Wick is in development, not yet a registered investment adviser, and currently offers no advisory services. joining the waitlist creates no advisory relationship.
Chapter 5
Algo trading risk management
sizing your position. setting your kill switch. deciding in advance what you'll hold through.
Published by Wick·
Every strategy draws down. The ones worth running have done it before — and recovered. The track record shows you exactly that.
Algo trading risk management isn't about avoiding losses — every strategy draws down. It's about deciding in advance how much you're willing to lose, under what conditions you'll stop, and making those decisions before you're under the pressure of a live drawdown. The worst time to set a stop-loss is when you're already in one.
kill switch
Halts the strategy immediately — no new orders are placed. Open positions may still need to be closed separately. The kill switch is the most immediate control available. Its location and behavior varies by platform.
position limits
Set the maximum capital any strategy can deploy — defined as a percentage of account, a fixed dollar amount, or a notional cap. Prevents any single strategy from exceeding its defined allocation, even if the strategy's own logic would take a larger position.
strategy stop-loss
Define a maximum loss threshold for a strategy. If it's hit, the strategy halts new entries. This is different from a trade-level stop-loss — it's a portfolio-level circuit breaker that kicks in when cumulative losses reach a defined point.
Position sizing in practice
How should you size your position in an algo strategy?
Position sizing is where much of the real risk management happens — and a decision that often gets made by feel rather than design. Getting it right doesn't require you to be a quant. It requires honesty about what drawdown you can absorb without stopping the strategy. How much capital you allocate determines the real-world impact of every drawdown. Not the percentage on the leaderboard. The actual number in your account.
position sizing example
Strategy max drawdown-24% (from its live track record)
Account size$50,000
Allocated at 20%$10,000 deployed → worst case loss: $2,400
Allocated at 50%$25,000 deployed → worst case loss: $6,000
Allocated at 100%$50,000 deployed → worst case loss: $12,000
The strategy's max drawdown doesn't change. What changes is what it means to your account. Algo trading risk management starts with being honest about which of those numbers you could absorb without stopping the strategy — then sizing accordingly. Running a strategy at a position size that would force you to stop during a normal drawdown defeats the purpose of having risk controls at all.
Setting a position size you can hold through the strategy's historical max drawdown — without panicking or stopping — matters more than picking the highest-returning strategy on the leaderboard.
What a drawdown actually feels like
What does a drawdown actually feel like to live through?
Two strategies. Same ballpark returns. Completely different psychological experiences.
equity curve comparison
conservative
aggressive
Common questions
Algo trading risk management — FAQs
What is a drawdown in algo trading?
A drawdown is the decline in value from a strategy's peak to its lowest point before recovering. If a strategy reaches $110,000 and then falls to $80,000 before climbing again, the drawdown is -27%. Maximum drawdown is the largest such decline over the strategy's entire history — it tells you the worst loss you would have experienced if you ran the strategy from the worst possible start point.
How much of my portfolio should I allocate to an algo strategy?
A commonly used starting point is allocating no more than you could sustain through the strategy's historical max drawdown without stopping it. If a strategy has a historical max drawdown of 15%, and you're comfortable losing up to 5% of your total portfolio at once, size your allocation so that a 15% strategy drawdown equals a 5% portfolio impact. Start smaller than you think you need to — you can always scale up after observing live behavior.
What is position sizing in algo trading?
Position sizing determines how much capital you allocate to an algo strategy relative to your overall account. It is the primary lever you control as a strategy runner — the strategy's logic is fixed, but the real-world dollar impact of every trade and drawdown is determined by how much capital you put behind it.
What is a stop-loss in algo trading?
A stop-loss is a pre-set threshold that automatically pauses or stops a strategy when losses exceed a defined level. It is set before going live — not during a drawdown. In algo trading, stop-losses protect against catastrophic loss in the event of unexpected market conditions or strategy malfunction, separate from the strategy's own internal risk logic.
What is a kill switch in algo trading?
A kill switch is a manual control that immediately stops a running strategy — halting all new orders. It doesn't automatically close existing open positions, which may need to be managed separately. The kill switch is the most important control to locate before going live. In a fast-moving market or unexpected event, you want to know exactly where it is.
What's the difference between a stop-loss and a kill switch?
A stop-loss is a pre-set threshold — if the strategy loses more than X%, it automatically halts new entries. It's a rule-based circuit breaker that triggers without intervention. A kill switch is a manual control you activate yourself. Both are important: the stop-loss protects you when you're not watching, the kill switch gives you immediate control when you are.
key takeaway
Drawdowns are not failures — they are a normal part of any strategy's lifecycle. The question is not whether a strategy loses, but whether you have the controls to manage it when it does.