# Monitoring Positions, P&L and Risk

> Running algorithms need constant monitoring. Learn the key health, trading and risk metrics to track, how to design useful alerts and how to avoid alert fatigue.

Source: https://learn.tradelabsai.com/algo-trading/live-monitoring/  
Track: Algorithmic Trading · Level: Advanced · Updated: 2026-10-03  
Publisher: TradeLabs AI (https://tradelabsai.com). Education, not financial advice.  
Cite as: TradeLabs Learn, "Monitoring Positions, P&L and Risk", https://learn.tradelabsai.com/algo-trading/live-monitoring/

An automated strategy that nobody watches is a risk waiting to happen. Live monitoring means tracking the health of the system, the behaviour of the strategy and the risk it is taking, in real time, with alerts that reach a human when something looks wrong. Good monitoring answers three questions at a glance: is the system running and connected, is it doing what it should, and is the risk within limits? The technical tooling is covered in [Monitoring and Logging Systems](https://learn.tradelabsai.com/infrastructure/monitoring-and-logging-systems/); this lesson focuses on what to watch and why.

## Three layers to monitor

| Layer | Question | Example metrics |
|---|---|---|
| System health | Is it running and connected? | Process alive, heartbeat, CPU and memory, connection status, data latency |
| Strategy behaviour | Is it doing what it should? | Signals generated, orders sent, fill rate, rejects, trades vs expected |
| Risk and P&L | Is the risk acceptable? | Positions, exposure, daily P&L, drawdown, limit usage |

## Key metrics

| Metric | Why it matters |
|---|---|
| Market data freshness | Stale data leads to bad decisions. Alert if no update in a set number of seconds |
| Heartbeat | A regular "I am alive" signal; silence means a crash or hang |
| Order reject rate | Rising rejects suggest connection, permission or logic problems |
| Fill rate and slippage | Execution quality versus expectations. See [Slippage Analysis](https://learn.tradelabsai.com/orders/slippage-analysis/) |
| Position vs expected | Detects missed fills or duplicate orders |
| Broker reconciliation | Internal positions must match the broker's. See [Trade Accounting and Reconciliation](https://learn.tradelabsai.com/industry/trade-reconciliation/) |
| Daily P&L and drawdown | Compare against normal ranges and limits |
| Limit utilisation | How close each risk limit is to triggering |

## Designing good alerts

| Principle | Explanation |
|---|---|
| Actionable | Every alert should require a decision or action |
| Tiered | Info, warning and critical levels with different channels |
| Specific | Say what broke, where and the current value |
| Rate limited | One alert per problem, not hundreds of repeats |
| Routed | Critical alerts reach a phone, not only an email inbox |

**Example: A silent data feed**
At 10:42, a strategy's market data feed for one symbol stops updating, but the connection stays open, so no error appears. The strategy keeps seeing the last price, $50.10, while the real price falls to $48.70. A freshness monitor set to alert when a symbol has not updated for 30 seconds during market hours fires at 10:43. The trader pauses the strategy, finds that the subscription dropped and resubscribes. Without the freshness check, the strategy might have bought at stale prices, believing the market was steady.

## Alert fatigue

When alerts fire too often, people start ignoring them, including the important ones. Reduce noise by:

- **Removing alerts** that never lead to action.
- **Raising thresholds** that trigger on normal variation.
- **Grouping related alerts** into one message.
- **Reviewing alert history** weekly and tuning.

## Dashboards

A simple dashboard showing system status, positions, P&L, open orders and recent errors lets a human check everything in seconds. Keep it focused: if everything is highlighted, nothing is.

## Daily routines

| When | Check |
|---|---|
| Before the open | Systems connected, positions reconciled, limits set, data flowing |
| During the session | Alerts, P&L against expectations, unusual activity |
| After the close | Reconcile positions and fills, review errors and slippage, archive logs. See [Post-Trade Analysis](https://learn.tradelabsai.com/start-here/post-trade-analysis/) |

## Frequently asked questions

### What should I monitor on a trading bot?

System health (heartbeat, connections, data freshness), strategy behaviour (orders, fills, rejects) and risk (positions, P&L, drawdown and limit usage).

### How do I get alerts from a trading bot?

Send them through messaging services such as email, SMS, Telegram or chat webhooks, with critical alerts routed to a phone. See [Alerts and Webhooks](https://learn.tradelabsai.com/programming/alerts-and-webhooks/).

### What is a heartbeat in trading systems?

A regular signal a process sends to show it is alive; if it stops, the monitoring system raises an alert.

Next, learn how to make systems fail safely in [Alerts, Error Handling and Reconnection](https://learn.tradelabsai.com/algo-trading/error-handling/).

## Continue learning

- Next lesson: [Alerts, Error Handling and Reconnection](https://learn.tradelabsai.com/algo-trading/error-handling/)
- Previous lesson: [From Backtest to Live: Paper, Shadow and Canary](https://learn.tradelabsai.com/algo-trading/backtest-to-live/)
- Related: [From Backtest to Live: Paper, Shadow and Canary](https://learn.tradelabsai.com/algo-trading/backtest-to-live/): Why live results almost always trail backtests, how to measure the gap, and a staged plan for taking a strategy from backtest to paper trading to real money.
- Related: [Monitoring and Logging Systems](https://learn.tradelabsai.com/infrastructure/monitoring-and-logging-systems/): The tools behind trading observability: structured logs, metrics, dashboards, tracing and alerting with Prometheus, Grafana and log stacks, and how to set them up.
- Related: [Risk Controls and Kill Switches](https://learn.tradelabsai.com/algo-trading/risk-controls-and-kill-switches/): Pre trade risk checks and kill switches stop a trading algorithm before a bug becomes a disaster. Learn the essential limits, how to layer them and how to test them.
- Related: [Alerts, Error Handling and Reconnection](https://learn.tradelabsai.com/algo-trading/error-handling/): Trading systems face rejected orders, disconnects, bad data and partial fills. Learn how to classify errors, retry safely, use idempotent orders and fail closed.
- Related: [Logging, Audit Trails and Incident Response](https://learn.tradelabsai.com/algo-trading/audit-trails/): An audit trail records every signal, order, change and fill so trading can be reconstructed later. Learn what to log, the regulatory rules and how it helps traders.
- Related: [Trade Accounting and Reconciliation](https://learn.tradelabsai.com/industry/trade-reconciliation/): Reconciliation checks that internal records of trades, positions and cash match brokers, custodians and clearing houses. Learn the process, common breaks and fixes.