1. Bring in trades
Use manual entry, CSV import, platform webhooks, MetaTrader exporters, crypto exchange sync, or API integrations. The first goal is complete closed-trade history.
A complete operating manual for onboarding, trade history, institutional analytics, Prop Firm Guardian, Pro API access, and the most common fixes. Search, deep-link, and resolve issues without waiting for support.
Try a broader term such as "broker", "CSV", "Guardian", "API", "billing", or "Monte Carlo".
VantageGrid works best when you use it as a closed improvement loop: connect data, review trades, tag behavior, study risk, update playbooks, and enforce guardrails.
Use manual entry, CSV import, platform webhooks, MetaTrader exporters, crypto exchange sync, or API integrations. The first goal is complete closed-trade history.
Tag setup, mistakes, emotion, confluence, session, screenshots, and notes. Context turns a trade log into behavioral intelligence.
Use Institutional Lab, Prop Firm Guardian, AI Coach, Calendar, Replay, and Playbooks to turn data into rules for the next session.
Start with the connection type that matches your trading workflow. If a direct connector is not available for your broker, CSV import and webhook/API routes still let you maintain a reliable journal.
For OANDA, verify timezone and fee handling carefully. A shifted timestamp can distort session analysis, daily P&L, and Calendar heatmaps.
VantageGrid stores exchange credentials encrypted at rest. You should still rotate any exchange key that was pasted into the wrong tool or shared externally.
CSV mapping and templates
API keys and live sync
Validate imported history
Interface map: first connect data, then validate trades, then add behavioral context.
| Method | Best for | What to verify |
|---|---|---|
| Manual entry | Small accounts, backfilled examples, missed trades, screenshots, and discretionary review. | Entry, exit, side, quantity, fees, account name, and close timestamp. |
| CSV import | Historical account history from brokers, prop dashboards, OANDA, TradeStation, NinjaTrader exports, and generic broker files. | Column mapping, timezone, duplicate handling, realized P&L, commissions, and account labels. |
| Webhook or platform bridge | Live journaling from NinjaTrader, MetaTrader, TradingView, TradeLocker, DXtrade, and automation tools. | API key, endpoint path, order IDs, heartbeat status, and whether partial fills are matched correctly. |
| Pro API | External AI agents, execution bots, risk pre-checks, and custom operating workflows. | API key security, Pro subscription status, rate limits, and idempotent client behavior. |
After any bulk import, open Reports and Calendar. If the equity curve, daily P&L, and trade count look wrong, fix mapping before adding more data.
Tags are the bridge between what you felt, what you planned, and what happened financially.
Use stable names such as "London Breakout", "VWAP Reclaim", or "Liquidity Sweep". Do not rename the same setup every week or analytics will split the sample.
Mark events such as FOMO, revenge, late entry, early exit, oversized risk, moved stop, no plan, or ignored news. These feed the Cost of Discipline and Psychology Heatmaps.
Also tag Patience, Following Plan, Correct No-Trade, Reduced Size, and Stopped After Rule Hit. The platform should learn your best behavior, not only your mistakes.
Minimum useful habit: tag every closed trade with one setup tag and one process tag within 24 hours. That is enough to activate most behavioral analytics.
Use the top status area for market session, sync state, trade count, Guardian status, and account context. It is designed for fast scanning before and during a trading session.
Win rate, profit factor, expectancy, P&L, drawdown, and quality score show whether the edge is healthy. Use them together, not in isolation.
Review mistake costs, live session coach, circuit breaker, mission plan, and trade permission before taking new risk.
Use Dashboard for monitoring, Trades for trade-level review, Analytics for deeper analysis, Playbooks for rules, and AI Coach for synthesis.
The Institutional Lab turns trade history into risk-adjusted, behavior-aware analytics. It is not just "how much did I make?" It is "was the process durable enough to scale?"
Sortino, drawdown, risk of ruin, tail behavior, and recovery factor show whether returns are stable enough to trust.
Setup Intelligence, Strategy Lab, session windows, and market-regime analysis reveal where the edge actually appears.
Psychology Heatmaps, mistake leaks, quality grades, and daily journal context reveal whether the trader is executing the plan.
Sortino is a risk-adjusted return metric that focuses on downside volatility instead of all volatility. A strategy with choppy upside and controlled downside can look better on Sortino than on Sharpe.
Sortino = excess return / downside deviation
Kelly estimates the theoretical growth-optimal fraction of capital to risk when win probability and payoff ratio are known. In practice, traders usually use fractional Kelly because full Kelly can produce severe drawdowns.
