Test it in steps, each of which catches a different way a system fails, and put money behind it only once it has passed all of them. A challenge fee or a first live deposit buys you an answer you can get first, at your desk, for the cost of the data.
1. Backtest it with your real costs
Run the rules over historical bars with the commission, spread and slippage your broker or prop firm actually charges. A system that looks profitable with free fills can lose money once every fill pays the spread. Set costs per instrument, since they differ. See How realistic do slippage and spread need to be in a backtest?
2. Look at the trades, one by one
A summary hides the trades that make it. Open the run and check that each entry and exit happened where your rules say it should, at a price the bar allows. Reamer Research's visual trade diagnostics show every closed trade on its own bars beside every value your strategy printed, so a rule that fires on the wrong bar shows up at a glance.
3. Run it on data you did not tune on
A system tuned until the backtest looks good learns the past. Hold back a stretch of data, or walk forward through it, and judge the system only there. See How do I know if my backtest is overfit?
4. Shuffle the order of the trades
The same trades in a different order can produce a much deeper drawdown. Reamer Research's Monte Carlo resamples a run's per-trade returns and reports loss, ruin and maximum-drawdown percentiles, seeded so the numbers repeat. If you are buying a prop firm challenge, also replay the trades through its daily loss limit and drawdown rules: see How do I backtest a strategy against prop firm rules?
5. Paper trade the same orders
A backtest cannot test your connection, your data feed or your broker's order handling. Paper trading can. Reamer Research's relay sends the same orders your strategy returns in a backtest to your broker's paper or demo account through the open-source receiver. When you switch to the live account, the strategy stays the same. See Can I paper trade before going live?
6. Start small and compare
Go live with the smallest size your account allows, and compare the live fills, spread and slippage with what the backtest assumed. Where they differ, put the live figures back into step 1.
Why repeatable results matter here
Every step above compares one run with another: before and after a change, tuned data against held-back data, backtest against paper. Reamer Research gives the same result to the byte from the same data, settings and seed, and settles a bar that touches both your stop and your target on a seeded tick path inside the bar. So when a result changes, your system or your data changed.
Reamer Research runs on your own machine, on Linux, macOS, or Windows through WSL2. The free download runs steps 1 to 4 in Evaluation Mode, on one ticker over up to 30 days of bars. A licence, $20 per seat per month, cancel any time, removes every limit and adds paper and live trading through the relay.