About GapperLab

Last updated: September 19, 2026

Risk notice. GapperLab is research software, not financial advice and not a broker. Trading smallcaps involves substantial risk of loss. Backtest results are simulations over historical data: they are not a track record, they are not a prediction, and past behaviour of a rule set does not imply future results.

1. What GapperLab is

GapperLab is a research platform for building and testing trading strategies on smallcaps.

It does two things. It lets you express a strategy as an explicit set of rules (entry conditions, stop loss, take profit, position sizing, session windows) and run that rule set over historical intraday data to see what it would have done. And it runs the same rule set against the live market during the session, so the strategy you tested is the strategy you watch.

The design goal behind the product is narrow and deliberate: the backtest and the live scanner should agree. A platform where the tested rules and the live rules quietly drift apart produces confidence that is not earned, which is worse than producing nothing.

2. What GapperLab is not

Being explicit about this matters more than the marketing copy, because the categories below are regulated activities and GapperLab performs none of them.

Not a broker or an executing venue. GapperLab does not hold client money, does not route orders and does not execute trades on your behalf.
Not investment advice. Nothing on this site is a personal recommendation. The platform evaluates the rules you configure; it does not tell you which rules to configure.
Not a signals service. There is no curated list of trades to copy. Alerts fire from your own strategy definitions, against your own filters.
Not discretionary asset management. No funds are managed and no performance is promised.
Not a track record. The figures shown anywhere on this site are the output of simulations over historical data, labelled as such.

3. What we hold ourselves to

Three commitments, stated in terms of what you can check rather than how it is put together.

What you test is what runs. A strategy you backtest and the same strategy watching the live session are evaluated against the same rules, and that agreement is verified automatically rather than assumed. If the two ever disagreed about whether a rule fires, that would be treated as a defect in the product, not as a rounding difference.
Costs are inside the result. Slippage, commission and short-locate fees are applied to the numbers you see, not quoted separately and left for you to subtract. What that costs a strategy is set out on the methodology page.
The limits are published, not buried. Every simulation makes assumptions that flatter it. Ours are written down in one place, including the ones that make backtested results better than reality.