Every week, we sit down and review signal performance. Not just to count wins and losses — but to understand what the data is telling us. This post walks you through the exact framework we use. Apply it to your own trading week, whatever system you're running.
---
What We Track Each Week
Raw P&L is the least useful number on our review sheet. Here's what we actually care about.
R-Multiples
We measure every signal outcome in R — multiples of the initial risk. A 2R win means the trade returned twice the amount risked. This normalises results across different position sizes and instruments. It's the only fair way to compare a scalp on EUR/USD with a swing trade on crude oil.
Win Rate by Mode
We segment signals by type — scalp, intraday, swing. Win rate across all three combined is almost meaningless. A system can have a 40% win rate on scalps and 65% on swings. Blending them hides that completely.
Recovery Tier Status
We track rolling drawdown across three tiers. Tier 1 is normal variance. Tier 2 triggers a review. Tier 3 triggers a full system audit. Knowing which tier you're in helps you size positions appropriately — not emotionally.
Pair-Level Performance
Some pairs behave differently week to week. We track performance per instrument, not just in aggregate. If EUR/GBP underperforms consistently across four weeks, that's a signal worth investigating. One week of poor results is noise. Four weeks is a pattern.
---
Framework for Evaluating Your Own Week
Here's how to review your trading week the way we review ours.
Step 1 — Convert everything to R
Take every trade. Divide the outcome by your initial stop-loss distance (your 1R). Now you have a clean, comparable dataset.
Step 2 — Separate by mode
Don't mix your scalps with your swing trades. They operate in different statistical environments. Evaluate each mode on its own terms.
Step 3 — Look at drawdown, not just P&L
You might end the week flat. But if you were down 6% on Wednesday and clawed it back, that's a risk management story worth examining. Maximum intraweek drawdown tells you more than your Friday closing balance.
Step 4 — Check your execution against the signal
This is where most traders lose edge. Did you take the signal as intended? Did you move stops early? Did you skip three trades and then overtrade to compensate? Execution variance is often bigger than system variance.
---
Common Patterns We See
After reviewing hundreds of weekly datasets, certain patterns repeat.
Session-based divergence is consistent. London open tends to produce cleaner breakout conditions. The New York-London overlap is typically the highest-volume window. Asian session signals on major pairs often show lower average R outcomes. This isn't universal — but it's a pattern worth tracking.
Pair clustering happens regularly. When EUR/USD has a difficult week, EUR/GBP and EUR/CHF often share similar characteristics. Correlated pairs under the same macro pressure will behave similarly. Trading five correlated pairs isn't five independent opportunities — it's one bet, five times.
Scalp vs swing divergence is common in trending weeks. Scalps get chopped. Swings capture the move. The reverse is true in ranging markets. Identifying market regime — even loosely — helps you weight your signal types appropriately.
---
When a Bad Week Isn't a Bad System
This is the most important concept in systematic trading. One week is roughly 20–40 signals, depending on your approach. That's a tiny sample size. Statistical edge requires hundreds of trades to express itself reliably.
A losing week with good process is often more valuable than a winning week with poor discipline. If your entries were valid, your sizing was correct, and you followed your rules — the outcome is almost secondary. Markets have variance. Your job is to control what you can control.
We never make system changes after a single bad week. That's how good systems get dismantled prematurely.
---
Frequently Asked Questions
How many trades do you need before results are meaningful?
Most researchers suggest a minimum of 100 trades per setup type before drawing conclusions. Even then, forward-testing matters more than backtesting alone.
Should I review results daily or weekly?
Weekly is better for most traders. Daily reviews encourage emotional over-reaction to normal variance.
What's a healthy weekly win rate?
It depends entirely on your average R per trade. A 40% win rate with 2.5R average winners is a strong system. Win rate alone tells you almost nothing.
---
See It In Action
Want to see how we apply this framework to live signals? Join our free Telegram channel and watch how we communicate signal logic, entry rationale, and R-targets in real time. No delay between free and VIP on signal content — only on delivery timing.
Explore Trade Forge signals → tradingforge.io
---
Trade Forge provides algorithmic trading signals for educational and informational purposes only. Nothing published here constitutes financial advice, investment advice, or a recommendation to buy or sell any financial instrument. Trading involves significant risk of loss. Past signal performance is not indicative of future results. You should seek independent financial advice before making any trading decisions. Trade Forge operates in accordance with FCA and ASIC regulatory standards.