The HC Commodities Podcast — Episode Summary

Podcast: The HC Commodities Podcast
Host: Paul Chapman, HC Group
Guest: Hans Balgobin (Systematic Trading Expert; Shell, Millennium, HSBC, Uniper)
Episode: From Systematic Trading to Structural Edge
Date: September 30, 2025

Episode Overview

This episode dives into the evolving world of systematic trading within energy and commodities markets. Host Paul Chapman welcomes Hans Balgobin, a seasoned systematic trader, to demystify the mechanics, philosophy, and strategic significance of systematic approaches. The discussion moves from definitions through real-world model-building to the future impact of data and automation, distinguishing between quantitative and systematic methodologies and their implications for organizational decision making.

Key Discussion Points & Insights

1. Defining Systematic and Quantitative Trading

Timestamps: [01:28]–[06:58]

• Systematic vs. Quantitative:
  • Systematic trading is about applying a repeatable, rational set of rules to market data, resulting in consistent decision-making.
  • Quantitative relates to the use of advanced mathematical, statistical, and computational tools for trade and risk management.
• Clarity:
  • The main value of systematization is clarity of purpose, leading to benefits such as speed, reduced human error, and transparent rationale for actions.
  • "If you want one word, it's about clarity." — Hans [05:01]

2. Model Structure & Methodology

Timestamps: [06:58]–[22:08]

• Inputs and Human Analog:
  • Systematic systems emulate human decision-making by crystallizing “the state of the world” — incoming data like price feeds, fundamentals, and even Twitter news [07:34].
  • Two Model Approaches:
    • Data-driven models search for repeatable patterns in historic data (risk of overfitting).
    • Thesis-driven models start with a hypothesis and use data to confirm or refute it, incorporating humility about model limitations.
• Overfitting & Humility:
  • Overfitting—creating models too finely tuned to historical data—remains the greatest hazard [11:30].
  • "Overfitting is the biggest issue in all sorts of trading, human and quantitative system overfit." — Hans [11:30]
• Frameworks:
  • Hans favors directed acyclic graphs: information flows one-way, preventing “look ahead bias".

3. Core Model Components and Challenges

Timestamps: [15:39]–[26:10]

• Regression/Causality:
  • Everything ultimately traces back to regression—identifying what variables move with or cause shifts in market outcomes [15:53].
• Backtesting:
  • Critical to stress-test strategies, accounting for slippage and real execution costs.
  • Blind walk-forward tests emulate real-world sequential trading.
• Risk Management:
  • Key techniques include mean-variance optimization and dynamic sizing according to confidence in signals and prevailing risk [22:08].
  • Sizing is particularly complex in commodities due to lower correlation/diversification relative to equities.

4. Execution and Compliance

Timestamps: [26:49]–[37:04]

• Frequency Spectrum:
  • High frequency = microseconds–minutes, requiring direct algorithmic links to exchanges.
  • Mid/Low frequency = longer holding periods (days to months), blending systematic and human decisions [32:41].
  • Compliance requirements escalate with speed; in some jurisdictions, algorithm code and identifiers must be filed with regulators [30:39].
  • "For example, if you…by mifid, you need to…attach a tag with your algo ID" — Hans [30:39]
• Execution Strategies:
  • Algorithmic methods like TWAP/VWAP manage execution to minimize detection and slippage.
• Sector Suitability:
  • The systematic edge appears first in liquid, data-rich markets (e.g., gas/power/Nymex) but is now permeating broader commodity classes.

5. Building Systematic Capability

Timestamps: [37:17]–[43:38]

• Investment:
  • Largest upfront costs: clean, comprehensive historical and live data; exchange access.
  • Team consists of data engineers, quant developers, and people with strong coding/math abilities, ideally oriented toward humility and creative hypothesizing.
  • Domain knowledge is valuable but microstructure insights and platform skills are equally prized.

