Ford–Fulkerson Flow as a Filter for ICT Market Structures

Summary

The article describes an Expert Advisor that represents swings, Fair Value Gaps, Order Blocks, and liquidity pools as nodes in a directed graph. It assigns edge capacities using tick volume, prior price reaction, distance, and structure type, then applies a Ford–Fulkerson maximum-flow calculation, implemented with an Edmonds–Karp search, to estimate whether a path from price to a liquidity target is sufficiently strong. ICT structures supply the directional setup; flow acts as a qualification filter, with flow also linked to position sizing and measurements aggregated across timeframes.

The document provides implementation settings and describes a backtest over a stated two-month period, but the supplied excerpt omits the equity curve and detailed results. It therefore does not let readers assess performance, costs, robustness, or whether the capacity scores predict follow-through. Thresholds and scoring weights are design choices that would need independent validation, and the article’s claims about the framework are not supported here by comparative or out-of-sample evidence.

Key ideas

• The system encodes detected price structures as nodes and assigns capacities to their directed connections.
• Maximum flow is used to screen whether a path toward a liquidity target has enough modeled capacity.
• ICT structures determine the candidate direction, while the flow score qualifies the setup and informs position size.
• The design combines flow measurements from multiple timeframes using configurable weights.
• The excerpt omits detailed backtest results, so it does not establish the filter’s predictive value or robustness.

Tags

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