Trend following vs mean reversion: the two ways to make money

Strip away the indicators and the jargon and almost every trading strategy ever written is one of two bets. Either you believe a move that has started will continue — trend following — or you believe a move that has stretched will snap back — mean reversion. They are mirror images. They make money in opposite conditions. And a startling number of traders run one without knowing which one it is.

Trend following — betting on continuation

A trend-following strategy enters in the direction of an existing move and bets it persists. Breakout entries, moving-average crossovers, momentum strategies — all trend following. Its signature payoff profile is distinctive: a low win rate, often well under 50%, paired with a high reward-to-risk ratio. It loses small and often, when trends fail to materialise, and wins big and rarely, when it catches a move that runs. Most of the year's profit comes from a few trades. The hard part is psychological: you must take many small losses patiently to be present for the few large wins.

Mean reversion — betting on the snap-back

A mean-reversion strategy does the opposite: it enters against a stretched move, betting price returns to an average. Buying oversold dips, fading band touches, range trading — all mean reversion. Its payoff profile is the inverse of trend following: a high win rate, often well above 50%, paired with a low reward-to-risk ratio. It wins small and often, collecting the snap-backs, and loses big and rarely — when a move it faded does not reverse but keeps going. The hard part is also psychological, but different: one un-reverted trend can erase many wins, so the rare loss must be controlled ruthlessly.

They need opposite regimes

Here is the unifying point. Trend following needs trending markets and gets chopped up in ranges. Mean reversion needs ranging markets and gets steamrolled in trends. Neither is better — they are tools for different weather. Crypto, helpfully, supplies plenty of both: long ranging stretches punctuated by violent trends. The implication is that no single strategy is always right, and a strategy underperforming is often not broken — it is simply in the wrong regime. This is exactly why a regime filter like ADX is so valuable: it tells each strategy when its weather has arrived.

Combining them

Because they profit in opposite regimes, a trend strategy and a mean-reversion strategy are naturally diversifying — when one is struggling, the other often is not, and the combined equity curve is smoother than either alone. Two ways to combine them: run both continuously as a portfolio and let their offsetting drawdowns net out, or use a regime filter as a switch that activates whichever one suits the current market. The first is simpler and robust; the second is more efficient but adds the risk of misjudging the regime.

Which should you build first?

Pick the one whose psychology you can actually tolerate. If a long string of small losses while waiting for a big win would make you abandon the system, trend following will defeat you regardless of its edge. If watching a faded move keep running against you would make you freeze, mean reversion will. The math of both works; the trader is the variable. Build the one you can hold through its worst stretch.

Noon Barbari's strategy designer ships reference templates for both families — trend (EMA cross, MACD) and mean reversion (RSI, Bollinger) — and the multi-strategy portfolio feature lets you run both together and see the combined, smoother curve. Backtest each across a trending and a ranging period to see the regime dependence for yourself.

Trend following and mean reversion are the two halves of trading. Know which one you are running, know which weather it needs, and you stop mistaking a regime change for a broken strategy.