Mean Reversion

Quick answer: Mean Reversion is a futures approach that assumes returns are negatively autocorrelated — that a market stretched away from its average tends to snap back — so it fades the extreme, winning often but risking rare catastrophic losses when the move keeps going.

In simple words

Mean reversion is the opposite bet to trend following. It assumes that when a market moves too far, too fast away from its recent average, it tends to snap back toward that average. So the trader sells strength and buys weakness — fading the move rather than joining it. This wins often, because prices really do wobble around an average much of the time. The danger is the rare occasion when the move does not come back but keeps going. Those few losses can be very large, which is why it is described as picking up pennies in front of a steamroller.

What it looks like

Professional explanation

The assumption: returns revert, not persist

Mean reversion rests on the claim that returns are negatively autocorrelated over the chosen horizon — a move up is more likely to be followed by a move down, and the market oscillates around an average rather than trending away from it. This is the exact opposite of the trend-following assumption, and both cannot be true for the same instrument on the same timeframe at the same time. When the reversion regime holds, fading extremes wins frequently. When it does not — when a trend takes hold — the same fades stand directly in front of the move, and the losses come from exactly the regime the method assumes away.

The mirror-image return profile: pennies before a steamroller

Mean reversion has the opposite skew to trend following: many small wins and rare, severe losses — a long left tail. Most of the time the fade works and books a modest gain; occasionally the move does not revert and inflicts a loss far larger than any single win. This is why it is described as picking up pennies in front of a steamroller. The frequent wins make it feel reliable and tempt the trader to add size, which is precisely the behaviour that turns the rare steamroller loss from painful into ruinous.

Why it and trend following are mirror opposites that both survive

Trend following bets returns persist; mean reversion bets they revert. On the same instrument and timeframe only one can be right at once, so they cannot both be correct there simultaneously. Yet both survive as strategies because they apply to different instruments and different timeframes — an index may mean-revert intraday while trending over months, and a commodity may trend where an index range-trades. The existence of both is not a contradiction; it is evidence that the correct regime assumption is local, not universal, and no rule tells you which one you are in now.

Undefined risk meets the steamroller

A futures position has no structural cap on loss, and mean reversion deliberately positions against the immediate move. Combine the two and the rare non-reverting event becomes especially dangerous: the trader is short a rising market or long a falling one precisely when it accelerates, and on a leveraged, gap-prone index the stop that was meant to cut the steamroller loss can be leapt over entirely on an overnight move. The high win rate encourages larger size, so the position tends to be biggest exactly when the regime is about to break.

Construction

1. Identify a market oscillating around an average rather than trending, accepting that this regime judgement is a bet that cannot be confirmed in advance.
2. Wait for price to stretch away from that average by an amount judged extreme relative to recent behaviour.
3. Take a futures position against the stretch — short an overextended rise, long an overextended fall — betting on a return toward the mean.
4. Define the adverse level beyond which the reversion thesis is abandoned, recognising this stop stands in front of a possibly accelerating move and can be gapped through.
5. Exit near the average as the snap-back completes, since the edge is the frequent modest reversion, not an open-ended run.

Market outlook

A trader may study mean reversion when a market appears to be oscillating around a stable level — range-bound, with moves that overshoot and return — rather than trending. The approach assumes negative autocorrelation: stretches away from the average tend to reverse. It is invalidated the moment a trend takes over, because then every fade stands in front of a persistent move and the rare catastrophic loss materialises. The regime that destroys it — a sustained trend — gives no advance notice, and the very frequency of small wins beforehand tends to lull the trader into larger size just as the reversion assumption is about to fail.

Risk profile

Mean reversion carries undefined risk, and its risk is uniquely treacherous because the return profile hides it. The futures position has no long option leg capping the loss; the structural bound is zero for a long or nothing for a short. The trader fades the immediate move, so a failure to revert means being short a rising market or long a falling one as it accelerates. The stop meant to cut that loss can be gapped through on a leveraged index. Worst of all, the high win rate encourages size, so the position is often largest exactly when the rare non-reverting move — the steamroller — arrives.

