Actionable insights — Has The Grand Game Just Changed?

8:48 1. Mark every crisis on the long-term yield chart — the breaking point falls as leverage rises

1. Plot the long-term Treasury yield over several decades and mark each financial crisis at the rate level where it struck.
2. Connect the marks. If each crisis hit at a lower yield than the last, the system's tolerance for rates is falling — because debt was added as rates declined, so every basis point now costs more.
3. Use the trend line, not the last peak, as your estimate of the current pain threshold. Today's yield being "below the old highs" is no comfort if it is above the falling line.
4. When yields sit above that line with no crisis, don't conclude the chart is broken — look for what is suppressing the symptom (here, the expectation of a rescue) and for stress already visible off-screen.

• Bankruptcy counts and ratings downgrades rising while the index is calm; forced sales of private assets below their marks; the 30-year holding above the falling crisis line.

12:50 2. Judge the Fed put by repeated non-response to drawdowns, not by speeches

1. Treat investor belief in a rescue as a conditioned reflex (Pavlov): four decades of rewards don't get unlearned from a hawkish speech.
2. Define the test event: an equity drop of 5–10% within days or weeks.
3. Score the central bank's reaction each time — any dovish commentary, emergency liquidity, or talk of cuts is a "put still alive" data point; silence is a "put retiring" data point.
4. Require repetition. One non-response doesn't change behaviour; only a string of them does. Until then, assume markets will keep pricing the rescue ("extend and pretend" to the next expected cut).
5. Check it against history: a new chair who talks hawkish and then capitulates in the first real sell-off (Greenspan 1987) doesn't end the put — he founds it.

• The Fed's first comments after a fast 5–10% equity drop; whether rate-cut pricing snaps back after each sell-off; the long end rising while equities make highs (the '87 setup).

21:40 3. Look through spreads to the absolute borrowing cost at the roll

1. Don't stop at credit spreads. A narrowing spread over a rising Treasury yield can still mean a much higher coupon.
2. Compare today's all-in yield for a borrower class with the rate at which its existing debt was issued (for junk, the zero-rate trough).
3. Multiply that gap by the maturity wall — the amount that must refinance — to size the interest-expense shock.
4. Add the competition for capital: the same lenders are being asked to fund the government's deficit, munis, consumers and any capex boom. Weak issuers get priced out first; the signal is a high-profile deal that "really does not go well."

• High-yield index yield-to-worst (not just OAS); the next 12–24 months of maturities; a failed or badly re-priced refinancing by a recognizable name.

19:47 4. Find the funding market that breaks first — the 2007 ABCP template

1. For any bubble, separate the asset (housing then, AI capex now) from its financing. The bust reaches creditors with a lag — and shows up first in the most short-term, confidence-dependent funding line.
2. Identify today's equivalent: the borrowing costs and CDS on the issuers funding the boom.
3. Track the divergence: credit demanding a rising risk premium while equities stay euphoric is the flash-point setup, not a contradiction.

• CDS on the largest AI borrowers; new-issue concessions on hyperscaler bonds; any short-term or off-balance-sheet funding vehicle for data centers that struggles to roll.

32:43 5. The checking-account test for "firepower" claims

1. When an official cites a cash balance as ammunition, ask: is this a surplus or a float? Is the money already committed to outgoing payments?
2. Net it against the flow it must fund. A balance that exists only because of the calendar (receipts in before spending goes out) is a checking account, not a war chest.
3. Any use of it must be replenished — by more borrowing — so the net effect on supply is zero or negative. Discount the claim accordingly.
4. Then ask what the claim is for: if it can't do what it says, it is probably meant to change positioning or sentiment.

• TGA drawdowns followed by bill issuance to rebuild it; buyback sizes vs the deficit run-rate; any formal move on the gold certificate price.

33:32 6. Carry the positioning caveat on your own call — and ask who policy is aimed at

1. Hold a directional view (here, bearish on long bonds), but check speculative positioning in the same instrument. A record crowded short is the one thing that can move price hard against a correct fundamental view.
2. List the catalysts that would force covering (a whiff of fiscal discipline, less long-dated issuance, a loud official threat).
3. Read official announcements through that lens: a claim that is fundamentally hollow but aimed at the crowded side may be a deliberate squeeze.
4. Tag the move as unsustainable if nothing fundamental changed — fade the rally rather than flip the view — and time it against the political calendar the policymakers care about.

• CFTC positioning in long-bond and ultra-bond futures; a sharp long-end rally on an announcement with no fiscal change; mortgage rates into October.

40:23 7. Against official intervention, count the ammunition

1. Establish the direction of the fundamentals. A central bank can push its currency down indefinitely (printing) but can't push it up indefinitely.
2. If it is defending against the fundamentals, estimate its finite ammunition — reserves available, what it has spent per round.
3. Size your position to survive the remaining rounds ("suck it up for another however many iterations"); the defender's spending is public and countable, yours isn't.
4. Map the second-order risk of a successful defence: propping up a funding currency threatens every trade financed in it.

• Japan MoF intervention totals and FX reserve changes; USD/JPY after each intervention; JGB yields vs US yields as the carry spread.

44:11 8. Split aggregate balance-sheet strength into the top 10 and the rest

1. When someone argues "record earnings mean corporates can easily absorb higher rates," disaggregate: the top ~10 index members versus the other 490, then versus all US companies.
2. Test even the leaders on free cash flow, not earnings — the strongest aggregates can hide leaders that have flipped FCF-negative.
3. Apply rising interest expense to the weakest cohort first; that's where bankruptcies come from, well before the index notices.

• Interest coverage for the equal-weight or small-cap index vs the cap-weighted one; FCF turning negative at mega-caps; margin debt peaks.

48:10 9. Read a long-dated supply announcement for its near-term effect on expectations

1. For any supply deal that can't deliver physical volume for years, separate its long-term value from its near-term job: shaping price expectations.
2. Ask why it was announced now — against which price shock and which political deadline.
3. Then check the near-term physical balance independently (inventory rebuilding, structural demand). If that balance is tight, the expectations cushion limits the upside but doesn't reverse it.
4. Widen the lens: whose supply does the deal take away? The geopolitical effect can be immediate even when the barrels aren't.

• Strategic-reserve refill announcements; oil holding up after Iran de-escalation headlines; China's crude import mix.