Backdating
Also known as: backdated track record, backfilled results, simulated history, back-tested as live
What is it?
Backdating is presenting a simulated history as if it were live trading - the equity curve starts years before the strategy existed, because those years were computed afterwards rather than traded. It is easy to spot once you know the shape. A signal service launches in March 2026 and publishes a curve running from January 2022, smooth through every major event, with a maximum drawdown of 6% and no gaps. The first fourteen months of that curve are a backtest: the rules were written knowing what happened, the losing variants were discarded, and the entries were filled at prices nobody was quoting to a real account.
| What you are looking at | Jan 2022 - Mar 2026 | Mar 2026 onward |
|---|---|---|
| How the trades were produced | Computed after the fact | Traded forward |
| Spread and commission paid | Assumed | Actually charged |
| Signals missed to downtime | None | All of them counted |
| Rules changed with hindsight | Possible | Impossible |
| Maximum drawdown shown | 6% | 14% |
The visible tell is usually the seam - the live portion after launch is bumpier, has a deeper drawdown, and a lower win rate than everything before it, because it is the only part that met a spread. The distinction is not academic. A backtested curve does not pay slippage on the trades it would have missed, does not skip signals that arrived while the platform was down, and never sizes down after a bad month. A verified live record from a third-party tracker starts on the day the account was connected and cannot be extended backwards.
Ask for the start date of the verified portion, compare the drawdown before and after it, and treat anything before that date as a hypothesis. Past performance, simulated or live, does not indicate future results. Your capital is at risk. See /risk-warning.
Why it matters: A curve that predates the strategy is a backtest wearing a live label, and the segment before the verification date is the part that never paid a spread.
It inflates every headline statistic a decision is made on, so capital gets allocated to results that were computed rather than earned.
Real-world example
A service launched in March 2026 published an equity curve from January 2022 with a 6% maximum drawdown; the post-launch segment alone showed 14%.
How SignalBots handles it
SignalBots labels backtested figures as backtested and keeps them separate from published signal history, so you can see which part of a record was traded forward rather than computed after the fact. See /risk-warning.
Pro tip
Ask one question - on what date did third-party verification begin - and measure the drawdown and win rate only from that date forward.
Common pitfalls
Judging a record by how long the curve is. Length is free when the early years are simulated; only the verified segment carries information.
Frequently asked questions
Is backdating illegal?
Publishing simulated results is generally permitted where they are clearly disclosed as hypothetical. What crosses the line in most jurisdictions is presenting them as actual trading, and disclosure rules vary by regulator.
Are backtests useless then?
No - a backtest is a legitimate research tool for rejecting bad ideas cheaply. It becomes misleading only when its output is displayed as a trading record instead of as a simulation.
What does third-party verification actually prove?
That the trades shown came from a real account the tracker was connected to, from the connection date onward. It does not vouch for anything before that date, and it does not make the strategy suitable for you.
Why is the live portion usually worse?
Because it pays costs the simulation did not - spread, commission, slippage, missed fills, and downtime - and because the rules can no longer be adjusted with hindsight.
How much verified history is enough?
Enough trades to cover more than one market regime, so a quiet trending stretch is not the whole sample. A short verified record is honest but tells you very little either way. Your capital is at risk.
Trading involves substantial risk of loss. Historical and backtested results do not guarantee future performance. Read the full risk warning.
