Strategy Best Practices

To achieve a high match rate and precise slippage data, your Pine Script strategy should be optimized for real-world signal generation.

Supported Trade Types

Important Limitation

Pyramid trading is not currently supported by Algo Trade Analytics.

Our matching algorithm is designed for simple entry/exit trade sequences:

Supported	Trade Flow
✅	Enter Long → Exit Long (buy to open, sell to close)
✅	Enter Short → Exit Short (sell to open, buy to close)
❌	Pyramid entries (adding to winning positions)
❌	Scaling into positions over multiple orders
❌	Partial position exits

What is pyramid trading? Pyramid trading is a strategy where a trader incrementally increases the size of a position as the market moves in their favor. For example, buying 10 shares, then adding 10 more when price rises, then adding 10 more again.

Why isn’t it supported? Our reconciliation engine expects a 1:1 relationship between entry signals and exit signals. Multiple entries into the same position cannot be accurately matched against backtest data or broker fills at this time.

How to configure your strategy:

strategy("My Strategy", pyramiding=1)  // ← Set pyramiding to 1

Critical Settings

Add these parameters to your strategy() declaration to align historical backtesting with real-time alerts.

1. calc_on_every_tick=true

By default, strategies only calculate at the close of a bar.

• Why: Live alerts fire intrabar. If this is false, your backtest will assume perfect entry at the close, while your real signal might fire mid-candle, creating artificial “slippage.”
• Action: Always set to true for strategies you intend to audit.

2. process_orders_on_close=false

• Why: Setting this to true forces backtest entries to the next bar open, which rarely matches real-time alert behavior.
• Action: Leave at false (default) to ensure signals are recorded the moment conditions are met.

3. calc_on_order_fills=false

• Why: This setting forces the strategy to recalculate logic immediately after an order is filled intrabar. While precise, it can cause “repainting” where a signal appears and disappears within the same bar in backtesting, but fires permanently in real-time.
• Action: Generally keep this false unless your strategy specifically relies on intrabar pyramiding.

High-Fidelity Signal Integration

When building your alert_message, use the positionDirection field and include complete metadata for the highest reconciliation accuracy. Basic & Pro

Recommended Entry Template

longMessage = '{"auth_key":"YOUR_KEY","ticker":"' + syminfo.ticker + '","positionDirection":"entryLong","orderType":"market","marketPrice":' + str.tostring(close) + ',"qty":10,"metadata":{"timestamp":"' + str.tostring(time) + '","strategy":"My Strategy","signal":"Entry Signal"}}'

if longCondition
    strategy.entry("Long", strategy.long, alert_message=longMessage)

Important Fields

• positionDirection: Use entryLong, exitLong, entryShort, or exitShort for explicit trade intent.
• marketPrice: Always use str.tostring(close) to capture the exact price at the moment of the signal. This is the baseline for slippage calculation.
• metadata.timestamp: Use str.tostring(time) to help the AI reconcile latency gaps.
• metadata.strategy: Include your strategy name for grouping and reporting.
• metadata.signal: Describe the specific signal type for detailed analytics.

Testing for “Alert Stability”

Before going live, verify your strategy’s Logic Stability without financial risk: All Plans

1. Run the strategy on a 1-minute chart in TradingView.
2. Set a webhook to the Hub.
3. Let it run for 24 hours.
4. Export the Strategy Tester CSV and compare it to your captured events in Backtest vs Alerts.
5. A high match rate indicates stable logic. Low match rates suggest logical issues (repainting or lookahead) that must be fixed.