By Nummius · October 31, 2026
Backtesting tests a strategy’s rules on historical data to see if they would have had an edge. Forward testing runs the same rules on new data the strategy hasn’t seen. Paper trading tests you: whether you can execute the strategy in real time without breaking the rules. They answer different questions, so use all three, in that order, before trading real money.
What each method answers
| Method | Question it answers | Main blind spot |
|---|---|---|
| Backtesting | “Would these rules have made money in the past?” | Overfitting; look-ahead bias; unrealistic fills |
| Walk-forward / out-of-sample testing | “Do the rules hold up on data they weren’t tuned on?” | Still historical; markets change |
| Paper trading / forward testing in real time | “Can I execute these rules live, consistently?” | No real money, so lighter emotions; perfect fills |
| Small real trading | “Does it work with real fills and real emotions?” | Small sample at first |
Backtesting: does the idea have an edge?
A backtest runs your exact rules over past data and reports win rate, average win and loss, drawdown and expectancy. It’s fast and covers many years and market conditions.
Watch for:
- Overfitting: tuning settings until the past looks perfect.
- Look-ahead bias: using information that wasn’t available at the time.
- Survivorship bias: testing only stocks that still exist.
- Ignoring costs: commissions, spreads and slippage.
Our guide on how to backtest a trading strategy covers these in detail.
Walk-forward testing: does it hold up?
Walk-forward analysis splits history into segments. You tune the strategy on one segment, test it on the next unseen segment, then roll forward and repeat. If results hold up across the unseen segments, the edge is more likely to be real rather than a fit to the past.
Paper trading: can you execute it?
A strategy with an edge is useless if you don’t follow it. Paper trading in real time, or with market replay, tests the human part:
- Do you take every valid signal, or skip some out of fear?
- Do you take signals that aren’t valid?
- Do you hold to the stop and target?
Compare your paper results with the backtest. A big gap usually means execution problems, not a broken strategy.
The order to use them
- Backtest the rules on plenty of history, with costs included.
- Walk-forward test on data the rules weren’t tuned on.
- Paper trade the rules for 30–50 trades, in real time or replay.
- Go live small and keep journaling.
- Compare live results with the backtest and paper results regularly.
If a step fails, go back a step rather than pushing forward.
Common mistakes
- Skipping the backtest and paper trading an idea with no edge.
- Skipping paper trading and assuming you’ll follow the rules.
- Changing the rules mid-test.
- Judging on too few trades.
In Nummius
Nummius Core includes manual and systematic backtesting for stocks and futures, with 80+ indicators and walk-forward validation, so you can test rules on unseen data. The free Trading Arena gives you replay and duels to practice execution, and the journal compares your plan with what you did. Core has a 14-day free trial; see pricing. For practice options, read paper trading vs real trading.
Frequently asked questions
Is backtesting better than paper trading?
They do different jobs. Backtesting tests whether the rules had an edge; paper trading tests whether you can execute them. Use both.
What is forward testing?
Testing a strategy on data it wasn’t developed on, either with walk-forward analysis on later history or by trading it in real time on paper.
How many trades do I need to test a strategy?
More is better. Many traders want at least 30–50 trades for paper testing and hundreds in a backtest before trusting the results.
This article is for educational purposes only and is not financial advice. Trading involves risk, including the loss of money. Past performance, backtests and practice results do not guarantee future results.