Every week, we run through the same structured review process. Not to celebrate wins or explain away losses — but to understand what the data is actually telling us.
This post walks you through exactly how we do it. Use it as a template for your own weekly review.
---
What We Track Each Week
Raw pip counts feel satisfying. They're also largely meaningless without context.
Here's what we actually measure.
R-multiples. Every signal is sized against an initial risk amount — one R. A trade that returns 2R made twice the amount risked. A trade that loses 0.5R stopped out early. This normalises performance across different pairs, timeframes, and volatility environments.
Win rate by signal mode. We separate scalp signals from swing signals. They behave differently. A scalp strategy with a 65% win rate and a 1:1 reward-to-risk profile isn't the same as a swing strategy with a 40% win rate and a 1:3 profile. Both can be profitable. Mixing them in one win rate figure hides useful information.
Recovery tier status. We track cumulative drawdown thresholds. If a signal stream moves into a defined drawdown tier, we flag it. This isn't panic — it's governance. It tells us whether to reduce position sizing or hold steady.
Pair-level and asset-level performance. Some pairs trend. Some range. Some become erratic around macro events. We break performance down by individual instrument, not just by asset class. A strong week in crypto doesn't mask a rough week in commodities indices — both get reviewed separately.
---
Framework for Evaluating Your Own Week
If you're reviewing your own trading week, start here.
Use R, not pips. Ask: How many R did I make or lose this week? A 200-pip gain on a 50-pip stop is 4R. A 200-pip gain on a 400-pip stop is 0.5R. The same "win" can mean completely different things.
Separate drawdown from P&L. You can end a week flat and still have taken significant drawdown mid-week. That matters. Maximum intraweek drawdown tells you whether your position sizing is sustainable — not just whether you ended in green.
Log the process, not just the outcome. Did your entries follow your rules? Did you move stops early? Did you hold trades through your planned exit? Outcome and process can diverge. A rule-breaking win is not a good week. A rule-following loss is not a bad week.
Calculate expectancy. Multiply your average win (in R) by your win rate, then subtract the average loss (in R) multiplied by your loss rate. Positive expectancy over a meaningful sample is the goal. One week doesn't tell you your expectancy — but tracking it weekly builds the data set that eventually does.
---
Common Patterns We See
Across hundreds of weekly reviews, a few patterns repeat.
Session-based performance divergence. London open and New York overlap sessions tend to produce cleaner signal follow-through. Asian session signals — particularly on forex and certain commodities — often see more noise and false breakouts. This isn't always true, but it's worth tracking in your own data.
Pair clustering around events. When macro data drops — think CPI, NFP, central bank decisions — correlated pairs move together. A bad week across multiple USD pairs might not be four independent losing signals. It might be one macro event, expressed four times. Understanding correlation stops you from over-weighting a single event.
Scalp vs swing divergence in ranging markets. When markets consolidate, scalp signals often perform better. Swing signals need trend — they suffer in flat conditions. If your swing results are weak, check the broader trend structure before questioning the system.
---
When a Bad Week Isn't a Bad System
This is probably the most important section.
A single week is 5 trading days. Depending on signal frequency, that might be 8–15 signals. That's not a statistically significant sample. Not even close.
Professional strategy evaluation typically requires 100+ trades before expectancy estimates become reliable. One losing week inside a positive-expectancy system is normal. Expected, even.
The question isn't "did we win this week?" The question is: Is the system behaving within expected variance?
Two consecutive losing weeks? Still within normal variance for most strategies. Six weeks of consistent underperformance across all asset classes, all sessions, all signal modes? Now you start investigating.
Discipline means not changing your process based on noise.
---
FAQ
What's a good weekly R result?
It varies by strategy, but a target range of +1R to +3R per week is reasonable for most systematic approaches. The consistency of that range over 20+ weeks matters more than any single number.
How do I track R-multiples if I'm using a free signal service?
You apply the R framework yourself. Treat each signal's stop-loss distance as 1R. Calculate your outcome from there. The math works regardless of how signals are delivered.
How often should I reassess a strategy based on performance?
Review weekly. Reassess at 50-trade and 100-trade milestones. Don't make structural changes based on fewer than 30 signals.
---
Want to see our signals in real time? Our free Telegram channel delivers the same signals as VIP — with a short delay. It's a great way to follow along, build your data set, and apply these frameworks to live signal flow.
👉 Join the free channel at tradingforge.io
---
This content is educational only and does not constitute financial advice. Trading involves significant risk of loss. Past performance is not indicative of future results. Trade Forge operates to FCA and ASIC-equivalent compliance standards. Always conduct your own research before making any trading decisions.