Skip to main content

CustomStoplossWithPSAR Strategy: A "Textbook Example" of Stoploss

Nickname: Stoploss Textbook
Day Job: Quant world's "tutor" — specializes in teaching custom stoploss
Timeframe: 1 hour (no need to stare at screens)

1. What's This Strategy?​

Simply put, CustomStoplossWithPSAR is:

  • An educational example strategy (not for making money, for learning)
  • Demonstrates how to use the custom_stoploss() function
  • Shows how to use PSAR for dynamic stoploss

Like a driving school car: "Don't mind me going slow, I'm here to teach you how to drive!" 🚗

2. Core Settings: It's a "Learning Template"​

Stoploss Rules​

Hard stoploss: Cut at -20% loss (safety net)
Custom stoploss: PSAR dynamic stoploss (the main event!)

Translation: This strategy's focus is teaching you how to implement dynamic stoploss with custom_stoploss(). The trading logic is secondary.

3. When to Buy: Simple to the Point of Pain​

🎯 Buy When PSAR Drops​

Core Logic: PSAR decreasing

In Plain English:

"PSAR switched from rising to falling, trend might be reversing, buy!"

Code Translation:

# Entry condition
PSAR < PSAR from previous candle

Roast: This entry condition is so simple it hurts, but hey, it's an example strategy — the focus is teaching stoploss techniques! 🤣

4. Stoploss Mechanism: This Is the Star​

4.1 PSAR Dynamic Stoploss​

Trigger: Price drops below PSAR

In Plain English:

"PSAR is a parabolic indicator that follows price. When price rises, PSAR rises with it. When price falls, PSAR becomes your stoploss level!"

Code Translation:

# Custom stoploss function
def custom_stoploss(...):
    Get latest PSAR value
    Calculate stoploss ratio = (current price - PSAR) / current price
    Return stoploss ratio

Classic Lines:

  • "This is the main character, entry logic is just the sidekick!"
  • "Learn this and you can write your own dynamic stoploss!"

4.2 Hard Stoploss Safety Net​

Lost 20%? → Cut it (last line of defense)

In Plain English:

  • When PSAR stoploss fails, there's still -20% as backup

5. Exit Logic: No Technical Exits​

5.1 Technical Exit: Doesn't Exist​

# Exit signal
dataframe.loc[:, "sell"] = 0  # No sell signals set

In Plain English:

"I exit via stoploss, who needs sell signals?"

Roast: This strategy is truly "minimalist" — it even skipped sell signals 🤣

6. This Strategy's "Personality"​

✅ Pros (Praise Time)​

  1. High learning value: Top choice for learning custom_stoploss()
  2. Clean code: 70+ lines, understandable in 30 minutes
  3. PSAR application: Shows the right way to use PSAR for stoploss
  4. 1-hour friendly: No screen-staring, perfect for office workers learning
  5. Clear comments: Well-commented code, conscientious example

⚠️ Cons (Roast Time)​

  1. Not for live trading: Entry logic too simple, will lose money
  2. No trend filter: Buys regardless of up or down trend
  3. No BTC correlation: Doesn't care if Bitcoin crashes
  4. No technical exit: Exits purely via stoploss
  5. Example nature: It was never meant to make money

7. When to Use It?​

PurposeRecommendationReason
Learning custom stoplossHighly recommendedThis is what it was designed for
Learning PSAR applicationHighly recommendedStandard PSAR stoploss example
Live tradingNot recommendedEntry logic too simple
Modified for liveMaybeAdd more conditions and protections

8. Bottom Line: How's This Strategy?​

One-Liner​

"A textbook example that can't make money but can teach you to make money"

Who Should Use It?​

  • ✅ People wanting to learn custom_stoploss()
  • ✅ People wanting to learn PSAR application
  • ✅ Quant newbies (simple code)
  • ✅ People wanting to develop their own strategies

Who Should NOT Use It?​

  • ❌ People wanting to make money directly in live trading
  • ❌ People looking for mature strategies
  • ❌ People who don't want to modify code

My Advice​

  1. Learning first: Don't think about live trading, learn first
  2. Understand the principles: Get how custom_stoploss() works
  3. Modify and improve: Add your own logic on top
  4. Test live: Consider live trading only after modifications

9. What Can You Learn?​

9.1 Core Tech: custom_stoploss()​

This is one of Freqtrade's official examples, specifically demonstrating custom_stoploss() usage.