Kelly = W - ((1 - W) / R)
Risk-of-ruin estimates the chance that normal variance pushes the account past a failure threshold. In VantageGrid this can be modeled through Monte Carlo paths and Guardian drawdown limits.
If P&L is positive but Sortino is poor, Kelly is extreme, or risk-of-ruin is high, the strategy may be profitable but fragile. Reduce size, tighten rules, or collect more data before scaling.
The engine samples from your historical closed-trade returns with replacement, then generates many possible future trade sequences. It asks: if your edge behaves like the past, what account paths are plausible?
This is the percentage of paths where drawdown breaches the configured danger threshold. If it is above your tolerance, reduce risk per trade or tighten stop conditions.
The 50th percentile outcome. It is more useful than the average when extreme outliers distort the mean.
Best case and worst case use percentile bands. Treat the 5th percentile as a serious planning scenario, not an impossible disaster.
Monte Carlo fan chart reading: focus on the lower band and maximum drawdown, not only the best path.
Higher highs, controlled drawdowns, stable trade frequency, and recoveries that do not require oversized wins.
Flat or falling curve while trade count rises, larger losses after winning streaks, or profit concentrated in only one day or one setup.
Equity drops that coincide with tags like FOMO, revenge, no plan, or oversized risk usually point to process failure, not market randomness.
Reduce size, pause weak setups, review Replay, write a Daily Journal debrief, and require Playbook compliance before returning to normal risk.
Psychology Heatmaps group closed trades by tags such as FOMO, Revenge, Patience, Following Plan, Hesitation, or Moved Stop. For each tag, VantageGrid calculates trade count, net P&L, and win rate.
If the radar is empty, add tags to closed trades. The feature needs enough journal data to separate real patterns from noise.
Asset Correlation inspects open trades and estimates net exposure by base and quote asset. For example, long EUR/USD adds EUR exposure and short USD exposure. Multiple positions can quietly create a concentrated USD, JPY, crypto, or index bet.
You are net long that asset across open positions.
You are net short that asset across open positions.
If exposure exceeds a threshold, the Guardian warns that one news event could hit several trades at once.
The correlation matrix highlights instruments that often move together or hedge each other.
FTMO rules often focus on daily loss, maximum loss, and profit target. Topstep and MyFundedFutures-style accounts often require careful trailing drawdown handling. Always enter the exact rules from your current account contract.
Create separate Prop Firm accounts for each challenge or funded account. Do not combine personal and prop data if the rules are different.
Check Guardian status before the session, after each large trade, and near the daily loss threshold. If a rule changes, update the account before the next trade.
Guardian Lock is a risk-control state. When current P&L, drawdown, or configured risk rules approach a dangerous area, VantageGrid can warn you, mark trade permission as unsafe, and return a denial from Pro API Guardian checks.
Guardian Lock is a guardrail, not a broker-side kill switch unless you explicitly wire your own execution system to honor it. If your platform can still place trades, you are still responsible for stopping.
VantageGrid is monitoring and intelligence software. It depends on the data it receives from your broker, platform, bridge, webhook, browser, network, and server. During fast markets or outages, data can be delayed, incomplete, duplicated, or unavailable.
The Pro API lets trusted AI agents and automation clients retrieve compact trading context and validate proposed trades before execution.
Interactive Swagger/OpenAPI documentation is available at /api/docs. This Help Center explains the operational contract in plain English.
Authorization: Bearer vgai_your_api_key_here
X-VantageGrid-API-Key: vgai_your_api_key_here
VantageGrid stores Pro API keys using hash-based lookup and encrypted storage. You still must treat the visible key as a secret because anyone holding it can call your Pro API until it is revoked.
Returns a compact, LLM-ready account context designed for models such as GPT, Claude, Gemini, and custom agents. Use it before asking an agent to coach, summarize, or reason about current trading behavior.
| Field | Description |
|---|---|
| recent_performance | Recent P&L, trade count, win rate, profit factor, expectancy, drawdown, and operating state where available. |
| emotional_patterns | Psychology tag summary such as FOMO, Revenge, Patience, Following Plan, net P&L, win rate, and sample size. |
| guardian_status | Current Prop Firm Guardian and risk permission state. |
| behavioral_bias_summary | A concise interpretation of likely behavioral bias based on tags, recent results, and risk state. |
curl -s https://vantagegrid.pro/api/v1/ai-context \
-H "Authorization: Bearer vgai_your_api_key_here"
Accepts a proposed trade and returns whether the trade is authorized based on the user's current drawdown and Prop Firm Guardian rules. External execution bots should call this before placing risk.
