This article is educational — a framework for understanding and evaluating regime-adaptive strategies. It is not personalised investment advice, not a recommendation of any product or security, and not a claim that any strategy is profitable for any individual.
“Regime-adaptive” has become a label every quant stock-signal pitch reaches for — and most of what wears it is a static screen with a market filter bolted on. The distinction matters, because a strategy that genuinely changes its behaviour with the market behaves very differently from one that simply trades less in a downturn. Here is what the term should mean, why it is the right thing to want, and the questions that separate a real one from a relabelled static system.
What “regime-adaptive” actually means
A market regime is a persistent state of the market — broad uptrend, narrow rally, choppy range, risk-off — with its own statistical character. A regime-adaptive strategy classifies the current regime and changes what it does in response: not just how much it trades, but which logic it runs. The academic roots go back to Hamilton’s (1989) Markov regime-switching work and later regime-shift asset-allocation research (Ang & Bekaert, 2002); the mechanics, in trading terms, are covered in regime-switching strategies.
Why a static strategy fails across regimes
Every edge has a home regime. A momentum-breakout screen prints money in a broad uptrend and bleeds in a choppy, mean-reverting tape; a mean-reversion system does the opposite. A single static strategy keeps firing the same logic into every market, so it gives back in its hostile regimes much of what it earns in its favourable ones. The fix is not a better single screen — it is matching the logic to the regime, which is the entire premise behind running four engines for four regimes instead of one.
How to evaluate one — five questions
Most of what is marketed as regime-adaptive will fail at least one of these. They are the questions worth asking of any system, ours included:
- Does it change behaviour, or just exposure? A real one switches which strategy runs by regime. A weak one only dials position size up and down while running the same screen.
- Is the regime call point-in-time? The classifier must use only data available before the decision — no look-ahead. A regime model fit with hindsight will dazzle in backtest and disappoint live.
- Will it sit out? A genuinely adaptive system has regimes in which it does nothing and holds cash. One that is always fully invested is not adapting; it is just rotating.
- Is the backtest survivorship-free and reproducible? Ask whether dead and delisted names are in the test, and whether the result reproduces to the dollar. The traps are catalogued in why backtests lie.
- Is it transparent about the losers? A credible operator shows the full trade distribution, not a curated highlight reel — the discipline behind a high-failure-rate, positive-expectancy setup.
Regime-adaptive is not market timing
A common objection: isn’t this just market timing in a lab coat? No. Market timing tries to predict where the market goes nextand bets on the forecast. Regime adaptation makes no forecast — it reacts to the regime that is already measurable today and runs the logic suited to it, switching only when the measured state changes. One is a prediction; the other is a response. The difference is the difference between guessing and classifying.
Where Shishin sits
By category, Shishin is a regime-adaptive system: a breadth-driven classifier gates four specialist engines, each running only in the regime it suits, with cash as a legitimate position. We publish the full five-year backtest and a live paper-traded track so the five questions above can be checked against us, not just asked of others. How the regime read becomes a concrete, ranked daily output is the subject of daily ranked signals by market regime.
Sources & further reading
- Hamilton, J. D. (1989). “A New Approach to the Economic Analysis of Nonstationary Time Series and the Business Cycle.” Econometrica, 57(2), 357–384. — the foundational Markov regime-switching model.
- Ang, A. & Bekaert, G. (2002). “International Asset Allocation with Regime Shifts.” Review of Financial Studies, 15(4), 1137–1187. — regimes applied to allocation.
- Jegadeesh, N. & Titman, S. (1993). “Returns to Buying Winners and Selling Losers.” Journal of Finance, 48(1), 65–91. — why a momentum edge is regime-dependent.