HydraNet: Momentum-Driven State Space Duality for Multi-Granularity Tennis Tournaments Analysis

TL;DR

This work tackles the elusive momentum phenomenon in professional tennis by introducing the Momentum Score ($MS$) and HydraNet, a momentum-driven state-space duality framework designed to model momentum across multiple granularities (points, games, sets, matches). It integrates explicit and implicit momentum via a four-stage architecture (Hydra learning, Versus Learning, CAAM, multi-granularity classification) and leverages cross-game propagation, micro/macro attention, and adversarial learning to capture dynamic player interactions. A million-scale Wimbledon and US Open cross-tournament dataset is built and used to validate MS modeling, with experiments showing strong point- and match-level predictions, diminished set-level performance, and insightful phenomena such as a halftime momentum effect. The work offers a new foundation for momentum modeling in tennis and other sports, with practical implications for coaching, strategy, and analytics, backed by rigorous ablations and multimodal missing-data analyses.

Abstract

In tennis tournaments, momentum, a critical yet elusive phenomenon, reflects the dynamic shifts in performance of athletes that can decisively influence match outcomes. Despite its significance, momentum in terms of effective modeling and multi-granularity analysis across points, games, sets, and matches in tennis tournaments remains underexplored. In this study, we define a novel Momentum Score (MS) metric to quantify a player's momentum level in multi-granularity tennis tournaments, and design HydraNet, a momentum-driven state-space duality-based framework, to model MS by integrating thirty-two heterogeneous dimensions of athletes performance in serve, return, psychology and fatigue. HydraNet integrates a Hydra module, which builds upon a state-space duality (SSD) framework, capturing explicit momentum with a sliding-window mechanism and implicit momentum through cross-game state propagation. It also introduces a novel Versus Learning method to better enhance the adversarial nature of momentum between the two athletes at a macro level, along with a Collaborative-Adversarial Attention Mechanism (CAAM) for capturing and integrating intra-player and inter-player dynamic momentum at a micro level. Additionally, we construct a million-level tennis cross-tournament dataset spanning from 2012-2023 Wimbledon and 2013-2023 US Open, and validate the multi-granularity modeling capability of HydraNet for the MS metric on this dataset. Extensive experimental evaluations demonstrate that the MS metric constructed by the HydraNet framework provides actionable insights into how momentum impacts outcomes at different granularities, establishing a new foundation for momentum modeling and sports analysis. To the best of our knowledge, this is the first work to explore and effectively model momentum across multiple granularities in professional tennis tournaments.

Figures (4)

• Figure 1: The conceptual diagram of $M_s$ modeling using the Hydra methodology with $M_p$, $M_e$, and $M_i$.
• Figure 2: Workflow of HydraNet: (a) Hydra Learning; (b) Versus Learning; (c) Collaborative-Adversarial Attention Mechanism learning; (d) Multi-Granularity Classification.
• Figure 3: Multi-Granularity Ablation Results for (a) Hydra module, (b) CAAM module and (c) Versus Learning method on the WID and USD Datasets.
• Figure 4: A case study on the multi-granularity analysis capability of the MS metric constructed by HydraNet.