5. Market Regimes

“The market regime is the operating environment for your strategies. Use the wrong tool in the wrong regime, and even great execution loses money.”

A regime is a persistent state of the market with characteristic statistical properties. Regimes change. Strategies that excel in one usually fail in another.

The four basic regimes

Regime	Characteristic	Best strategies	Worst strategies
Bullish trend	Higher highs, low vol, narrow drawdowns	Momentum, breakout, trend-following	Mean reversion, short selling
Bearish trend	Lower lows, rising vol	Short momentum, vol buying	Buying dips, naked shorts of vol
Quiet range	Low vol, sideways	Mean reversion, premium selling, condors	Breakout, momentum
Crisis / High vol	Vol spike, correlations → 1, gaps	Cash, tail hedges, gamma scalping	Most everything else

Detecting the regime

Trend filters

• Price vs 200-day MA — above = bullish regime; below = bearish.
• Slope of 200 SMA — rising/falling reinforces.
• % of stocks above 200 SMA (breadth) — > 70% = strong bull, < 30% = bear.

Volatility filters

• VIX level — low (<15) → quiet, normal/high → trending or stress.
• VIX percentile (1-year) — relative measure, more useful than absolute.
• Realized vol clusters — GARCH or simple rolling std.

Trend strength

• ADX > 25 → trending regime (any direction).
• ADX < 20 → range / consolidation regime.

Breadth & participation

• Advance-decline ratio.
• % of stocks at 52-week high vs low.
• Sector rotation patterns.

Combined regime classifier (example)

if Nifty > 200 SMA and slope_pos and VIX_pct < 60:
    regime = "BULL_TREND"
elif Nifty < 200 SMA and VIX_pct > 50:
    regime = "BEAR_TREND"
elif ADX < 20 and VIX_pct < 40:
    regime = "QUIET_RANGE"
elif VIX_pct > 80:
    regime = "CRISIS"
else:
    regime = "TRANSITION"

Why this matters — the strategy/regime fit

Strategy	Bull trend	Bear trend	Quiet range	Crisis
50-day momentum	⭐⭐⭐	⭐⭐ (short)	❌	❌
Breakout	⭐⭐⭐	⭐⭐	❌	⚠️ (real or fakeout?)
Mean reversion (RSI fade)	⚠️	❌	⭐⭐⭐	❌
Short premium (condor)	⭐⭐	⚠️	⭐⭐⭐	💀 (existential risk)
Long vol (straddles)	⚠️	⚠️	❌	⭐⭐⭐
Pairs / stat arb	⭐⭐	⚠️	⭐⭐⭐	❌ (correlations break)
Long index + cash	⭐⭐⭐	❌	⭐	❌

Most retail blow-ups happen because a strategy that worked for 18 months in a quiet bull suddenly faces a regime change — and the trader keeps doing the same thing. Recognize the change, switch strategies (or step back).

Regime change signals

Things that often precede transitions:

• Volatility expansion out of a long compression (BB squeeze on Nifty itself).
• Breadth divergence — index keeps rising but fewer stocks participate.
• Defensive sector outperformance mid-bull (FMCG, Pharma leading) → late-cycle.
• Yield curve inversion (10y − 2y < 0) — historically a recession signal 6–12 months out.
• Skew steepening — institutions hedging.
• Credit spreads widening — risk-off arriving in bond market first.

Adaptive systems

Two approaches:

1. Switching system

Maintain N strategies. Each day, deploy only the one(s) suited to the current regime.

if regime == "BULL_TREND":
    deploy: momentum_breakout, pullback_long
elif regime == "QUIET_RANGE":
    deploy: rsi_mean_reversion, iron_condor
elif regime == "CRISIS":
    deploy: cash_only, tail_hedges

Pros: clean attribution, simple risk controls. Cons: regime detection is lagging — you switch after the regime has already changed.

2. Always-on with regime-modulated sizing

All strategies always run, but position size is scaled based on regime fit:

• Bull trend: momentum at 1.0×, mean-reversion at 0.3×.
• Range: opposite weights.
• Crisis: everything at 0.2×.

Smoother equity curve, but more complex to manage.

Building a regime-aware portfolio

A robust portfolio combines uncorrelated strategies that win in different regimes:

Sleeve	Wins in	Losses tolerated in
Trend following (long-only equity)	Bull trends	Sideways, bears
Short premium (condors, weekly)	Quiet ranges	Trends, vol spikes
Long vol (calendars, OTM puts)	Crises, vol expansions	Quiet markets
Mean reversion (pairs / RSI)	Quiet, low-trend	Strong trends
Cash	All (preserves capital)	(always lower expected return)

Allocating across these → smoother equity curve. Diversifying across regimes is more valuable than diversifying across stocks within one regime.

Backtesting across regimes

Always partition your equity curve by regime (or by year, or by VIX bucket) and look at performance separately.

A strategy with 25% CAGR overall might be:

• 60% in bull markets, −15% in bears.
• Or +15%/+18%/+20%/+22% steady — much more robust.

Avoid strategies whose entire profit came from one regime. They will revert.

Common regime mistakes

1. “This time is different.” — It’s never different in the way you think. Markets do mean-revert.
2. Holding the same strategy through a regime change — the textbook way to give back a year of gains.
3. Over-fitting regime detection — too many states, too many parameters → unreliable.
4. Confusing volatility with regime — high vol can occur in both bull (melt-up) and bear (crisis) — need direction + vol jointly.
5. Forgetting that you are part of the regime — your psychology shifts with the market. Your “this isn’t working” feeling often coincides with regime change.

Reading list

• The Most Important Thing — Howard Marks (cycles, contrarian thinking).
• Manias, Panics, and Crashes — Charles Kindleberger.
• Big Debt Crises — Ray Dalio.
• Adaptive Markets Hypothesis — Andrew Lo.