Three regime indicators we will be watching into the back half of 2026

We do not forecast. We do, however, build positioning rules that depend on which regime the market is in. Three indicators will be doing most of the work for our books in the back half of 2026.

1. Cross-asset realised correlation

When equities, credit, and the dollar all start moving together, leverage gets cut everywhere simultaneously. We watch the twenty-session rolling correlation of S&P 500 returns to USD/CHF returns. Historically, when this reading rises above 0.5 it has preceded systematic de-grossing across multi-strategy books. The mechanism is straightforward: when the typical diversifiers stop diversifying, every levered investor sees their portfolio risk metric climb at the same moment, and the rational response is to reduce gross exposure.

2. The slope of the front of the rate curve

Specifically the spread between the three-month T-bill and the two-year Treasury. When this inverts and then re-steepens, carry trades historically suffer and crypto implied volatility rises. The economic intuition is that the front of the curve is the cleanest read on near-term policy expectations, and rapid changes there reprice every cost-of-funding assumption in the global macro book.

3. BTC funding rates

Perpetual-swap funding rates on BTC are the cleanest read on speculative positioning in digital assets. When they have been positive for more than thirty consecutive sessions and the front-month basis exceeds eight per cent annualised, the market is paying to be long, which means the marginal buyer is leveraged. When the trade goes the other way, it tends to go quickly, because the leverage has to clear.

None of these are predictive in isolation. The combination, two of three signalling stress, is what moves our systematic position-sizing logic from "normal" to "defensive."

Track record of the rule

We have made the move from normal to defensive five times since the rule went into production in 2019. Three of those moves were correct ahead of meaningful drawdowns. One was a clear false alarm. One is in progress. That is a hit rate we are comfortable with given the asymmetry of the trade: being briefly under-positioned is cheap, being fully positioned into a cascade is not.

We will publish the regime state in the monthly insights letter throughout the back half of 2026. If you want to see what the system is doing in close to real time, the easiest way is to subscribe.