MultiMA_TSL: The "Multi Moving Average" Trend-Following Master
Nickname: MA Master
Profession: Quant world's "collector" — collects EMA, ZEMA, TRIMA three moving averages
Timeframe: 5 minutes (short-term player)
1. What's This Strategy?
Simply put, MultiMA_TSL is:
- A strategy using 3 different MAs (EMA, ZEMA, TRIMA)
- A strategy with custom stoploss
- A strategy with protection mechanisms (Cooldown + LowProfitPairs)
Like a cautious buyer who only buys on sale: "EMA on sale? ZEMA on sale? TRIMA on sale? All satisfied? Then buy!" 🛒
Roast: This strategy is really "indecisive", three MAs for you to choose! 🤣
2. Core Settings: Simply "Offset Trading + Dynamic Stoploss"
Profit-Taking Rules (ROI Table)
Make 10% right after buying? → RUN!
Hold 30 minutes and make 5%? → RUN!
Hold 60 minutes and make 2%? → RUN!
Translation: This strategy is classic "trend-following thinking", 10% ROI is relatively high, expects to capture big trends!
Roast: 10% ROI? This strategy is ambitious, wants to catch big fish! 🎣
Stoploss Rules
Hard stoploss: Cut at 15% loss
Trailing stop: Activates after 1.8% profit, run if pulls back 1%
Translation: -15% hard stoploss is looser than common strategies, gives trend fluctuation room — classic "let profits run" thinking 💰
3. Entry Conditions: Three MA Modes to Choose From
This strategy's entry conditions use "offset trading" approach:
🎯 Core Logic: Offset Trading
Offset trading = Only buy when price is below certain percentage of MA
EMA offset: Price < EMA50 × 0.958 (95.8%)
ZEMA offset: Price < ZEMA30 × 0.963 (96.3%)
TRIMA offset: Price < TRIMA14 × 0.958 (95.8%)
In Plain English:
"EMA says 100 bucks? I only buy at 95.8! Only buy on sale!"
🎯 EWO + RSI Confirmation
Not enough that price is low, also need EWO wave oscillator confirmation:
(EWO < -20) OR ((EWO > 6) AND (RSI < 50))
In Plain English:
"Either deep pullback (EWO < -20), or trend pullback (EWO > 6 but RSI not overbought) — then consider buying"
🎯 Three MA Mode Comparison
| Mode | MA Type | Period | Offset | Characteristics |
|---|---|---|---|---|
| EMA mode | EMA | 50 | 0.958 | Standard, moderate response |
| ZEMA mode | ZEMA (zero-lag EMA) | 30 | 0.963 | Fast, almost no lag |
| TRIMA mode | TRIMA (triangular MA) | 14 | 0.958 | Smooth, filters noise |
Roast: Three MAs for you to choose, this is the legendary "three-choice dilemma"! 🤣
4. Protection: More Luxurious Than Previous Strategies
This strategy's protection is quite luxurious:
| Protection Type | Function | Plain English |
|---|---|---|
| CooldownPeriod | Pause 2 candles after entry | "Just bought, cool down 2 candles before next buy" |
| LowProfitPairs | Pause losing pairs | "This pair not making money, pause 20 candles" |
| Trailing Stop | Activates after 1.8% profit | "Follow price after making money, run if pulls back 1%" |
| Custom Stoploss | Quick exit when sell signal triggers | "Sell signal came? Set stoploss to 0.001, run fast!" |
Roast: Custom stoploss is a highlight of this strategy — almost immediate stoploss when sell signal triggers, doesn't let profits slip away! 🚀
5. Exit Logic: Simpler Than Entry
5.1 Technical Exit: Price Above Offset
Trigger:
(Price > EMA32 × 1.002) OR (Price > TRIMA48 × 1.085)
In Plain English:
"Price is above MA offset already (rose enough) — if you don't run now, what are you waiting for?"
Roast: EMA offset 0.2% to sell, TRIMA offset 8.5% to sell? That's a big gap! 🤔
5.2 Custom Stoploss: Quick Exit When Sell Signal Triggers
This is a big feature of this strategy:
def custom_stoploss(...):
if (sell_signal == 1) AND (buy_signal == 0):
return 0.001 # Almost immediate stoploss
return 1 # Default: don't trigger
In Plain English:
"Sell signal came? No matter how much profit, run fast!"