You'll Learn:

  • custom_stoploss() function signature and parameters
  • How to get latest candle data
  • How to calculate and return dynamic stoploss values
  • custom_info data storage tricks

9.2 Performance in Different Markets (Plain English Version)​

Market TypeRatingPlain English Explanation
📈 Slow bull /ranging upward⭐⭐⭐☆☆PSAR stoploss follows trend, but entry logic is simple
🔄 Wide-range oscillation⭐⭐☆☆☆PSAR triggers frequent stoploss in ranging markets
📉 One-way crash⭐☆☆☆☆No trend filter, prone to consecutive losses
⚡️ Extreme sideways⭐☆☆☆☆PSAR crosses frequently, signals chaotic

One-Liner Summary: Good for learning, skip for live trading

10. Want to Run This? Check These Settings First​

10.1 Pair Configuration​

SettingSuggested ValueRoast
Number of pairs10-20It's an example, don't overdo it
Quote currencyUSDTDon't use BTC/ETH as quote
Max open trades1-3Control risk
Timeframe1hMandatory, can't change

10.2 Hardware Requirements (This Strategy Is Friendly!)​

Minimal computation, very low VPS requirements:

PairsMin RAMRecommended RAMExperience
10-20 pairs512MB1GBRuns easily

Warning: This is one of the few quant strategies that can run on a Raspberry Pi 😅

10.3 Backtest vs Live​

custom_info behaves differently in live vs backtest:

In Backtest:

  • Can use current_time for indexing directly
  • Data is pre-loaded

In Live:

  • Need to use get_analyzed_dataframe() to get data
  • Data updates in real-time

Suggested Workflow:

  1. Understand the code logic first
  2. Backtest to see stoploss effect
  3. Paper trade to observe live behavior
  4. Modify and improve before considering live

11. Easter Eggs: The Author's "Little Tricks"​

Look closely at the code and you'll find some fun stuff:

  1. Detailed comments: Lots of comments in the code

    "Afraid you won't understand, I'll make it clear!"

  2. Entry logic ridiculously simple: Just PSAR decreasing

    "I'm focused on teaching stoploss, entry logic is just thrown together!"

  3. Sell signal set to 0: dataframe.loc[:, "sell"] = 0

    "Who needs sell signals? Stoploss isn't good enough?"

12. Final Thoughts​

One-Liner​

"A textbook that can't make money but can teach you to make money"

Who Should Use It?​

  • ✅ People wanting to learn custom stoploss
  • ✅ Quant newbies
  • ✅ People wanting to develop their own strategies

Who Should NOT Use It?​

  • ❌ People wanting to make money directly in live trading
  • ❌ People who don't want to modify code

Manual Trader Tips​

Manual traders can learn the PSAR stoploss approach:

  • PSAR can serve as dynamic stoploss reference
  • When price rises, PSAR rises with it
  • Consider exiting when price breaks below PSAR

13. ⚠️ Risk Reminder (Must Read This Section)​

This Is an Example Strategy!​

CustomStoplossWithPSAR is an educational example, not a production strategy:

This strategy's entry logic is overly simple, live trading may result in losses.

Simply put: It's a training car, not a race car!

Hidden Risks of Example Strategies​

In live trading, simple logic can lead to:

  • Consecutive losses: Entry conditions too simple, win rate may be low
  • Ranging market beatdown: PSAR triggers frequent stoploss in sideways markets
  • No trend protection: Will buy even in downtrends

My Advice (Real Talk)​

1. Treat this as a textbook, don't trade live directly
2. Understand how custom_stoploss() works
3. Add your own entry logic on top
4. Add trend filters, BTC correlation, and other protections
5. Consider live trading only after improvements

Remember: This is a training car. After learning to drive, you need to build your own race car!

Final Reminder: No matter how good the strategy, the market won't say hi before teaching you a lesson. Light position testing, staying alive is most important! 🙏