4:52 1. Read deliveries, not price — the standing-for-delivery gauge
The repeatable method
- Pull COMEX monthly delivery notices (contracts stopped) for gold and silver and convert to ounces and dollars (1 gold contract = 100 oz; 1 silver = 5,000 oz).
- Anchor to the historical norm: physical delivery has averaged <1% of contracts issued across a 36-year career — delivery is an anomaly, so a run of large deliveries is the tell.
- Compare the year-to-date total to whole prior years (2023 ≈ 23,000 silver contracts for the full year). When five months already exceeds a full year, "price is misdirecting."
- Separate the signal from the mechanism: deliveries don't automatically drain vaults — what matters is the desire to take ownership (numbered bars over paper warrants).
Here: ~34,000 silver contracts (~170M oz) and ~99,000 gold (~9.9M oz, ~$45B) delivered Jan–May 2026, plus ~$13B of gold in June — "the highest ever," read as sovereigns accumulating into the price decline (
6:38).
Watch for
- Monthly delivery totals breaking prior-year records; the big delivery months (Feb/Apr/Jun/Aug/Dec) where contracts come due.
8:10 2. Cross-check official numbers against the refineries
The repeatable method
- Never take reported central-bank purchase figures at face value — buyers under-report to avoid moving the market.
- Triangulate against physical throughput: the World Gold Council's refinery-flow estimate is the harder number.
- When the refinery number dwarfs the reported number, treat the gap as quiet accumulation — and note the direction: buying as the price falls is positioning, not panic.
Here: central banks reported 15 tons last quarter; the WGC refinery read implied ~244 tons — "15 times the official number" — bought into the crash (~1,000 tons/yr for four years).
Watch for
- Quarterly WGC refinery estimates vs reported official-sector purchases; any widening gap during price weakness.
9:20 3. Use the reserve-manager surveys as a forward signal
The repeatable method
- Read the annual central-bank reserve-manager surveys for intentions, not just flows: % planning to add gold, % expecting official gold holdings to rise, % expecting the dollar's reserve share to fall.
- When all three point the same way, treat it as positioning ("this isn't confusion") and weight it above price action and media narrative.
Here: ~45% of central banks expect to add reserves this year, ~90% expect official gold holdings to rise, ~74% expect the dollar's reserve share to fall — "the media is watching the wrong markets."
Watch for
- The yearly survey prints; rising "expect to add gold" / "expect dollar share to fall" readings.
29:52 4. Track repatriation & trust as the break trigger
The repeatable method
- Monitor sovereign gold-repatriation requests (metal demanded back from the New York Fed / Bank of England) as the clearest "ownership still means ownership" signal.
- Set it against the trust backdrop that didn't exist in 2011 — reserve confiscation/sanctions, SWIFT removal — which is what makes this cycle different from 1980/2012.
- Use the historical pattern: what closed the gold window (de Gaulle demanding delivery) and what broke Madoff was the same thing — delivery. The break comes when too many ask for the metal at once.
Here: France, Germany, the Netherlands, Czech, Poland, Hungary, Turkey, Austria and India repatriating — "it tells you where this is ultimately going" (
15:57).
Watch for
- New repatriation announcements; reserve-sanction precedents; new non-Western settlement rails (CIPS, mBridge, Shanghai/Dubai/Singapore/Hong Kong physical-metal contracts).
31:42 5. Discount the "available" inventory to its true free float
The repeatable method
- Don't trust headline exchange inventory. Strip the LBMA free float down to what's truly unencumbered (back out ETF-owned metal — ~140M oz headline may be ~40M oz real, per David Jensen).
- Size the paper claim against the metal: COMEX runs ~300–400 paper contracts per physical ounce; assume much "registered" silver is bank-earmarked (per Ed Steer, JP Morgan).
- Conclusion to act on: if everyone stood for delivery only a fraction would receive metal ("just like a bank") — which is why standing-for-delivery is rationed ("death by a thousand paper cuts"), not maximized.
Here: the takeaway is why deliveries come in steady billions every month for 18 months rather than one system-breaking demand — a forced majeure would "blow up" the exchange.
Watch for
- Registered-vs-eligible COMEX stocks; LBMA float net of ETF holdings; any month a single buyer tries to stand for an outsized amount.
36:01 6. Build a "tells" mosaic for a monetary-regime shift
The repeatable method
- Collect small institutional signals that individually look like curiosities but together suggest gold is being re-integrated into the system.
- Personnel: who gets hired and what they've written (a policy advisor who authored a gold-bond / parallel-gold-standard chapter).
- Official products & pricing: government mints pricing commemorative coins at extreme premiums to spot can be a soft revaluation signal — verify it's real (check the primary source) before acting.
- Policy proposals: gold-convertible Treasury bonds ("Sheltons"), mark-to-market of official gold — track whether they move from think-tank to mainstream.
Here: Paul Winfrey (Warsh's top advisor) wrote Project-2025 Chapter 24 on a parallel gold standard + gold bonds; the US Mint posted half-oz gold Liberty Bells at ~$10,000 (vs ~$2,000 of gold) on usmint.gov (
38:03).
Watch for
- Gold-friendly Fed/Treasury appointments; official gold-bond or mark-to-market proposals; mint pricing of bullion coins far above spot.
44:28 7. Define the capitulation, then dollar-cost-average through it
The repeatable method
- Define "the bottom" behaviorally, not as a price target: the level where the speculation is shaken out and the leveraged futures/options "expire worthless" (here, sub-$60 silver / sub-$4,000 gold).
- Don't try to pick it — accumulate on a fixed cadence (DCA every ~15 days); falling prices mean a better cost on the next tranche.
- Own metal as wealth, not a trade — sizing should let you ignore the daily volatility that "tugs on your emotions."
Here: Taggart adds every 15 days regardless of price; Schectman calls the bottom "pretty close" and frames metal as long-term wealth, not a get-rich vehicle (
28:27).
Watch for
- Capitulation tells (sentiment, options open interest expiring worthless); the dealer discount on physical (junk silver swinging from ~$2-under-spot toward fair value as the refiner backlog clears).
47:51 8. Harvest losses with the precious-metals tax swap
The repeatable method
- Know the rule difference: the 30-day wash-sale rule applies to securities, not commodities — so metal (and crypto) can be sold at a loss and immediately rebought.
- Execute: sell the physical at a loss and rebuy the identical metal (~2% round trip if vaulted; the dealer moves the bars to make it a real transaction).
- Book the realized loss against capital gains (or carry it forward against income) and reset your cost basis higher-into-lower — keeping the same position.
- Confirm with your own accountant; treat the loophole as potentially temporary (likely to close as crypto adoption grows).
Here: a January buyer down ~50% in silver / ~20%+ in gold can sell-and-rebuy, harvest the loss, and keep the metal — "paper-trail free" and basis reset.
Watch for
- Year-end loss-harvesting windows while metals are down; any IRS rule change extending wash-sale treatment to commodities/crypto.