Note: This feature only effective in live/dry-run mode, not in backtest!
5.3 ROI Exit: 3-Level Take-Profit
Profit Hold Time Trigger Exit
─────────────────────────────────────
10% Anytime Run when reached (big profit)
5% After 30 min Run when reached (medium profit)
2% After 60 min Run when reached (small profit)
In Plain English:
- Make 10% right after buying? → Pie from heaven, run!
- Hold 30 minutes and make 5%? → Not bad, run!
- Hold 60 minutes and make 2%? → Mosquito leg is still meat, run!
6. This Strategy's "Personality Traits"
✅ Advantages (Praise Session)
- Multiple MA types: EMA, ZEMA, TRIMA three options, adapts to different market rhythms
- Offset trading: Captures deep pullbacks, better entry prices
- Custom stoploss: Quick exit when sell signal triggers, locks profits
- EWO filter: Wave oscillator identifies trends and pullbacks
- Protection mechanisms: Cooldown + LowProfitPairs dual protection
- Hyperparameter optimization: Supports Hyperopt for key parameters
⚠️ Disadvantages (Roast Session)
- High complexity: Multiple MA + custom stoploss, debugging is painful
- No BTC correlation: Doesn't know when Bitcoin crashes
- Parameter sensitivity: Offset amounts and MA periods need optimization for different pairs
- Backtest differences: Custom stoploss not effective in backtest, may differ from live
- Medium computation: Multiple MA types increase computational burden
7. Applicable Scenarios: When to Use It?
| Market Environment | Recommended Action | Reason |
|---|---|---|
| Ranging market | Highly recommended | Offset trading best for ranging pullbacks |
| Uptrend | Recommended | Custom stoploss can ride green candles |
| Downtrend | Pause or light position | No long-term trend filter, easy to lose consecutively |
| High volatility | Adjust parameters | May need to adjust offset amounts |
| Low volatility | Lower ROI | Lower ROI threshold for small moves |
| BTC crash | Pause | Big brother crashed, watch first |
8. Summary: How's This Strategy?
One-Sentence Review
"A trend-following master using three MAs, with custom stoploss and protection mechanisms"
Who Should Use It?
- ✅ People who like multi-indicator combinations
- ✅ People who can accept some complexity
- ✅ People with quantitative foundation
- ✅ Friends with VPS RAM 2GB or more
Who Should NOT Use It?
- ❌ People who like simple strategies (this strategy has many conditions)
- ❌ People wanting to bottom-fish in downtrends (no long-term trend filter)
- ❌ People who don't want to optimize parameters
- ❌ Pure quantitative newbies (need to understand multiple MAs)
My Recommendations
- Backtest first: Understand strategy behavior in different markets
- Add BTC filter: Consider adding BTC trend filter for extra protection
- Dry-run test: Test at least 1-2 weeks before live trading
- Start small: Begin with small capital, increase after confirming stability
9. What Markets Make Money with This Strategy?
9.1 Core Logic: Riding Trends with Multiple MAs
MultiMA_TSL is a trend-following strategy. Its core philosophy is:
"One MA might give false signals, but three MAs with offset confirmation should be more reliable, right?"
