can take from you. So you don’t have to further define your risk on an already risk-defined trade, do you?  I’m just asking… People that put stop orders on spreads, there is something wrong with you. I don’t care whether you bought it or you sold it. And if you don’t like the risk, I’ve got better news for you. Don’t do the damn trade. Does that sound reasonable? Does it? Here’s what happens when you stop out of a spread. You take the loss at the

Google turned $900 million into $94 billion. That’s the SpaceX stake Alphabet disclosed last week, 551 million shares, about 4.2% of the company, off a check they wrote in 2015. More than a hundredfold in ten years. Nvidia showed up with $21 billion, its second largest disclosed equity position after Intel. That one came in sideways, out of a $10 billion investment in xAI that turned into SpaceX shares when Musk folded xAI into the rocket company. AMD is on the cap table too. So is Norway’s sovereign wealth fund, the Saudi PIF, Fidelity, and the University of California. And Harvard. Harvard Management’s single largest stock holding is SpaceX, at $2.2 billion, which works out to 52% of their entire disclosed US equity portfolio. TSMC is second at about a sixth the size. Who would have thought Google is buying SpaceX? Isn’t that kind of like one of your competitors?