Most crypto traders spend 90% of their attention on what to buy — and almost none on how much. This is the wrong order. Position sizing is the variable that turns the same strategy into either a durable compounding machine or a guaranteed blow-up. It doesn't matter how good the signal is; the wrong stake size will destroy any edge over time.
Imagine a strategy with a 55% win rate and a 1:1 reward-to-risk — it wins 55 times out of 100, each win equal to each loss. By the maths, this strategy is profitable. Now run two versions of it:
Version A will go bankrupt in a short losing streak — even a normal one. A run of five losses in a row (which a 55% strategy hits regularly) removes 67% of the account. Recovery requires a 200% gain from the remaining balance. Most accounts never come back from that.
Version B survives the same losing streak with 95% of the account intact. It keeps trading. It lets the edge play out over hundreds of trades, which is the only timeframe on which a genuine edge becomes real money.
Same strategy. Opposite outcomes. Position size was the only difference.
Most systematic desks set a maximum risk per trade — the amount they are willing to lose on a single position — at 1–2% of the total account. Not the position size: the risk, meaning the distance to the stop loss expressed as a percentage of the account.
If a trade has a stop loss 5% away from the entry, and the desk risks 1% of the account, the actual position size is 1% ÷ 5% = 20% of the account. At 2% risk with the same stop, it's 40%. The stop distance drives the position size — not a fixed dollar amount per trade.
This means:
A 60% win rate with 25% position sizing will underperform a 50% win rate with 1% position sizing over any meaningful run. The first trader has a better signal but will periodically wipe out; the second will slowly, steadily accumulate. The market routinely rewards the second and destroys the first.
Professional desks are obsessive about this for a simple reason: you cannot trade well from a small or recovering account. Capital preservation is not conservative — it's the prerequisite for staying in the game long enough for the edge to matter.
Leverage multiplies both gains and losses — it does not change the win rate. A 5× leveraged position with a 10% stop means a 50% account loss on a single trade. At that point, the account needs a 100% return just to break even. The maths of drawdown recovery becomes exponentially harder as losses compound: a 50% drawdown requires 100% to recover; an 80% drawdown requires 400%.
This is why leverage without position sizing is not risk management — it's gambling with an eventual defined outcome.
If you follow or copy a trader, position sizing is the hidden variable that makes their track record meaningful or meaningless. Two traders with identical win rates and identical trades can produce completely different account outcomes if one risked 10% per trade and the other risked 1%.
Questions worth asking:
A service that can answer these precisely is one that takes risk management seriously. One that can't answer them — or deflects to "it depends on the trade" — is one where the exit from a bad month will be yours to manage, not theirs.
Every strategy on this desk runs with a fixed risk budget per trade — defined before entry, enforced by the stop. When the book is in a drawdown, we do not increase size to recover faster. When it's in a good run, we do not increase size because things are going well. The parameter is set by the research, not the mood of the week.
This is also why we show losses: the full picture — including how a strategy sizes and responds to loss streaks — is what makes a track record usable. Numbers without context are marketing.
The full strategy book — position sizing parameters, stage of validation, and the trades we killed — is published openly.
See it at glasshousedesk.com/lab. For questions, talk to us.