In the early hours of a deep autumn in 2019, I stared at the 5-minute K-line of ETH, my hands trembling. The night before, I saw it break through $200 and chased after a long position, thinking I could make a quick buck. Relying on a little unrealized profit in my account, I was too lazy to think much—stop-loss? Why bother, this market looks very stable.

As a result, at three in the morning, a sudden negative news hit the market, and ETH fell below $180 within half an hour. I watched the loss in my account jump from 5% to 15%, and I could only see the margin warning line slowly creeping up.

When I hurriedly closed my position, my account had already evaporated nearly 200,000. That morning I sat in front of the computer, smoked half a pack of cigarettes, and finally understood what veteran traders meant by 'stop-loss is a lifeline'—it's not just talk; it can really save your life.

After ten years of trading, I have transformed my bloody lessons into this practical strategy for stop-loss and profit-taking. Without further ado, let's get straight to the essentials!

1. The core of stop-loss: Replace 'luck mentality' with 'certain signals'.

The essence of stop-loss is not 'admitting mistakes', but 'filtering ineffective fluctuations with controllable losses'. The three methods I mentioned just happen to cover different scenarios of certainty:

  • Technical stop-loss: Rely on 'market consensus levels' for support. Support and resistance levels, trend lines, ATR, etc., essentially capture 'reversal signals recognized by the majority'. For example, if a double bottom support breaks below $500, it's not that I arbitrarily set $500, but rather to leave a buffer for normal fluctuations to avoid being washed out by 'false breakouts'; ATR stop-loss is even harsher, I directly follow market volatility, using 3 times ATR for high-volatility assets (like small coins) and 1.5 times for low-volatility assets (like BTC), completely ignoring 'reasoning with the market', only recognizing data.

  • Capital management stop-loss: Provide 'insurance for the account'. A fixed risk of 1%-3% of the principal is how I mathematically ensure that 'a single loss won't hurt too much', even if I make 5 consecutive mistakes, the account can still retain over 85%. This is what I often say: 'preserving the principal is the key to turning things around'; taking profits in batches is even better, I will split the positions into 2-3 parts, with strict stop-loss for the first part, and a wider stop-loss after confirming the trend for the second part, effectively controlling risk without easily missing the trend, perfectly solving the dilemma of 'fearing a rebound after a stop-loss'.

  • Breakout stop-loss: Capture 'trend death signals'. When the head and shoulders neckline or the upward trend line is broken, it essentially signals that the trend structure has been destroyed. At this point, not stopping losses is turning a 'small pullback' into a 'deep loss'. I have always emphasized 'leave the market immediately when the trend line is broken', which is actually about 'cutting the tail to survive' at the initial stage of a trend reversal.

2. The key to taking profits: Let profits 'have boundaries but no ceiling'.

The most counterintuitive aspect of taking profits is 'being afraid of missing out on gains while also fearing a reversal.' My strategy happens to address these two pain points:

  • Technical target level for profit-taking: Set a 'visible ceiling' for profits. Fibonacci extension levels and previous high resistance levels are where I find 'natural profit-taking points' in the market through historical patterns. For example, taking profit when BTC hits $60,000 is not because $60,000 is a magic number, but because there are large amounts of trapped and profit-taking orders, making it likely to pull back.

  • Dynamic profit-taking: Let profits 'run on their own'. Moving average profit-taking and trailing stop-loss are essentially about 'not guessing the top' and letting the trend carry the profits. For instance, reducing positions at the 5-day line and liquidating at the 20-day line is actually using the 'lagging nature' of moving averages to catch the tail end of a trend; 5% trailing stop-loss is even more extreme, the higher the price rises, the higher I set the stop-loss, which is like giving profits an 'automatic escalator'; if it keeps rising, I keep earning, and as soon as it pulls back, I lock in profits.

  • Cycle emotion profit-taking: Take advantage of 'market cycles' to lock in profits for safety. Taking profits 12-18 months after halving and reducing positions when the fear and greed index exceeds 90 is how I step out of K-lines to observe the larger cycle—during the later stages of a bull market when 'everyone is shouting that it will go up', it is precisely a risk accumulation period. What I have always said about 'withdrawing 5% when profits exceed 20%' is essentially about using cycle emotions to cool down greed.

3. The 'adaptability of stop-loss and profit-taking for different strategies': ** Don’t use short-term logic for trends, and don’t use trend logic for dollar-cost averaging - short-term trading **: Fast entry and exit, I set the stop-loss to be 'harsh and narrow' (2%-5%).

Because short-term fluctuations are fast, the time left for reactions is short, and the stop-loss is so wide that the profit cannot cover the risk; taking profits of 10%-20% is also reasonable. Short-term trading relies on 'win rate', not 'odds'; taking profit when it's good is the way to accumulate small gains.

- ** Trend trading: betting on 'big movements', stop-loss must be 'wide and stable' (5%-10%), because there will inevitably be pullbacks in a trend, and a too-narrow stop-loss can easily be shaken out by 'normal fluctuations'; taking profits of 50%-100% is a trade-off of 'time for space', during which moving averages and trend lines are my 'position comfort pills', as long as they hold, I will hold on.

- Investment strategy **: Profiting from 'cyclical dividends', so I have 'no single stop-loss' but 'interval profit-taking'. Investment relies on diversifying purchases to dilute costs, and there’s no need to obsess over single stop-losses; however, taking profits within a bull market range of 300%-500% is based on historical cyclical patterns—profit peaks for dollar-cost averaging in the Bitcoin bull and bear cycles generally occur 1-2 years after halving, and if you don't take profits at this time, you might just ride a rollercoaster.

Finally, a practical detail: ** Stop-loss and profit-taking should 'change with the characteristics of the asset'. **

In practical operations, I've found that strategies for mainstream coins like BTC and ETH need to be adjusted to be more 'aggressive' when applied to small coins (market cap below 1 billion):

  • The volatility of small coins is 3-5 times that of BTC. For technical stop-loss, I will use 3 times ATR and expand the stop-loss buffer to 10%-15% (for example, if the previous low is $1, set the stop-loss at $0.85), otherwise, I will be punctured by stop-loss in seconds.

  • Taking profits is even more about 'taking gains when they are good'. Small coins do not have the consensus support like BTC, so when the price rises by 50%, I will take profits in batches; don’t wait for the '1.618 extension level'—many small coins cannot complete a full trend, so take half of the profits first.


The core of my strategy has never been about 'setting precise levels', but about 'locking in risks with rules and maximizing profits with patterns', turning those years of 'the bloody taste of staring at the screen in the early morning' into 'the mechanical feeling of executing rules at the push of a button'—this is the key to turning trading from 'gambling' into 'strategizing'.

If you are also a tech enthusiast and are deeply researching technical operations in the cryptocurrency space, feel free to follow @钱包守护者 . You will gain more insights!