Reward profile

The reward is frequent and modest: most fades book a small gain as price returns toward the average, producing a smooth run of wins that feels dependable. Because the target is the mean rather than an open-ended move, the profit per trade is deliberately capped by the method's own logic — the trader exits near the average rather than letting it run. This is the mirror of trend following: many small wins instead of a few large ones. The catch is that the frequency of wins is not the same as positive expectancy, because a single non-reverting loss can erase a long string of them.

Margin requirement

The position is held against full naked-futures SPAN plus exposure margin, marked to market daily, with no spread benefit because there is no offsetting leg. Mean reversion is especially exposed here: fading a move means the position is often adverse before it reverts, so daily marks run against it early, and a move that keeps going can trigger a margin call at the worst possible moment. The high win rate tempts larger size, which raises margin and the damage of a square-off. Margin percentages and lot sizes are revised periodically.

Greeks exposure

Volatility impact

Implied volatility does not apply to the futures position — there is no vega — so the options market's expected-move pricing does not move the futures price, and the 'IV regime' field is not meaningful here. What matters is realised volatility: a rise in realised volatility means larger stretches away from the average and larger snap-backs, but also a higher chance that a stretch is the start of a trend rather than an overshoot. So a volatile environment both widens the potential reversion and raises the odds of the non-reverting steamroller — the volatility environment matters through actual price behaviour, not through any premium paid or received.

Time decay

There is no time decay. The futures position carries no theta, so waiting for a stretched move to revert costs nothing per day. This is a clean contrast with expressing mean reversion through options: a short-option version would collect theta (and take on its own undefined risk), while a long-option version would pay theta and fight the clock. The futures position simply waits, indifferent to time, until price reverts or the thesis is abandoned. The only time-linked cost is the roll spread if held across expiry, rarely a factor on a days horizon.

Practical examples

Lot sizes used above (NIFTY 75, BANKNIFTY 30) are those in force at the time of writing; NSE revises them periodically. Figures exclude brokerage, STT, exchange charges, stamp duty and GST, all of which materially affect small spreads.

Common mistakes

• Adding size because the fades keep working, so the position is largest exactly when the rare non-reverting move arrives and turns a manageable loss into a ruinous one.
• Mistaking a high win rate for positive expectancy, when a single steamroller loss can erase a long string of small wins.
• Averaging down into a losing fade on the belief that reversion must be near, when the move may be a trend that keeps accelerating.
• Fading a genuine trend as if it were an overshoot, standing in front of exactly the regime the method assumes away.
• Sizing off the small typical loss rather than the rare large one, then facing a margin call when the steamroller day comes.
• Placing the stop so close that normal reversion noise tags it, or so far that the rare loss is catastrophic — both failures of respecting the undefined downside.
• Ignoring that a gap can leap the stop entirely, so the loss on the one move that does not revert is realised far beyond the planned level.

Advantages & disadvantages

Adjustments & exits

• A trader may cap position size deliberately so that a single non-reverting move cannot exceed a fixed fraction of capital, accepting smaller wins in exchange for surviving the steamroller.
• A trader may require a stretch to be extreme by a demanding measure before fading, reducing the number of fades that turn out to be trend continuations at the cost of fewer opportunities.
• A trader may take partial profit as price approaches the average rather than waiting for a full return, locking in the frequent modest gain the method depends on before a reversion stalls.

Adjustment is a decision about risk, not a way to rescue a losing view. See Adjustments and Exit Planning.

Professional usage

Quantitative desks run mean reversion as a systematic, heavily risk-controlled programme across many instruments, sizing each fade small so no single non-reverting move can dominate, and modelling the reversion statistically rather than by eye. They cross-margin, hedge, and cut exposure when their models detect a shift toward trending — protections a retail trader rarely has. The concept is replicable by retail on an index future, but the risk controls that keep the steamroller from being fatal are the hard part, and without them a retail mean-reversion trader is most exposed precisely when the method feels most reliable.

Frequently asked questions

Voice search & related questions

Natural-language questions people ask about the Mean Reversion.

Sources & references

• NSE India — Equity derivatives (futures) education
• SEBI — Investor education and studies on individual trader outcomes in derivatives
• CME Group — Futures education

Last reviewed 9 July 2026. Educational content only — not investment advice.