14:50 1. Buy undervaluation, sell overvaluation — demand cash flow + a catalyst
The repeatable method
- Refuse to buy an asset just because it's going up. Require a real business with cash flow and management that knows what it's doing.
- Buy only things that are undervalued and carry a catalyst — an event that can change the market's perception and re-rate the price higher.
- Treat the market as a long-term compounding tool, not a slot machine: "buy undervaluation, sell overvaluation" beats the hype over time even though it's "boring" and "work."
- Stress-test the popular name's price against its cash flow before deciding it's a bubble (51% of S&P market cap above 10× sales is the flag).
Here: rejected SpaceX/Starlink/Grok and the coming Anthropic/OpenAI supply as "not investing"; favored cash-flowing oil producers and beaten-down miners with catalysts instead.
Watch for
- Hyped IPOs with no/negative free cash flow; the share of index market cap trading above 10× sales as a froth gauge.
31:00 2. Own the service / picks-and-shovels layer to be insulated from the commodity price
The repeatable method
- When you're bullish a commodity but can't tolerate its price volatility (or can't trade the futures safely), step one layer back to the services that the producers must keep paying for.
- Prefer overlooked, off-the-radar names (he points at "Singapore and Norway" listings) over the obvious majors everyone names.
- Verify the demand is cycle-locked, not price-sensitive: offshore projects are multi-year, so the commodity would have to sit at $40–50 for a year before activity pulls back.
- Pair the services names with a few producers that already cash-flow well even at a low price, so the book wins without needing a price spike.
Here: owns offshore oil services "regardless of the oil price"; held producers like PTAL throw off "tremendous cash flows" at $75–90.
Watch for
- Service/equipment names whose backlog is locked for years; the commodity price level that would actually force activity to pull back (the safety margin).
21:50 3. The inventory-vs-production tell — draws have a floor, production doesn't
The repeatable method
- When a commodity won't respond to an obvious supply shock, find what's absorbing it — here, strategic reserves and excess storage being drained to fake abundance.
- Track the reserve levels toward their operational minimums: SPR barrels, the Cushing settlement floor, weeks of consecutive product draws.
- Key distinction: an inventory draw has a floor (you can't go below operational minimums) while production does not — so once reserves near the floor, the buffer is gone and price must rise to ration.
- Use the physical price, not the futures screen, to see through jawboning (Singapore refinery price >$100 vs. ~$80s on the screen).
Here: SPR 450M→357M (min by early autumn), Cushing 33→24.5M, gasoline drawn 15 weeks, exports 3.9→6 Mbpd masking depletion → shortage "weeks to a couple months" out.
Watch for
- SPR / Cushing / product-stock levels vs. their operational floors; the gap between physical (refinery) prices and the futures screen.
47:42 4. Capitulation buy signal — the bullish-percent index near zero
The repeatable method
- For a sector you're long-term bullish on, watch the bullish-percent index (share of names in uptrends). A reading near zero / below ~20 marks extreme capitulation — "the end of the move."
- Confirm the businesses are still fundamentally sound at the current commodity price (cash flows intact) — you want washed-out sentiment, not broken fundamentals.
- Buy when "everybody else doesn't want it"; don't try to nail the exact bottom — aim to capture 70–80% of the move and add as it falls.
- Use the easy major as the low-risk expression, or hunt the beaten-down juniors for more upside.
Here: gold-miner bullish-percent index "reached zero" (unseen since 2016) → added a junior, flagged AEM down 40% as the "just buy that" major; Tavi Costa's price-vs-cash-flow chart confirms fundamentals.
Watch for
- Sector bullish-percent index at/near zero with profitable underlying companies; sentiment washouts where price — not fundamentals — collapsed.
44:15 5. Central-bank gold buying as value-buyer confirmation
The repeatable method
- Treat central banks as disciplined value buyers of gold: they accumulate on dips and pull back at highs.
- When their buying resumes after a pullback (and forced sellers who were defending currencies reverse), read it as confirmation the value is being recognized.
- Don't fight the short-term headwind: when synchronized rate-cutting flips to hikes (tightening liquidity, stronger dollar, rising real rates), expect gold pressure — but anchor on the trend in real rates, not the level.
Here: China's central-bank purchases picking up again after ~2 flat years; Turkey/others stopped selling — but he flags rising global rates as the near-term offset, staying long-term bullish.
Watch for
- Central-bank purchase data turning back up on a price dip; the direction (trend) of real rates as the swing factor.
54:16 6. Second/third-order effect hunting — the "page-16 story" screen
The repeatable method
- Take a front-page event and trace its non-obvious downstream effects two and three steps out (Don Coxe's "look for the page-16 or page-20 story moving to page one").
- Find a company you already own/like that is accidentally positioned to benefit from that downstream effect — "put yourself in a situation to get lucky."
- Quantify the windfall to confirm it's material, not a rounding error.
- Remember it cuts both ways — the same chain can be a disaster for the firms on the wrong side.
Here: Hormuz closure → ~20–30% of world sulfur (a Gulf-refining byproduct) removed → sulfur price doubled → IVN's smelter byproduct now ~$1M/day, on top of copper >$6/lb.
Watch for
- Byproduct/feedstock chains disrupted by a geopolitical event; an existing holding that captures the windfall without new risk.
50:45 7. Scarcity → own the land/royalty layer where the buildout is unopposed
The repeatable method
- Spot a buildout colliding with a hard physical limit (here: data centers vs. power + water + neighbor backlash; industrial power up ~50% in five years).
- Locate where the buildout can proceed unopposed — stranded gas/water on near-empty land — and own the entity that collects royalties/fees there rather than the capital-intensive operator.
- Look for the same model "a generation behind" the marquee name (early, under-followed, run by people who "get it").
- Don't invent a ticker for a name the speaker won't disclose — note it as a peer reference and wait for the name.
Here: TPL and LB (LandBridge) for West-Texas land/water/power (Loving County ~100 people); plus an unnamed Arizona land company "like a TPL situation a generation behind" outside Phoenix (no ticker disclosed).
Watch for
- Land/royalty owners in resource-rich, low-population regions; local moratoriums/power-price spikes as the demand signal.
17:41 8. Issuance + Berkshire-cash near tops as a market-top tell
The repeatable method
- Track net equity issuance of non-financial corporates: a surge of new supply (IPOs, secondaries) competes for the same finite liquidity and caps other assets.
- Overlay it with Berkshire's cash level — both tend to peak together "around tops" (Buffett hoards cash when he can't find value; companies issue stock when prices are rich).
- When supply floods in and other names stop rising because money is being pulled to chase it, treat it as a late-cycle warning, not a green light.
Here: net equity issuance near recent highs before SpaceX, BRK.B cash "way up" — the classic top-cluster; other names flat as money rotates into the IPO.
Watch for
- Net equity issuance vs. S&P market cap spiking; Berkshire's cash level relative to past tops; breadth stalling as supply hits.