Actionable insights — The S&P 500 / Gold Ratio

0:48 1. Denominate the asset in gold before comparing across decades — the dollar is not a fixed ruler

1. Whenever a comparison spans more than a business cycle, ask what the numeraire did over that window. A nominal price series across 50–100 years measures the currency as much as the asset.
2. Re-express the series as a ratio to a hard, non-issuable denominator — gold is the conventional choice — so that inflation and currency debasement cancel out of both legs.
3. Read the resulting line as a relative-performance series, never as a price. It says which of two assets is winning, and nothing about whether either is going up in your own currency.
4. Apply the same trick anywhere the unit of account is doing the work: index / gold, house prices / median income, market cap / GDP, wages / CPI.

• Long-horizon charts drawn in nominal dollars — treat them as unfalsifiable until re-denominated. Conversely, when a ratio chart is offered, check what the denominator itself did: a falling ratio can be a gold bull market rather than an equity bear market, and the two call for different actions.

3:31 2. Identify the regime before picking the asset — the explicit lesson of the century

1. Start from the premise that no asset class wins forever and that leadership persists for decades, not quarters — long enough that a single regime feels like a permanent law to anyone who invested through it.
2. Before choosing a security, answer the prior question: which regime is currently in force — paper assets leading, or hard assets leading?
3. Set exposure to that answer rather than going all in on the side that has just worked. The historical winners "weren't the ones who went all in on one side."
4. Re-ask the question on a slow cadence. A regime call that changes monthly is not a regime call; a multi-decade cycle can only be reassessed against multi-year evidence.

• The three continuation conditions the video itself names: real interest rates, central-bank gold buying, and whether AI-driven earnings growth keeps justifying equity valuations. Each is observable and each would, if it turned, argue the regime is turning with it.

3:10 3. Treat a ratio as a rotation gauge with no timing content — and hold its author to that

1. For any indicator offered as predictive, separate the two claims inside it: the coincidence claim (this level has historically accompanied X) and the prediction claim (this level causes or foretells X). Only the first is usually supported.
2. Count the observations. A century of a two-asset ratio yields roughly four or five full regime turns — far too few to establish a hit rate, however clean the chart looks in hindsight.
3. Demand a falsifier: what level, over what horizon, would prove the indicator wrong? A gauge with no falsifier is a description of the past, and should size nothing.
4. Use it for allocation drift — nudging exposure between paper and hard assets — and never for entries, exits or leverage.

• The tell of a promotional framing: a headline claim of prediction, a body that only ever demonstrates coincidence, and a disclaimer late enough that most viewers never reach it. When they conflict, the page — and the position — should follow the disclaimer.

2:21 4. Flag the rare configuration — record equity highs and a hard-asset bull market — as something to investigate, not to trade

1. Watch for two assets that normally move against each other making highs at the same time. Risk assets at records alongside a monetary hedge at records is the specific case: it is not the usual behaviour of either.
2. Ask what single set of conditions could produce both. Here: liquidity and a narrow growth story bidding equities, while central-bank diversification out of the dollar, persistent inflation and geopolitical risk bid gold — i.e. the buyers are not the same people and are not expressing the same view.
3. Write down the two competing interpretations before choosing: (a) equity strength is masking risk not yet priced; (b) it is the early innings of a long rotation into hard assets, as after the 1970s and after 2000.
4. Because nothing in the configuration distinguishes (a) from (b) in real time, the correct output is a research agenda and a modest exposure tilt, not a trade. Decide in advance which later evidence would resolve it.
5. Check the breadth beneath the index high. An index carried by a handful of names is a different fact from a broad advance, and it is the version consistent with interpretation (a).

• Equal-weight versus cap-weight index divergence (the breadth check); central-bank gold purchase reporting; real yields; and whether gold's advance persists through an equity drawdown, which is the evidence that separates a genuine regime rotation from two unrelated bull markets.