Forward-testing on demo; measuring win rate, R-multiple, expectancy

A backtest tells you whether your rules would have worked. A forward test tells you whether you can make them work — in real time, with real spreads and slippage, while the outcome is still unknown. It's the final exam before risking real money, and the stage where most of the gap between "great on paper" and "disappointing in practice" shows up. Done properly, it turns your strategy from an idea into evidence.

What you'll learn

• What forward testing reveals that backtesting can't
• How to set up a demo forward test that mirrors live trading
• How to measure results in R: win rate, average win and loss, expectancy, and drawdown
• How to compare forward-test and backtest results fairly
• Decision criteria for going live — and how to scale up safely

1. What forward testing reveals

Common discoveries: setups happen at inconvenient times, costs are higher than modelled, and — most often — the trader doesn't follow their own rules as precisely as in the backtest.

2. Setting up the forward test

• Freeze the rules. Use one strategy version for the whole test.
• Mirror live conditions. Same broker type, same account currency, and the same risk per trade you'd use live. A $100,000 demo account traded with $100 positions teaches you nothing about real-world sizing.
• Treat it as real. Follow the full trading plan: pre-session routine, checklist, journal, loss limits.
• Set the sample in advance. Aim for at least 50 trades, and ideally a number comparable with your backtest. Decide the size before you start so you're not tempted to stop at a good moment or keep going after a bad one.

3. Measuring in R

For every trade, record the R-multiple:

R-multiple = profit or loss on the trade ÷ amount risked

• Risked $50, made $110 → +2.2R
• Risked $50, lost $50 → −1R
• Risked $50, lost $62 because of slippage → −1.24R

Measuring in R makes results comparable across markets, position sizes, and account sizes — and shows you instantly when slippage or management mistakes make losses bigger than −1R.

The core metrics

4. Comparing forward test and backtest

Forward-test results are usually somewhat worse than backtests. That's expected: real costs, imperfect execution, and the absence of hindsight all take a toll.

Worked example

(Illustrative.)

Interpretation: the win rate held up, but average winners shrank — likely because of costs, slippage, or exiting early. The edge is smaller but still positive, drawdown stayed within the backtest's range, and rule adherence was high. This is a reasonable candidate for a small, cautious live start — and a clue to investigate: why are winners smaller?

5. Deciding whether to go live

Write your criteria before the forward test starts. For example:

Go live (small) if all are true:

• At least 50 forward-test trades completed
• Expectancy above zero after costs
• Maximum drawdown no worse than 1.5 × the backtest's
• Rule adherence at least 90%

Otherwise: return to review — the rules, the execution, or both.

Scaling up safely

• Start live at a fraction of your planned risk — for example 0.25–0.5% per trade.
• Increase in steps (for example 0.5% → 0.75% → 1%) only after each stage shows results consistent with the forward test — say, every 30–50 trades.
• If live results fall outside your drawdown limit, step back down and review.

6. It never really ends

Markets change. Keep measuring:

• Track expectancy in rolling blocks (for example, every 50 trades).
• Compare each block with your forward-test benchmark.
• A sustained drop is a signal to review — not necessarily to abandon — the strategy.

This ongoing measurement is exactly what later tiers build on: advanced risk management (Tier 4) and the live evaluation in Tier 5.

Common beginner mistakes

• Treating demo casually — oversized trades, skipped rules, no journal.
• Stopping the test early after a good run, or abandoning it after a bad one.
• Measuring in dollars only, hiding slippage and inconsistent sizing.
• Expecting forward results to match the backtest exactly.
• Going live at full size straight after the forward test.

Key terms

Practice

1. Write your forward-test setup: strategy version, sample size, risk per trade, and go-live criteria.
2. Start the forward test on demo and record every trade in R, with a checklist yes/no.
3. Every 10 trades, update win rate, average win and loss, expectancy, drawdown, and rule adherence.
4. At the end, complete a backtest vs forward test comparison table and make your decision using your pre-set criteria.

Quick recap

• Forward testing checks the rules and the trader in real conditions.
• Mirror live trading: same risk, same plan, full discipline — and set the sample size in advance.
• Measure everything in R; track expectancy, drawdown, and rule adherence.
• Expect forward results to be somewhat worse than the backtest — investigate the gap.
• Go live only when pre-set criteria are met, then scale up gradually.

---

You've completed the Practitioner lessons. Take each module's knowledge check, then the Practitioner exam to earn your Tier 3 badge.