Advertisement IC Markets

Plotting win rate against reward to risk for each setup, and what the picture showed

Document your personal trading journey. Track daily equity curves, review winning and losing streaks, share trade screenshots, and get constructive feedback.
FXS HongKong Scalper
Posts: 543
Joined: Sat Oct 24, 2026 10:15 pm

Re: Plotting win rate against reward to risk for each setup, and what the picture showed

Post by FXS HongKong Scalper »

Agreed, and the same logic is why we stopped ranking setups by the gap to the curve. What also matters is how many trades each point is built on. A setup at 35% and 2.2R with only 40 trades has a standard error on the win rate of about 7.5 points, which is twice the 3.75 point cushion, so it may not be above break even at all.

A practical version: plot each setup with a horizontal error bar on the win rate (roughly square root of p times 1 minus p over n). Setups whose bar crosses the curve go on watch, not into the live plan. It makes the chart less exciting but much more honest.
FXS Official Team | Hong Kong Session Desk
Recommended broker for automated trading & scalping IC Markets
Fairman
Posts: 6955
Joined: Tue Jul 21, 2026 7:11 am
Location: Abuja

Re: Plotting win rate against reward to risk for each setup, and what the picture showed

Post by Fairman »

The part of this that jumped out at me is the setup that "felt good" but sat on the line. That's me. I had a pullback entry with a 72 percent hit rate that felt like a machine, and when I plotted it, average win was 0.6R against 1R loss. Break-even at 0.6R reward needs about 62.5 percent, so 72 percent leaves around 0.15R of edge before costs. With 1.2 pips of spread on a 6 pip target, cost eats 0.2R, and the edge goes negative.

A setup that wins only 38 percent with 2.8R winners sits at about +0.45R expectancy. It hurts to trade because you lose six of ten, but it's the one that paid the bills.

Doing this per quarter makes sense, as long as each point has enough trades. I wouldn't trust a dot built on fewer than 60 or so, and I'd draw a rough error bar so I know which dots might be noise.

Do you include costs in R before plotting, or plot gross and show the cost gap separately? And have you found setups that move across the line from one quarter to the next?
It’s Fairman :geek:
Post Reply