Paper: arXiv 2604.12082

Authors: Alessandro Falezza

Abstract

Battery energy storage systems (BESS) participating in multi-market electricity trading require price forecasts to optimize dispatch decisions. A widely held assumption is that forecast accuracy, measured by standard metrics such as mean absolute error (MAE), drives trading performance. We challenge this assumption using a hierarchical three-layer optimization system trading simultaneously on frequency containment reserve (FCR), automatic frequency restoration reserve (aFRR), day-ahead, and continuous intraday (XBID) markets in Germany and Switzerland over 2020-2025, with real market data from Regelleistung.net and Swissgrid. We find that rank correlation (Kendall tau), rather than MAE, is the primary predictor of intraday dispatch value: forecasts above an empirical threshold of tau approximately 0.85-0.95 capture up to 97-100% of perfect-foresight revenue, while persistence forecasts with near-zero tau capture only 33%. This threshold is stable across market regimes and volatility levels, and reflects the ordinal structure of the dispatch problem. Furthermore, under reserve market constraints, FCR capacity revenue exceeds XBID by 6.5x per MW, making capacity allocation – not forecast accuracy – the primary driver of total revenue. In the Swiss market, hydrological surplus anomalies are significantly associated with balancing market revenue (p = 0.0005), a mechanism absent from existing German-focused literature. These findings reframe forecast evaluation for BESS operators: the relevant question is not what the MAE is, but whether the forecast achieves tau-sufficiency.

Complexity vs Empirical Score

  • Math Complexity: 4.5/10
  • Empirical Rigor: 8.5/10
  • Quadrant: Street Traders — practical and empirical, lighter on theory

Why this score: The paper presents strong empirical evidence using real high-frequency market data across multi-year regimes to challenge standard forecast error metrics in storage dispatch. It successfully demonstrates that ordinal rank correlation (Kendall’s tau) is a far better proxy for real-world trading revenue than MAE, providing actionable insights for quantitative trading strategies in energy storage.

Research Flowchart

  flowchart TD
    A[Research Goal: Evaluate Forecast Metrics vs. Battery Trading Performance] --> B[Data Inputs: 2020-2025 German & Swiss Markets: Regelleistung, Swissgrid, Hydrology]
    B --> C[Methodology: Hierarchical 3-Layer Optimization: FCR, aFRR, Day-Ahead, XBID Intraday]
    C --> D[Forecast Evaluation: Compare Metric Performance: MAE vs. Rank Correlation Kendall Tau]
    D --> E[Finding 1: Tau Threshold 0.85-0.95 Captures 97-100% Intraday Value vs. 33% for Persistence]
    D --> F[Finding 2: Reserve Capacity Revenue Exceeds Intraday Trading by 6.5x per MW]
    D --> G[Finding 3: Swiss Hydro Surplus Anomalies Significantly Impact Balancing Revenue p=0.0005]

Cited References (11)

  1. Maximizing Battery Storage Profits via High-Frequency Intraday Trading — → 2504.06932
  2. Coordinated Trading Strategies for Battery Storage in Reserve and Spot Markets (2024) — link
  3. The Value of Battery Energy Storage in the Continuous Intraday Market: Forecast vs. Perfect Foresight Strategies (2025) — link
  4. BESS Revenue Index 1h — German Energy Storage Revenue Index (2024) — [link](https://www.regelleistung-online.de/ german-energy-storage-revenue-index/ bess-revenue-index-1h/)
  5. The Value of Coordination in Multimarket Bidding of Grid Energy Storage (2023)
  6. Statistical and Economic Evaluation of Forecasts in Electricity Markets: Beyond RMSE and MAE — → 2511.13616
  7. Predict-then-optimize or predict-and-optimize? A n empirical evaluation of cost-sensitive learning strategies (2022) — link
  8. Decision-Focused Learning with Machine Learning Proxies for Energy Storage Systems (2024) — [link](https://www.mech.kuleuven.be/en/tme/research/ energy-systems-integration-modeling/ pdf-publications/wp-esim2024-5.pdf)
and 3 more
  1. Integrating Renewables in Electricity Markets: Operational Problems (2013) — link
  2. A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition (1989) — link
  3. Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor (2018) — link

2 of 11 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.