Intermarket analysis

The logic is intuitive enough that it became a fixture on trading desks. Interest-rate markets are famously quick to price new information, and a spike in credit stress tends to hit equities before it fully shows up in economic data. Chain enough of these relationships together and you get a story where one market's move today writes tomorrow's headline in another. The appeal is obvious. If bonds really lead stocks, you don't need a crystal ball. You need a calendar.

That story bundles two separate claims, and they get treated as one far too often. The first: related markets move together. True, and not surprising. Equities and real yields share common macro drivers, so they correlate for good reason. The second claim is bigger: that one market's move today reliably predicts another market's direction tomorrow. Co-movement is a fact you can measure in a correlation coefficient. A tradeable lead is a forecast, and a forecast has to survive a test, not just a chart where two lines happen to rhyme.

So we tested it properly. Synchronous bars only, no borrowing a few hours across time zones to manufacture a lag that isn't really there. Then a volatility-null benchmark, so a signal doesn't get credit just because every asset moves together on a high-volatility day. That alone can look like prediction and isn't. And an honest count of every lag window and asset-pair combination we tried in both directions, the same discipline behind multiple testing, because run enough combinations and one will look predictive by pure chance.

Of more than 400 directional configurations, essentially one held up. Not one asset class. One configuration. Everything else failed the way most of what we reject fails: no real edge, full stop, before a cent of cost even entered the math. The apparent lead was two correlated markets moving on the same macro trigger, arriving within hours of each other. Coincidence, not causation.

Hold onto the distinction. Contemporaneous co-movement between markets is real, and mostly explainable by shared macro drivers. A genuine tradeable lead, one market's move forecasting another's direction before it happens, is almost never there once you test for it honestly. The textbook version of intermarket analysis survives on charts. It does not survive on out-of-sample bars.