- EMA faith: Classic exponential moving average, widely used
- ZEMA faith: Zero-lag version, faster response
- TRIMA faith: Triangular smooth, filters noise
9.2 Performance in Different Markets (Plain English Version)
| Market Type | Performance Rating | Plain English Explanation |
|---|---|---|
| 📈 Slow bull/ranging up | ⭐⭐⭐⭐☆ | Offset entry + custom stoploss works well |
| 🔄 Wide ranging | ⭐⭐⭐⭐☆ | Pullback entry suitable for ranging |
| 📉 Single-sided crash | ⭐⭐☆☆☆ | No trend filter, may buy and get stuck consecutively |
| ⚡️ Extreme sideways | ⭐⭐⭐☆☆ | Signals reduce, but offset filters false signals |
One-sentence summary: Makes money in ranging and uptrends, be careful in crash markets
10. Want to Run This Strategy? Check These Configs First
10.1 Pair Configuration
| Configuration | Recommended Value | Roast |
|---|---|---|
| Number of pairs | 15-30 | Moderate signal frequency |
| Max positions | 3-5 | Control risk, don't be greedy |
| Position mode | Fixed position | Recommended fixed, control risk |
| Timeframe | 5m | Mandatory, can't change |
10.2 Hardware Requirements (Moderate Level)
This strategy uses multiple MAs, moderate computation:
| Number of Pairs | Minimum RAM | Recommended RAM | Experience |
|---|---|---|---|
| 15-30 pairs | 1GB | 2GB | Normal |
| 30-60 pairs | 2GB | 4GB | Comfortable |
Warning: Custom stoploss only effective in live mode — backtest results may be optimistic!
10.3 Backtest vs Live Differences
Backtest problems:
- Custom stoploss not effective
- May miss some exit behaviors
- Results may be overly optimistic
Recommended process:
- Backtest with default parameters first
- Adjust parameters, see how results change
- Dry-run test at least 1-2 weeks
- Small capital live test
- Confirm stability before adding capital
Don't go all-in immediately, backtest isn't everything!
11. Easter Egg: Strategy Author's "Little Thoughts"
Look carefully at the code, you'll find some interesting things:
-
Three MA types: EMA, ZEMA, TRIMA all available
"Can't decide which MA is best? Use all three!"
-
Offset mechanism: Buy below MA, sell above MA
"Buy low, sell high — the oldest trick in the book!"
-
Custom stoploss: Only works in live mode
"Backtest is fantasy, live trading is reality!"
-
Protection mechanisms: Cooldown + LowProfitPairs
"Don't trade too fast, don't lose too much!"
12. Last But Not Least
One-Sentence Review
"Three MAs + offset trading + custom stoploss — suitable for trend followers who like multiple confirmations"
Who Should Use It?
- ✅ Multi-MA believers
- ✅ People wanting to learn offset trading
- ✅ 5-minute timeframe traders
- ✅ People with some quantitative experience
Who Should NOT Use It?
- ❌ MA phobia patients
- ❌ People who want simple strategies
- ❌ People unwilling to backtest and verify
- ❌ Pure newbies
Manual Trader Recommendations
You can reference MultiMA_TSL's approach:
- Observe multiple MA types simultaneously
- Buy when price below MA×offset
- Use trailing stop to protect profits
- Set cooldown between trades
But doing it manually is too tiring, let the robot handle it 🤖
13. ⚠️ Risk Reminder Again (Must Read This Section)
Backtest Is Beautiful, Live Trading Needs Caution
MultiMA_TSL's historical backtest performance may look very good — but there's a trap:
Custom stoploss only works in live mode, so backtest results may be overly optimistic.
Simply put: "Backtest is a dream, live trading is reality."
Hidden Risks of Complex Strategies
In live trading, multi-MA combinations may cause:
- Signal conflicts: Different MAs may give different signals
- Parameter sensitivity: Change one parameter, entire strategy behavior changes
- Computation load: Multiple MAs increase computation, may miss timing
- Backtest/live gap: Custom stoploss differences cause behavior gaps
No Trend Filter Pit
MultiMA_TSL's biggest problem: Doesn't care how the broad market moves.
When Bitcoin crashes, it may still be buying — this is called "counter-trend bottom-fishing", easy to get buried.
Recommend adding a BTC trend filter yourself:
- Don't buy when BTC is below key MA
- Or use other broad market indicators to judge
My Recommendations (Real Talk)
1. Backtest with default parameters first, see strategy behavior
2. Add a BTC trend filter yourself, don't go against the trend
3. Dry-run test at least 1-2 weeks, confirm custom stoploss works
4. Small capital live test, don't exceed 10% of total capital
5. Regularly check strategy performance, stop timely if not working
Remember: No matter how smart the strategy, it can't fight black swans. Light position test, staying alive is most important! 🙏
Final reminder: Multiple MA confirmations are good, but over-relying on historical parameters is dangerous. The market always changes, no strategy makes money forever!