Forward Test (Paper Trading)
Also known as: paper trading, demo test, out-of-sample live test
What is it?
A forward test means running a strategy on live, unfolding market data that it has never seen before, usually on a demo or simulated account, to check whether it behaves as expected before you risk real money. Where a backtest replays old historical data, a forward test moves with the market in real time, so the strategy faces genuinely new conditions it could not have been fitted to. This is sometimes called paper trading because no real capital is at stake while you observe the results. Forward testing matters because it catches problems that a backtest can hide.
On live data you finally see the real spreads, commissions, and slippage you would actually pay, and you find out whether the strategy still works on data it was never optimised against. A strategy that looked flawless on historical data can behave very differently once it meets fresh conditions. To be useful, a forward test should run long enough to span more than one type of market, since a calm two weeks proves little about how the approach handles a trending or volatile stretch. Keep in mind what a forward test does and does not tell you.
A successful forward test increases your confidence that an edge is real, but it is still a sample of past behaviour, not a forecast. Past performance does not guarantee future results, there is no risk-free strategy, and once you go live your capital is genuinely at risk.
Why it matters: Forward testing catches problems a backtest hides, like real spreads, slippage, and behaviour on data the strategy was never fitted to.
Forward testing is the key check between a backtest and risking real capital.
Real-world example
After a strong backtest, a strategy runs on demo for a month; if live behaviour matches, confidence in it grows.
How SignalBots handles it
SignalBots' demo mode lets you forward-test a feed on live data before switching it to a real account. See /risk-warning.
Pro tip
Forward test for long enough to span different conditions; a calm two weeks proves little about a trending or volatile regime.
Common pitfalls
Skipping forward testing and deploying straight from a backtest into live money.
Frequently asked questions
How long should I forward test?
Long enough to cover varied market conditions, typically several weeks to months, not just a single calm or trending stretch. The more conditions it survives, the more meaningful the result.
Is forward testing the same as paper trading?
They overlap. Paper trading is the practice of placing simulated trades with no real money, and a forward test uses that approach to validate a strategy on new, live data over time.
If a forward test goes well, will live trading match it?
Not necessarily. A demo can have slightly better fills and no emotional pressure than a funded account. A good forward test raises confidence but does not guarantee identical live results, and capital is at risk.
Can I skip the backtest and just forward test?
You can, but it is slow because you can only test as fast as the market unfolds. A backtest quickly screens out poor ideas, and forward testing then validates the survivors on fresh data.
Why does demo behaviour sometimes differ from real money?
Real accounts can face wider spreads, slower fills, and the psychological strain of risking actual capital. These differences are exactly why a forward test is a check, not a promise of future profit.
Trading involves substantial risk of loss. Historical and backtested results do not guarantee future performance. Read the full risk warning.