"Never take a trade under 3:1 risk-reward." You've heard this advice. It sounds smart. It's also incomplete.
Risk-reward ratio matters, but not in the simplistic way most traders think. The real math involves probability, and ignoring that leads to worse trading decisions.
Here's how risk-reward actually works.
The Basic Math
Risk-reward ratio compares what you might lose to what you might gain.
The formula: Risk-Reward Ratio = Potential Loss / Potential Gain
Example:
- Entry: $100
- Stop loss: $95 (risking $5)
- Target: $115 (potential gain $15)
- Risk-Reward: 1:3 (risking $1 to make $3)
On paper, this looks good. You can be wrong twice and right once and still profit. But this ignores something critical: how often does the trade actually reach the target?
The Missing Variable: Win Rate
Risk-reward means nothing without probability. A 1:10 risk-reward trade that only wins 5% of the time is terrible. A 1:1 trade that wins 70% of the time is excellent.
Expectancy formula: Expectancy = (Win Rate × Average Win) - (Loss Rate × Average Loss)
Let's compare two traders:
Trader A (high R:R, low win rate):
- Risk-reward: 1:3
- Win rate: 30%
- Expectancy: (0.30 × $3) - (0.70 × $1) = $0.90 - $0.70 = +$0.20 per $1 risked
Trader B (low R:R, high win rate):
- Risk-reward: 1:1
- Win rate: 60%
- Expectancy: (0.60 × $1) - (0.40 × $1) = $0.60 - $0.40 = +$0.20 per $1 risked
Same expectancy. Different approaches. Neither is inherently superior - they're mathematically equivalent.
The Break-Even Point
For any risk-reward ratio, there's a minimum win rate needed to break even.
Break-even win rate = 1 / (1 + Reward/Risk)
- 1:1 R:R requires 50% win rate to break even
- 1:2 R:R requires 33% win rate to break even
- 1:3 R:R requires 25% win rate to break even
- 1:5 R:R requires 17% win rate to break even
This is why high R:R isn't automatically better. Yes, you can be wrong more often. But will you be? The further your target, the lower the probability of reaching it.
The R:R Trap
Here's how traders misuse risk-reward:
Trap 1: Artificial targets
You want 3:1 R:R, so you place your target 3× your stop distance from entry. But the target is in no-man's land - no structure, no reason for price to reach it. You've created the illusion of good risk-reward without actual edge.
Trap 2: Ignoring probability
A target at major resistance has lower probability than a target at minor structure. Stretching for better R:R often means worse probability - and worse expectancy.
Trap 3: One-size-fits-all
Different setups have different natural targets. Breakout trades might run 5:1. Mean reversion trades might only offer 1:1. Forcing uniform R:R across all trades ignores what the market is actually offering.
Structure-Based Targets
Better approach: let market structure determine your targets.
Natural target locations:
- Prior swing highs/lows
- Major support/resistance levels
- Volume profile nodes
- Measured move projections
- Fibonacci extensions (if you use them)
Calculate R:R after identifying these levels, not before. If the natural target only offers 1.5:1, that's the trade's actual risk-reward - not some arbitrary multiple of your stop.
Then ask: given this R:R and my estimated probability, is expectancy positive?
Partial Profits and R:R
Many traders take partial profits, which complicates R:R calculation.
Example:
- Risk: $100
- Take 50% off at 2:1 (+$100)
- Move stop to breakeven
- Let remaining 50% run to 4:1 (+$200) or get stopped at breakeven ($0)
If the 4:1 target hits 40% of the time after the first target is reached:
- Average winner: $100 + (0.4 × $200) = $180
- Original R:R appears lower, but you've locked in profit while maintaining upside
Partial profits reduce headline R:R but can improve psychological execution and reduce variance.
What Actually Matters
Stop optimizing for risk-reward ratio. Optimize for expectancy.
Good trades have:
- Positive expectancy (R:R × win rate produces profit)
- Structure-based stops (invalidation makes sense)
- Structure-based targets (reason for price to reach them)
- Sufficient R:R for your win rate (math works)
The minimum R:R you should accept depends on your win rate:
- 60% win rate: minimum 1:1
- 50% win rate: minimum 1.5:1
- 40% win rate: minimum 2:1
- 30% win rate: minimum 3:1
Know your historical win rate. Use it to filter trades.
The Bottom Line
Risk-reward ratio is half the equation. Win rate is the other half. Together they create expectancy - the only number that determines long-term profitability.
Don't force artificial R:R targets. Don't ignore probability. Don't treat all trades the same.
Let structure define your stops and targets. Calculate R:R from reality, not fantasy. Then verify the math works for your actual win rate.
Confluence scoring can help estimate probability. When cycle phase, volume regime, and multi-timeframe alignment all agree, probability is higher - and lower R:R becomes acceptable. When signals conflict, higher R:R is needed to compensate for lower probability. The math always governs.
Augury Grid's 5-point scoring system quantifies setup quality, giving you data to estimate probability on each trade. Higher confluence scores correlate with higher win rates, allowing you to accept lower R:R on the best setups while demanding higher R:R on marginal ones.
See confluence scoring →