Creating Actionable Signals
A good trading signal is actionable. It tells the trader:
- Where to enter
- Where to place the stop loss
- Where to take profits
Always include Entry, Stop Loss, and at least one Take Profit in your signals.
Signal Structure
Complete BUY Signal
{
"points": [
{
"time": 1765540800,
"type": "low",
"price": 1.1725,
"label": "BUY",
"color": "#3b82f6",
"shape": "arrowUp",
"size": 2
},
{
"time": 1765540800,
"type": "low",
"price": 1.1695,
"label": "SL",
"color": "#ef4444",
"shape": "square",
"size": 1
},
{
"time": 1765540800,
"type": "high",
"price": 1.1755,
"label": "TP1",
"color": "#22c55e",
"shape": "circle",
"size": 1
},
{
"time": 1765540800,
"type": "high",
"price": 1.1785,
"label": "TP2",
"color": "#22c55e",
"shape": "circle",
"size": 1
},
{
"time": 1765540800,
"type": "high",
"price": 1.1815,
"label": "TP3",
"color": "#22c55e",
"shape": "circle",
"size": 1
}
],
"metadata": {
"signal_type": "BUY",
"entry": 1.1725,
"sl": 1.1695,
"tp1": 1.1755,
"tp2": 1.1785,
"tp3": 1.1815,
"risk_pips": 30,
"risk_reward": "1:3"
}
}
Risk Management
Calculate Risk/Reward
def calculate_risk_reward(entry, stop_loss, take_profit):
"""Calculate risk/reward ratio."""
risk = abs(entry - stop_loss)
reward = abs(take_profit - entry)
return reward / risk if risk > 0 else 0
# Example
entry = 1.1725
sl = 1.1695
tp = 1.1815
rr = calculate_risk_reward(entry, sl, tp)
print(f"Risk/Reward: 1:{rr:.1f}") # Output: Risk/Reward: 1:3.0
Minimum R:R Requirements
| Signal Quality | Minimum R:R |
|---|
| Standard | 1:1 |
| Good | 1:2 |
| Excellent | 1:3+ |
Signals with R:R below 1:1 are generally not worth taking. Consider filtering these out.
Visual Hierarchy
Size Recommendations
| Element | Size | Why |
|---|
| Entry | 2 | Most important - catches attention |
| Stop Loss | 1 | Secondary - always visible |
| Take Profits | 1 | Secondary - shows targets |
Color Consistency
Always use consistent colors:
| Element | Color | Hex |
|---|
| BUY Entry | Blue | #3b82f6 |
| SELL Entry | Orange | #f97316 |
| Stop Loss | Red | #ef4444 |
| Take Profit | Green | #22c55e |
| Neutral/Info | Yellow | #eab308 |
Label Best Practices
- Use short labels:
BUY, SELL, TP1, SL
- Be consistent across signals
- Include level numbers for TPs:
TP1, TP2, TP3
DON’T
- Avoid long labels:
Buy Entry Here at Support
- Don’t use special characters that may not render
- Don’t mix naming conventions
Time Alignment
Always use the time value from the /bars endpoint. Never calculate timestamps manually.
# CORRECT
bars = get_bars("EURUSD", 60)
signal_time = bars.iloc[-1]["time"] # Use time from API
# WRONG
signal_time = int(datetime.now().timestamp()) # May not align with bars
Multiple Signals on Same Bar
You can place multiple points on the same bar by using the same time value:
points = [
{"time": bar_time, "type": "low", "price": entry, "label": "BUY", ...},
{"time": bar_time, "type": "low", "price": sl, "label": "SL", ...},
{"time": bar_time, "type": "high", "price": tp1, "label": "TP1", ...},
]
Signal Spacing
For visual clarity, add small offsets to prices:
def create_buy_signal(bar, entry_price, sl_price, tp_prices):
"""Create a BUY signal with proper spacing."""
points = []
# Entry - slightly below the low
points.append({
"time": int(bar["time"]),
"type": "low",
"price": entry_price,
"label": "BUY",
"color": "#3b82f6",
"shape": "arrowUp",
"size": 2
})
# SL - with offset below entry
points.append({
"time": int(bar["time"]),
"type": "low",
"price": sl_price,
"label": "SL",
"color": "#ef4444",
"shape": "square",
"size": 1
})
# TPs - with offsets above entry
for i, tp in enumerate(tp_prices, 1):
points.append({
"time": int(bar["time"]),
"type": "high",
"price": tp,
"label": f"TP{i}",
"color": "#22c55e",
"shape": "circle",
"size": 1
})
return points
metadata = {
# Signal details
"signal_type": "BUY",
"entry_price": 1.1725,
"stop_loss": 1.1695,
"take_profits": [1.1755, 1.1785, 1.1815],
# Risk metrics
"risk_pips": 30,
"reward_pips": 90,
"risk_reward": "1:3",
# Strategy info
"strategy": "ICT Order Block",
"timeframe": "H1",
"confluence_factors": ["OB", "FVG", "HTF Trend"],
# Confidence
"confidence": "high", # low, medium, high
# Timestamp
"generated_at": datetime.utcnow().isoformat()
}
Filter Low-Quality Signals
Before submitting, filter out weak signals:
def is_quality_signal(entry, sl, tp, min_rr=1.5):
"""Check if signal meets quality criteria."""
rr = calculate_risk_reward(entry, sl, tp)
return rr >= min_rr
# Filter signals
quality_signals = [
s for s in all_signals
if is_quality_signal(s["entry"], s["sl"], s["tp"])
]
Update Frequency
| Timeframe | Recommended Update |
|---|
| M1 | Every 1 minute |
| M5 | Every 5 minutes |
| M15 | Every 15 minutes |
| H1 | Every hour |
| H4 | Every 4 hours |
| D1 | Once per day |
Don’t update too frequently - it wastes API calls and can cause visual flickering.
Error Handling
Always handle API errors gracefully:
def safe_submit(points, max_retries=3):
"""Submit with retry logic."""
for attempt in range(max_retries):
try:
return submit_indicator(points)
except requests.exceptions.HTTPError as e:
if e.response.status_code == 429:
# Rate limited - wait and retry
wait = 60 * (attempt + 1)
print(f"Rate limited. Waiting {wait}s...")
time.sleep(wait)
elif e.response.status_code >= 500:
# Server error - wait and retry
time.sleep(5 * (attempt + 1))
else:
raise
raise Exception("Max retries exceeded")
Checklist
Before deploying your indicator:
Verify time alignment
Signals appear on the correct candles
Check visual hierarchy
Entry is prominent, TPs/SL are secondary
Validate R:R ratios
All signals meet minimum R:R requirements
Test on multiple timeframes
Indicator works on H1, H4, D1
Add error handling
Gracefully handle API errors
Include metadata
Track signal performance
Next Steps