| Request field | Purpose |
|---|---|
| instrument | Symbol such as NQ, ES, EUR/USD, XAU/USD, BTC/USD, US30, or a broker-specific symbol. |
| side | Buy, sell, long, or short. |
| quantity | Proposed size or number of contracts/lots/units. |
| entry_price | Expected entry price if known. |
| stop_loss | Stop price used to estimate potential loss. |
| account_id | Optional account selector. Use the correct prop account when rules differ. |
curl -s https://vantagegrid.pro/api/v1/guardian-validate \
-H "Authorization: Bearer vgai_your_api_key_here" \
-H "Content-Type: application/json" \
-d '{
"instrument": "NQ",
"side": "buy",
"quantity": 1,
"entry_price": 18500.25,
"stop_loss": 18480.25,
"account_id": 42
}'
{
"authorized": true,
"reason": "Trade is within configured Guardian limits.",
"guardian_status": {
"daily_drawdown_remaining": 875.00,
"lock_active": false
}
}
from __future__ import annotations
from dataclasses import dataclass
from typing import Any, Dict, Optional
import requests
class VantageGridError(RuntimeError):
pass
@dataclass
class ProposedTrade:
instrument: str
side: str
quantity: float
entry_price: Optional[float] = None
stop_loss: Optional[float] = None
account_id: Optional[int] = None
class VantageGridClient:
def __init__(self, api_key: str, base_url: str = "https://vantagegrid.pro", timeout: float = 10.0):
self.base_url = base_url.rstrip("/")
self.timeout = timeout
self.session = requests.Session()
self.session.headers.update({
"Authorization": f"Bearer {api_key}",
"Content-Type": "application/json",
"User-Agent": "VantageGrid-AI-Client/1.0",
})
def _request(self, method: str, path: str, **kwargs: Any) -> Dict[str, Any]:
url = f"{self.base_url}{path}"
response = self.session.request(method, url, timeout=self.timeout, **kwargs)
if response.status_code == 401:
raise VantageGridError("API key is missing, invalid, or revoked.")
if response.status_code == 403:
raise VantageGridError("Active Pro subscription required.")
if response.status_code == 429:
raise VantageGridError("Rate limit exceeded. Back off before retrying.")
if response.status_code >= 400:
raise VantageGridError(f"VantageGrid API error {response.status_code}: {response.text[:300]}")
return response.json()
def ai_context(self) -> Dict[str, Any]:
return self._request("GET", "/api/v1/ai-context")
def guardian_validate(self, trade: ProposedTrade) -> Dict[str, Any]:
payload = {
"instrument": trade.instrument,
"side": trade.side,
"quantity": trade.quantity,
"entry_price": trade.entry_price,
"stop_loss": trade.stop_loss,
"account_id": trade.account_id,
}
payload = {key: value for key, value in payload.items() if value is not None}
return self._request("POST", "/api/v1/guardian-validate", json=payload)
if __name__ == "__main__":
client = VantageGridClient(api_key="vgai_your_api_key_here")
context = client.ai_context()
print("Behavioral bias:", context.get("behavioral_bias_summary"))
decision = client.guardian_validate(ProposedTrade(
instrument="NQ",
side="buy",
quantity=1,
entry_price=18500.25,
stop_loss=18480.25,
account_id=42,
))
if decision.get("authorized"):
print("Trade allowed by Guardian.")
else:
print("Trade blocked:", decision.get("reason"))
Execution bots should fail closed. If VantageGrid is unreachable, the safest default is not to place the trade.
Common header fixes: map `Symbol`, `Market`, or `Instrument` to instrument; map `Realized P/L`, `Profit`, or `Net PnL` to P&L; map `Commission` or `Fees` to fees.
If no VantageGrid extension is installed, use the TradingView webhook setup from Account Settings. The webhook flow is the supported fallback when browser UI integrations change.
Check platform bridge status, API key, webhook URL, internet connection, and whether the trading platform is running. If the key was regenerated, update it in the bridge or Expert Advisor.
Review account selection, daily P&L, drawdown remaining, stop distance, quantity, and prop-firm rule configuration. If the proposed trade has no stop loss, an external bot may be denied because downside cannot be estimated safely.
VantageGrid uses essential cookies for login and security. With your permission, functional cookies remember product preferences and analytics helps us improve VantageGrid. No ad tracking. You can change this anytime in Cookie settings. Privacy Policy