What stop-loss and take-profit orders do
A stop-loss is an order that closes your position if price reaches a level you set in advance. A take-profit is the mirror image: it closes the position when price reaches your target. Together they turn a trade into a defined bet with a known worst case and a planned best case, made while you are calm instead of while you are watching a candle move against you.
On most exchanges these can be attached to the position at the moment you open it, so the protection exists before you can get distracted or hesitate.
A stop-loss is not a prediction that you will lose. It is the price at which you agree your idea was wrong.
Where to place a stop-loss
The most common beginner method is a fixed percentage: "I always use a 2% stop." It is simple and it ignores the market. A calm pair and a violent one do not need the same distance, and the same pair does not behave the same way in a quiet week and a news week.
Two ideas fix this.
Structure. Put the stop where the setup stops making sense. For a long, that is typically below the most recent swing low or below the support zone you based the trade on. If price returns there, the reason you entered has failed. See support and resistance for how to find those zones.
Volatility. The ATR (Average True Range) measures how far price typically moves per candle. A stop closer than about one ATR to your entry is inside normal noise and will be hit by ordinary movement. Many traders use 1.5 to 2 times the ATR as a starting point, placed beyond the structural level.
Hypothetical example: you buy at 100, the recent swing low is 98, and the ATR on your timeframe is 2. A stop at 97 sits under the swing low with a little room. The distance is 3, or 3%. The same idea on a calmer pair with an ATR of 0.5 might need only a 1% stop. Neither number is "right" in isolation.
Then size the position from the stop, not the other way around. If you risk 1% of a 1,000 account, that is 10. With a stop 3% away, the position is about 333. If 333 feels too large or too small, you change the risk budget, never the honesty of the stop.
Risk-reward and why a 60% win rate can lose money
Risk-reward compares what you stand to lose with what you stand to gain. Risking 10 to make 20 is 1:2.
Win rate is the statistic everyone quotes and the least useful one on its own. Take a trader who wins 60% of trades, with an average win of 10 and an average loss of 20:
- 60 wins x 10 = +600
- 40 losses x 20 = -800
- Net: -200 over 100 trades, before any fee
Now a trader who wins only 40% of the time, average win 20, average loss 10:
- 40 wins x 20 = +800
- 60 losses x 10 = -600
- Net: +200
Fees push both lower. On futures, a round trip costs a fraction of a percent of the position, and it is paid on every trade, winners included. A strategy with a thin edge can be wiped out by the cost of trading it. Always judge a system on expectancy per trade after fees, which is covered in risk management in crypto trading.
Taking profit in parts
A single target forces a choice: exit early and leave gains on the table, or hold for more and watch a winner turn into a loser. Partial take-profits split the decision.
A 30/40/30 ladder is one example of the concept. You close 30% of the position at a first target, 40% at a second, and let the final 30% run with a trailing stop. The first part pays for some of the risk, the middle part banks a real gain, and the last part keeps exposure to a larger move. The percentages are a design choice, not a law. The point is to decide the exit plan before entering, including what happens to the remainder.
Partial exits cost something too: if the trade runs straight to a big target, you earned less than with a full hold. You are trading some upside for a smoother result.
Moving the stop to break-even, and what it costs
After a first target is hit, many traders move the stop to the entry price so the rest of the trade cannot lose. It is a reasonable tool and it is not free.
The cost is that price often comes back to retest the entry before continuing in your direction. A stop at break-even gets hit by that retest, and you exit a trade that later goes exactly where you expected. You also still pay fees, so a "break-even" exit is slightly negative.
Moving the stop too early turns more winners into scratches. Many traders wait until price has cleared a structural level, or trail the stop behind swing points or a multiple of the ATR instead of jumping straight to entry.
Stop hunting and wicks
Crypto candles have long wicks, often through obvious levels where many stops sit. Whether a specific move was a deliberate hunt or just liquidity being taken is usually impossible to prove from the outside, but the practical lesson is the same: a stop placed exactly on a round number or one tick under an obvious low is easy to touch.
Give the stop room beyond the level rather than on it. Watch which price your exchange uses to trigger stops: many offer a choice between last traded price and mark price, and mark price is less sensitive to a single thin-book spike. Also know the order type. A stop-market order fills at the next available price, which may be worse in a fast move. A stop-limit order can fail to fill at all if price jumps through your limit, leaving you in the trade.
Liquidation price versus stop-loss
With leverage, your position has a second, hidden exit: liquidation. If losses use up your margin, the exchange closes the position itself, at its price, often with an extra fee.
Hypothetical example: a long at 100 with 10x leverage and isolated margin is liquidated a bit before 90, because of maintenance margin. A stop at 97 closes the trade with a 3% price loss. A stop at 88 is beyond liquidation and will never execute, because the exchange got there first.
Rule of thumb: your stop should be comfortably inside the liquidation price, and if it cannot be, the leverage is too high for that setup. The details are in crypto leverage explained.
The most expensive mistake: a signal with no stop
The costliest habit in retail trading is acting on a call that contains only a pair and a direction: "BTC long." It feels like information. It is missing the three things that define a trade: where to enter, where you are wrong, and how big to go.
Without a stop you do not know your risk, so you cannot size the position. Without a size, the same signal can cost 0.5% of your account or 30%, depending on what you happened to click. And because you never decided where you are wrong, the market decides for you, usually at the worst moment and often through liquidation.
For that reason, Darwin Lab's free Telegram channel, which posts only the pair and the direction without entry, stop or size, is not a tradeable setup on its own. The system traded real money from April to June 2026 and has run on paper since. If you use any call as an idea, build the stop, the target and the size yourself before you place an order.
A short checklist before any trade
- Where is the idea invalid? That is the stop.
- How far is it from entry, in percent and in ATR?
- How much of the account is lost if it is hit? Keep it small and fixed.
- What size follows from that, and is the stop inside liquidation?
- What is the plan for the target, and for the remainder after partials?
This article is educational and not financial advice. Futures trading carries a high risk of loss.
Related: Risk management in crypto trading | Crypto leverage explained | What is a trading signal? | Join the free channel