The two tests
| Backtesting | Forward testing | |
|---|---|---|
| Data | Historical | Live, as it happens |
| Time to run | Minutes for years of data | Weeks or months of real time |
| Capital at risk | None | None: simulated orders only |
| What it shows | Whether the rules had an edge in the past | How the rules and connected feed behave now; fills remain simulated |
| Main weakness | Hindsight: look-ahead bias, overfitting, idealised fills | A short sample: a few weeks cannot prove an edge |
| Also called | Historical simulation | Paper trading, virtual trading, incubation |
What backtesting tells you
A backtest answers questions that need many trades: whether the rules had positive expectancy, how deep the drawdowns were, how the strategy behaved in trending and ranging years, and how sensitive it is to its parameters. A three-year test of an intraday strategy covers several hundred trades and several market regimes in minutes.
Its weaknesses come from hindsight. The data is clean and complete, fills are assumed, and the person running the test knows what happened next. The result is only as honest as the method. Our guides to common backtesting mistakes and avoiding overfitting list the traps.
What forward testing tells you
A forward test runs the strategy in real time against live prices, with orders simulated or placed in a paper account. It answers the questions a backtest cannot:
- Do the signals fire when they should? Live data arrives with delays, late ticks and occasional gaps. A rule that evaluates on a candle that has not finished forming behaves differently from the same rule on historical data.
- Do the fills match the assumptions? Compare the available bid-ask data and simulated execution assumptions. A paper fill does not establish the slippage or liquidity of a real order.
- Does the connection hold? Logins, sessions, order rejections and broker limits appear only in live operation.
- Can you follow it? Watching a strategy lose three trades in a row without touching it is a test of the trader, not the rules.
Its weakness is sample size. A month of forward testing might include fifteen trades, which say nothing reliable about the edge. Forward testing verifies the machinery; it does not prove the strategy.
Why the results differ
A strategy that backtests well and forward tests poorly has usually met one of these:
- Look-ahead bias in the backtest: signals used information that was not yet available. See look-ahead bias.
- Overfitting: the rules matched the past data's noise.
- Costs and slippage larger than the backtest assumed.
- Data differences: the backtest used clean closes; the live feed delivers ticks with noise.
- A changed market: the regime that suited the strategy has ended.
A forward test that fails is cheaper than a live account that fails. Resist adjusting the rules to fit the forward test's few trades; that is overfitting in a new form.
Walk-forward testing: the bridge
Walk-forward testing sits between the two. It splits historical data into windows, optimises the strategy on one window and tests it on the next, then rolls forward. Each test window is "forward" relative to the data the rules were fitted on, so it simulates forward testing many times over, with years of data. Our guide to walk-forward testing explains the method.
A repeatable validation sequence
- Backtest your stated rules on the history your broker makes available, with explicit costs and fill assumptions.
- Validate on data that was not used to choose the rules, where sufficient history exists.
- Forward test (paper trading) on live prices. Compare signals, simulated fills, feed gaps and connection behaviour.
- Investigate differences before changing rules. Neither a fixed number of trades nor a short test proves an edge.
- If you later choose live trading, review the broker connection, quantity, exits and mode separately. A successful simulation does not guarantee live execution.
Forward testing in QuantPower
Forward test · Paper trading · live prices, no real money. Historical candles come from your broker. If they are unavailable, the correct state is Data is not available. QuantPower is a software vendor and does not store or provide data. Check release/account availability before relying on a feature.
QuantPower's paper trading runs a strategy on live prices from your connected broker account and records simulated orders, fills and P&L, so a forward test uses the same strategy definition that will later trade live. The backtesting engine runs the historical test first, and strategy performance reports show both side by side.
Start your free trial
Connect your broker, build a strategy and test it on paper before a single live order goes out.
Frequently asked questions
Is forward testing the same as paper trading?
Paper trading is the usual way to forward test: the strategy runs on live prices and orders are simulated. In QuantPower, Forward test means paper trading on live prices with no real money. A later live test is a separate activity.
How long should I forward test a strategy?
Two to four weeks is enough to check signals, fills and the connection. It is not enough to prove an edge; that comes from the backtest and out-of-sample tests.
Can I skip backtesting and only forward test?
You can, but a few weeks of forward testing cannot tell you how the strategy behaves in a sharp fall or a volatility spike. Both tests answer different questions.
What if my forward test results differ from my backtest?
Expect some difference from slippage and data. A large difference points to look-ahead bias, overfitting or a configuration error. Check those before changing the rules.
Does forward testing use real money?
No. QuantPower Forward test uses live prices and simulated orders, with no real money. Live trading is a separate mode and involves financial risk.