Definition and scope
Paper trading is a way to practice or test a trading workflow by placing simulated orders in a platform, using virtual balances and a simulated order-matching process. It is meant for learning the mechanics of order entry, position management, and reporting—rather than for proving future profitability.
When you evaluate paper trading, separate what is stable (the general concept and the questions you ask) from what is variable (how a specific simulator generates prices, fills orders, and calculates costs). Because there is no live capital at stake, the key question is not “Did it make money?” but “Does the simulator reproduce the conditions that would matter in real execution?”
How it works: the inputs to evaluate
A useful evaluation checklist starts with the rules your simulator applies.
- Price and feed assumptions
- What price does the simulator use for fills (bid/ask/mid), and how often does it update?
- Are prices derived from recorded market data, a live-like feed, or synthetic values?
- If you can select instruments, verify how symbol mapping is handled (for example, whether session times or rollovers are modeled).
- Order execution and “fill” logic
- How does the engine decide whether an order is filled fully or partially?
- For limit and stop orders, what triggers the fill, and at what timestamp?
- Are slippage and partial fills possible, or are fills effectively assumed to be ideal?
- Costs and account math
- Which costs are included (such as spreads, commissions, and financing/rollover where relevant)?
- How are swaps/financing and commissions calculated—by time held, notional size, or another method?
- Confirm the units and rounding used in profit and loss calculations.
- Risk controls and practical workflow
- Does the simulator enforce margin rules, leverage limits, and order rejection logic in a realistic way?
- Are stop-loss and take-profit orders handled with the same execution assumptions as manual closes?
Evidence or example: turn observations into verification
To make the evaluation concrete, run small, controlled checks where you can state assumptions.
- Execution test (assumption: a repeatable trigger): Place a single order type (for example, a limit) and document what condition must occur for a fill. Then compare that with the platform’s trade log fields (order status changes, timestamps, fill price).
- Cost test (assumption: constant position size): Open the same position under identical simulated conditions and verify whether reported costs (spreads/fees/financing if included) scale with size and holding time as you expect.
- Reporting test (assumption: consistent accounting): Cross-check that balance/equity changes align with the trade log (fills, commissions, and any carry). If the simulator’s figures do not reconcile, treat paper results as unreliable.
The goal is not to “prove” a strategy, but to verify that the simulator’s accounting and execution model behave consistently with its stated documentation.
Material limitations and risks (what can fail)
Paper trading often fails as evidence of real-world performance due to systematic differences between simulation and execution.
- Perfect or near-perfect fills: Many simulators fill orders more cleanly than real markets, which can hide slippage and partial-fill risk.
- Simplified spreads and costs: If the simulator uses a simplified spread model or ignores certain costs, outcomes can be overstated.
- Timing mismatch: If order triggers use different timestamps or lower-resolution price data, stop/limit behavior may not match real conditions.
- Market-state gaps: Events like volatility spikes, reduced liquidity, or price gaps can be underrepresented if the simulator smooths or regularizes data.
- Historical-to-future mismatch: Even if a simulation uses historical data, historical relationships do not establish future results.
A practical “failure mode” to look for is when results change dramatically after you adjust assumptions (for instance, switching the order-filling model or enabling cost components). Large sensitivity suggests the simulation is not robust.
Verification and next question
To evaluate a paper trading setup independently, you can use a clear acceptance checklist:
- You can explain exactly how fills are determined for each order type.
- You can list which costs are included and how they are calculated over time and size.
- You can reconcile trade log entries with balance/equity changes.
- You understand where the simulator is likely to be optimistic (ideal fills, simplified spreads, or timing resolution).