MACD Signals with Probability-Based Position Sizing

Summary

This interview profiles a novice Automated Trading Championship participant and summarizes his GBPUSD Expert Advisor. The robot uses MACD to generate buy or sell signals, then applies manually estimated probabilities of subsequent price movement to decide whether to trade and how large a lot to place relative to the account balance. The developer describes using mathematical models alongside practical experimentation, but says the algorithm lacks a rigorous justification and may risk the entire deposit.

He reports assessing the system in the tester and using optimization and statistics to estimate signal probabilities, while choosing parameters for maximum profit. He also says the approach appeared to work better with considerable price oscillations and that the chosen currency pair was essential to its calculations. These are personal observations from a contest, not a validated forecasting method: the probability estimates and trading rules are not disclosed, and the interview provides no detailed test design or risk-adjusted results. The participant stresses drawdown and profitability as important measures and acknowledges uncertainty about deploying the strategy live.

Key ideas

• The described Expert Advisor uses MACD signals and estimated conditional probabilities to decide whether to trade.
• Lot size is calculated using the probability assessment and current account balance.
• The developer estimated probabilities manually and checked behavior with tester statistics and optimization.
• The strategy was designed specifically for GBPUSD and was reported to perform better amid larger price swings.
• The interview provides no reproducible rules or robust performance evidence, and its developer warns that the system could lose the deposit.

Tags

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