A structural break is a lasting shift in a stock's price behavior — a re-rating, an acquisition, a fundamental change in the business — that takes future prices outside the range the stock's own trailing history would have suggested.
A structural break usually traces to a concrete event: an acquisition that changes the business overnight, a demand surge that re-rates what the market is willing to pay, a regulatory shift, or a management change that alters the company's trajectory. What these have in common is that they break the assumption every history-based model relies on — that the recent past is a reasonable guide to the near future. A mechanical stock split, by contrast, is adjusted out of the price series and is not itself a structural break.
A platform-wide coverage average can look healthy while a handful of individual names are badly miscalibrated, because good coverage on most tickers offsets poor coverage on a few names that went through a structural break. That is why Quantustik publishes per-ticker calibration figures rather than the aggregate alone — checking the specific ticker you actually trade matters more than trusting the platform-wide average.
A wide confidence band, or a per-ticker backtest accuracy that lags the platform average, is a signal that history-based forecasting has less to say about that specific name — treat it as a reason for caution and smaller position sizing, not as noise to look past. See the named AI Lab post-mortem for a worked example of two real tickers this affected and how the bands were honestly widened in response.
Read the full post-mortem: Where our model fails: structural breaks.
A lasting shift in how a stock behaves — driven by something like an acquisition, a re-rating, or a fundamental change in the business — that takes prices outside the range the stock's own trailing history would have implied.
Not one built on historical price-path statistics. By construction, such a model can only widen or narrow a band around what the past distribution suggests is plausible — a structural break is a departure from that distribution, not something the model can see coming.
No — a mechanical stock split is adjusted out of the price series before any modeling happens. A structural break is a genuine shift in the business or how the market values it.
By publishing per-ticker calibration figures rather than only an aggregate average, and by widening confidence bands where the historical record shows real miscalibration — an honesty fix, not a claim that the model gained foresight it never had.
Browse all S&P 500 tickers to see this metric applied to individual companies.
Educational research only — not investment advice.