Testing Gap-Trading Ideas with Probability and Risk Analysis

Summary

The article presents a probability-based framework for evaluating automated trading ideas, using gaps as its example. It models a single-asset Expert Advisor through position, price, and equity changes, then simplifies the analysis by representing activity as a sequence of trades with defined profitability and risk. This lets the author study finite sequences of outcomes without trying to build a complete probabilistic model of price paths.

The proposed starting point is to look for evidence against a zero-drift random walk, under which trading is expected to yield no profit before costs and a loss when spread is included. The article gathers gap statistics, tests a gap-based strategy, and discusses choosing risk per deal. It reports that tests on the examined symbols showed profits in one period, but judges ordinary gaps too rare and small to support confidence in live trading. The analysis makes simplifying assumptions, including fixed spread and ignored slippage, and treats historical parameters as having little effect on outcome distributions; its conclusions therefore have limited scope.

Key ideas

• The framework represents an Expert Advisor’s activity as a sequence of trades with measurable risk and profitability.
• A trading hypothesis can be assessed by testing whether price behavior departs from a zero-drift random walk.
• The gap example uses collected statistics and backtests to evaluate both strategy results and risk per deal.
• The analysis assumes fixed spread, ignores slippage, and simplifies how historical conditions affect returns.
• Reported test profits do not establish that ordinary gaps are viable, since the examined gaps were rare and small.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.