Methodology
You can count this one yourself.
Macro Concordance finds the years in history whose macro conditions most resemble this one, and shows how those years resolved. The match is a plain count out of seven. There are no weights inside it, no proprietary score, and nothing a member has to take on trust.
Macro commentary has a reliable failure mode. Someone decides what kind of year this is, then finds the historical parallel that supports it. The parallel is chosen after the conclusion, which makes it decoration.
Macro Concordance runs that process in the opposite order and fixes the rule in advance. Seven macro assets each set a price range in January. Whether each one later closes a month above or below that range gives it a state for the year. Together those seven states form the year's macro fingerprint. The tool then compares that fingerprint against every year on record and counts how many of the seven match.
The years it surfaces are whatever the count produces. Nobody selects them, and the tool shows how those years went for the asset you choose rather than telling you what it thinks it means.
The reference range is set by one month, chosen in advance, applied identically to every asset and every year in the record. It is not the most sophisticated possible anchor and that is close to the point. A fixed rule cannot be adjusted later to make a pattern appear, which is the failure the whole tool is designed around.
Matching is a whole number from zero to seven. No weights, no multipliers, no statistical thresholds sit anywhere in it. A member can see each asset's state, see which historical years matched, and check the arithmetic. Most macro tools ask you to trust a number. This one shows the working.
Some of the seven move against risk appetite by nature: a rising dollar or a rising real borrowing cost is not the same kind of news as rising industrial demand. Those are flipped so a bullish reading means the same thing everywhere in the set. Without that alignment the count would be adding together things pointing in opposite directions.
Nothing changes until a month ends. A break of the January range mid-month does not register until the monthly candle closes beyond it. A macro regime that flipped every few days would not be a regime, and a tool that reacted intramonth would be measuring noise and calling it a backdrop.
The closest matches rest on the fewest years
This is the tension at the centre of the tool. A loose match returns plenty of historical years and tells you relatively little. A very close match is far more interesting and may rest on a handful of years, sometimes fewer. The percentages shown against the tightest tiers are arithmetic on a small number of observations, and they should be read that way. The tool lists the matching years by name so a member can see exactly how many are behind any figure.
The usable record is shorter than you would like
For most of the seven assets, reliable history begins in the early 2010s. Anything earlier is sparse or missing. Where data is absent the tool marks it as no data and abstains from scoring that asset, rather than filling the hole with an assumption and quietly inflating the match count.
Seven price-based inputs cannot see everything
The fingerprint is built from price behaviour in seven markets. It carries no view on policy changes, structural breaks, or anything happening outside those seven. Two years can share a fingerprint and differ in every way that turned out to matter. Historical base rates here describe what happened, and they are not forecasts.
No black box to take on trust.
Members see the seven macro assets, this year's fingerprint, and every historical year that matches it, listed by name.
Discussion