A strategy's rules matter, but so does how much capital it puts behind each trade - and the right size isn't a fixed number. It should change depending on the market regime, because the same size that's reasonable in a calm bull trend can be reckless in a violent, choppy one.
Why one fixed size is a mistake
Position size determines how much a single bad trade can cost. If size stays fixed regardless of regime, two things go wrong:
- In a high-volatility regime, a fixed size that was fine in calm conditions now risks a much bigger loss per trade, because the coin can move several times as far in the same amount of time.
- In a choppy regime, a fixed size sized for trending conditions gets used to take many more losing trades (false breakouts, failed reversals), compounding the damage of a market that doesn't reward the strategy's core assumption in the first place.
Sizing has to answer the same question every time: how much could this specific trade, in this specific regime, realistically cost if it's wrong?
How sizing changes by regime
Bull. Trend confirmation is more reliable, and reversals tend to be gradual rather than violent. This is where a trend-following strategy can reasonably size up - not recklessly, but with more conviction than in choppier conditions.
Bear. Directional longs are stood down or sized down sharply, since down-moves in crypto are often sharper and faster than up-moves. Market-neutral and short-premium approaches, which don't depend on picking a direction, become the more reasonable places to hold size.
Choppy / Neutral. No clear trend to lean on. This is where a trend-following strategy should size down, not up - every signal is more likely to be a false start. Mean-reversion strategies can be more appropriate here, but even they should size conservatively, since a genuinely choppy market can break in either direction without warning.
High-volatility. Regardless of direction, larger average moves mean a given position size represents more real risk. The reasonable response is smaller size, wider stops relative to normal, or standing down entirely until volatility normalizes.
Layer-1s, DeFi, and meme coins don't size the same way either
Regime isn't the only input - a coin's own category matters too. A high-beta layer-1 that can swing much harder than Bitcoin in either direction gets sized smaller than a large-cap major, even in the same regime. A meme coin, driven more by social momentum than fundamentals, gets an even smaller size and a faster exit, because its moves are less about a trend maturing and more about a narrative catching or fading quickly.
How zengtrade applies this
zengtrade's risk governor combines the current regime read with a coin's own category profile to size every position - not as a single global number, but as a function of both. A trailing "chandelier" exit lets a genuine trend run further before cutting it, while a fixed-target exit locks in a reversion trade once it's captured the move it was designed for. Position sizing and exit style aren't decorations on top of a strategy; in a real sense, they are the strategy - two systems with identical entry signals but different sizing and exit rules will produce very different outcomes over time.
None of this removes risk. It's a framework for taking a known, bounded amount of risk per trade rather than letting the market's mood decide it for you.
Authored by zengtrade Quantitative Research Group • Reviewed by Algorithmic Risk Committee: Every model, friction parameter (35 bps round-trip friction), and signal rule is backtested against live Binance spot data. zengtrade is strictly non-custodial and paper-first. Read our Regime Methodology and Risk Disclosures.
Educational content, not investment advice. zengtrade is paper-first and non-custodial.