6. Value & Impact Across the Organization

Timestamps: [45:27]–[52:05]

• Beyond Trading Returns:
  • Systematic processes add value by standardizing decision-making, aiding boards/managers, and “diarizing” institutional memory.
  • Proprietary data—such as a company's own trade history and operational nuances—provides a durable edge when interpreted by domain experts, even within public disclosure constraints [50:25].
  • "Your interpretation of public data because of your interaction with people on the ground is more intricate and more likely to be closer to the ground truth." — Hans [47:07]

7. Does Systematic Trading Work? Where Is It Most Effective?

Timestamps: [53:31]–[59:11]

• Repeatability and Alpha:
  • Systematic trading's utility is twofold: delivering repeatable, explainable outcomes and, when well-executed, creating alpha via disciplined execution.
• Which Markets?
  • It thrives in markets with deep, frequent data (e.g., short-tenor, high-frequency) but, as competition grows, the edge concentrates in less liquid or longer-tenor opportunities requiring human judgment.
  • "Data is important and in short term tenors you have more data…in short term, data helps you systematically." — Hans [58:04]
• Strategy Mix:
  • Hans advocates running sets of orthogonal models (trend, mean-reversion, relative value) and allocating dynamically.

8. Future Trends & The Outlook for Systematic Trading

Timestamps: [59:39]–[67:17]

• Democratization & Data Explosion:
  • Data availability is surging, pushing even “human” traders to adopt systematic tools.
  • IoT, better sensors (in batteries, grain silos, etc.), and AI forecasts will flood markets with actionable information.
• Industry Evolution:
  • As systematic methods dominate short-tenor, liquid markets, the “game” pushes humans toward less-crowded, longer-term judgments (e.g., geopolitics, regime shifts).
• Organizational Brain:
  • Systematic desks will increasingly support not just trading but broader strategic decisions, including asset M&A, long-tenor risk, and scenario analysis [65:59].
  • "The same scientific method can be applied to decisions of any tenor and scope." — Paul [65:59]

Notable Quotes & Memorable Moments

• Systematic vs. Quantitative:
  "Systematic is about being systematic, repeatable and rational...quantitative is about mathematical toolset...big overlap but not exactly the same." — Hans [01:57]

• Philosophy of Clarity:
  “If you want one word, it’s about clarity.” — Hans [05:01]

• Perils of Overfitting:
  "Overfitting is the biggest issue in all sorts of trading, human and quantitative." — Hans [11:30]

• Real-world Model Building:
  "I favor something called a directed acyclic graph, ...because information...is going to have to flow in one single time direction." — Hans [13:51]

• Data as the New Edge:
  "The most expensive piece is data...access to historical data...and live data is critical." — Hans [37:32]

• Execution Insights:
  "You can give algorithmic execution side of it for free to the regulator, but you would still be able to keep some stuff to yourself..." — Hans [31:34]

• Where Systematization Wins:
  "Systematic processes add value by standardizing decision making, aiding boards/managers, and 'diarizing' institutional memory." — Paul/Hans [46:35–47:07]

• Human Edge Remains:
  "The human are still the apex predators...they have a view of the minds of humans and other societies...so they should still have an edge on the AIs on the longer term, fewer sample side of things." — Hans [63:22]

Structural / Thematic Summary

Segment Guide

| Topic | Start Time | Key Points | |---------------------------------------------|-------------|-----------------------------------------------| | Defining systematic & quantitative trading | 01:28 | Clarity, comparison with traditional methods | | Model mechanics & overfitting | 06:58 | Inputs, types of modeling, humility, bias | | Model components, regression, risk, sizing | 15:39 | Backtesting, optimization, execution tactics | | Execution, market types, frequency | 26:49 | Compliance, high/mid/low frequency explained | | Building a systematic desk | 37:17 | Costs, skills, hiring, domain knowledge | | Systematic as organization-wide brain | 45:27 | Proprietary/public data blending, impact | | Effectiveness, best-fit markets | 53:31 | Liquidity, alpha creation, strategy types | | Future, data’s growing role | 59:39 | AI, sensors, systematic push & human role |

Conclusion & Takeaways

• Systematic trading is a crucial, fast-evolving organizational pillar in energy and commodity markets, driving clarity, efficiency, and a competitive edge.
• The core challenge is balancing mathematical rigor with human adaptability, especially in avoiding overfitting and navigating less liquid or longer-tenor environments.
• Future structural edges will be grounded in proprietary data interpretation, the fusion of AI and domain knowledge, and the democratization of systematic methods across all decision types—not just trading.
• As systematic strategies dominate liquid markets, human expertise shifts to longer-horizon, complex, and less data-saturated problems.
• Organizationally, building systematic capacity is as much about institutional learning and cross-team collaboration as it is about returns.

This episode offers an expert, in-depth look at the technical, philosophical, and strategic dimensions of systematic trading as it reshapes commodities, both for practitioners and broader organizations.