Momentum vs Mean Reversion
Two strategies give opposite instructions on the same chart. Which is wrong?
The idea
Momentum expects a move to persist; mean reversion expects deviation to shrink. Each works only under suitable regimes and horizons.
In the real world
A trend strategy buys strength; a range strategy sells an extreme back toward average.
Going deeper
The two strategies encode opposite assumptions about the current regime, and both have solid historical records because both regimes occur.
That makes regime identification the actual problem, and it is much harder than either strategy. Trending conditions reward buying strength and punish fading extremes; range-bound conditions do the reverse, and the transition between them is where most systematic losses happen. A strategy's historical record is a record of how often its assumed regime occurred in the sample, which is not a forecast of how often it will occur next.
Where it stops applying
Regimes are only reliably identifiable after the fact, and rules that classify them in real time are themselves fitted to history. Diversifying across both is often more honest than choosing.
Why it matters
Neither is wrong in general, so the useful question is which regime you are actually in.
Try this today
Define regime, horizon, entry, and failure condition before selecting either.
Test yourself
A momentum system says buy the breakout; a mean-reversion system says sell the extreme. Both have solid historical records. What decides which applies?
Show the answer
The regime and the horizon. Trending conditions reward persistence and punish fading; range-bound conditions do the reverse. Each strategy encodes an assumption about which state the market is in, and that assumption is the thing to check first.
Learn this in the feed Answering from memory, then again days later, is what makes it stick.