When does the mass follow the rate?

The rate and the mass of profit don't have to turn together. Capital can keep growing, and mass with it, for years after the rate has already peaked — a falling rate just means each additional unit of capital returns a little less, not that it returns nothing. Roberts uses exactly this cushion as a slump predictor.

peak/trough = a local extreme within a ±window-year neighbourhood  ·  lag = mass-peak year − rate-peak year

In plain terms: find every year the profit rate turned down after rising (a peak), and every year the profit mass did the same. Pair each rate peak with the next mass peak after it. The theory predicts the pair exists and the mass peak comes later — the rate turning down is the early warning, the mass turning down is what the crisis is waiting on. Profit mass is converted to real terms (GDP price deflator) first: nominal profit in current dollars grows almost every year on inflation alone, which leaves nothing for a peak-finder to find.