pith. sign in

arxiv: 2606.04068 · v2 · pith:M2TNKYXYnew · submitted 2026-06-02 · ⚛️ physics.pop-ph

Competitive Instability in Judo: The Hidden Mechanism within an AI-Driven Non-Linear Dynamics Framework

Attilio Sacripanti This is my paper

Pith reviewed 2026-06-28 07:21 UTC · model grok-4.3

classification ⚛️ physics.pop-ph
keywords judononlinear dynamicsinstability indexmultibody systembifurcation analysisattractor basinsfunctional instabilityTori-Uke dyad
0
0 comments X

The pith

Judo throws arise from two instability archetypes in the Tori-Uke dyad modeled as a constrained multibody system with a multiplicative Functional Instability Index.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper models the Tori-Uke pair as a coupled multibody system whose throws emerge from symmetry breaking and basin transitions. It isolates two archetypes—rotational collapse of Uchi-mata type and gravitational lever collapse of Seoi-otoshi type—as the pathways that generate every throwing technique. A dimensionless Functional Instability Index, built multiplicatively from geometric, dynamic and coupling terms via local bifurcation analysis, quantifies the critical moments of Kuzushi, Tsukuri and Kake. Fractional Brownian motion with local Hölder exponents describes global dyad displacement, while an AI video pipeline extracts Lyapunov exponents and attractor topology. The result is a proposed shift from technique lists to an instability-centered teaching framework and the claim of first theoretical foundations for predictive judo performance science.

Core claim

All competitive throwing techniques in judo evolve through one of two fundamental instability archetypes—rotational collapse or gravitational lever collapse—within a Tori-Uke dyad treated as a constrained multibody system, and these transitions are quantified by a single multiplicative Functional Instability Index derived from local bifurcation analysis that serves as a dimensionless order parameter for Kuzushi, Tsukuri and Kake.

What carries the argument

The Functional Instability Index It, a dimensionless multiplicative order parameter assembled from geometric, dynamic and coupling variables extracted by local bifurcation analysis of the constrained multibody Tori-Uke system.

Load-bearing premise

The Tori-Uke interaction can be treated as a constrained multibody system whose critical transitions are fully captured by local bifurcation analysis and the multiplicative Functional Instability Index without requiring sport-specific empirical calibration or validation data.

What would settle it

High-speed competition video in which the computed Functional Instability Index values fail to rise sharply at observed successful throws or in which many throws fall outside the two proposed archetypes would falsify the central reduction.

read the original abstract

This study presents a unified nonlinear dynamical framework for understanding, modelling, and teaching competitive judo. The Tori Uke Dyad is formalised as a constrained multibody system whose behaviour emerges from symmetry breaking, coupling dynamics, and transitions between attractor basins. Two fundamental instability archetypes rotational collapse Uchi mata type and gravitational lever collapse Seoi otoshi, suwari version are identified as the core pathways through which all throwing techniques evolve. A Functional Instability Index It is introduced as a dimensionless order parameter. Derived from local bifurcation analysis, it integrates geometric, dynamic, and coupling related variables through a multiplicative nonlinear structure, enabling the quantification of critical transitions such as Kuzushi, Tsukuri, and Kake. Fractional Brownian Motion models the global displacement of the Dyad, where the local H\"older exponent encodes the informational structure of the interaction. An AI based pipeline extracts instability signatures from high frequency competition video, providing objective measures such as finite time Lyapunov exponents, attractor topology, and coupling stiffness. Building on these principles, a three level teaching framework is proposed, shifting judo pedagogy from a technique centred to an instability centred approach.This study establishes the first theoretical foundations for a predictive science of judo performance and outlines future directions for empirical validation, athlete monitoring, injury risk modelling,cross sport applications and judo as neurological rehabilitative tool for Parkinson diseases.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 1 minor

Summary. The paper claims to formalize the Tori-Uke dyad as a constrained multibody system whose behavior emerges from symmetry breaking and attractor transitions. It identifies two instability archetypes (rotational collapse of Uchi mata type and gravitational lever collapse of Seoi otoshi type) as the pathways for all throwing techniques, introduces a dimensionless Functional Instability Index It derived from local bifurcation analysis via a multiplicative nonlinear combination of geometric, dynamic, and coupling variables to quantify Kuzushi-Tsukuri-Kake transitions, models dyad displacement with fractional Brownian motion whose local Hölder exponent encodes interaction structure, describes an AI pipeline extracting finite-time Lyapunov exponents and attractor topology from high-frequency video, and proposes a three-level instability-centered teaching framework. The manuscript asserts that this establishes the first theoretical foundations for a predictive science of judo performance.

Significance. If the claimed derivations, explicit index, and validation were supplied and held, the work would represent a novel application of nonlinear dynamics and bifurcation concepts to a sports context, potentially enabling quantitative performance prediction, pedagogical shifts, and extensions to injury modeling or rehabilitation.

major comments (3)
  1. Abstract: the central claim that the Functional Instability Index It is 'derived from local bifurcation analysis' and integrates variables 'through a multiplicative nonlinear structure' is unsupported because the manuscript supplies neither the explicit functional form of It, the underlying bifurcation equations, nor any derivation steps, rendering the index construction unverifiable and load-bearing for all subsequent claims about critical transitions.
  2. Abstract: the assertion that 'all throwing techniques evolve through one of two instability archetypes identified via local bifurcation analysis' is presented without any bifurcation equations, phase-space analysis, or concrete mapping of techniques to the archetypes, making the axiom untestable and central to the unified framework.
  3. Abstract: the AI-based pipeline is stated to extract 'finite time Lyapunov exponents, attractor topology, and coupling stiffness' from competition video, yet the manuscript contains no methods description, algorithm details, parameter values, or comparison against any empirical dataset, which is required to support the 'objective measures' and 'predictive science' claims.
minor comments (1)
  1. Abstract: minor typographical issues include 'suwari version' (likely intended as 'suwari' for kneeling), 'Parkinson diseases' (should read 'Parkinson's disease'), and escaped LaTeX in 'H"older exponent'.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive and detailed report. We agree that the abstract makes central claims whose supporting derivations, equations, and methodological details are not supplied in the current manuscript, rendering those claims unverifiable as presented. We will revise the manuscript by adding the required explicit content in dedicated sections while preserving the overall framework.

read point-by-point responses

  1. Referee: Abstract: the central claim that the Functional Instability Index It is 'derived from local bifurcation analysis' and integrates variables 'through a multiplicative nonlinear structure' is unsupported because the manuscript supplies neither the explicit functional form of It, the underlying bifurcation equations, nor any derivation steps, rendering the index construction unverifiable and load-bearing for all subsequent claims about critical transitions.

    Authors: We accept this assessment. The manuscript presents It conceptually as a dimensionless order parameter but does not include the explicit multiplicative form or the local bifurcation derivation steps. In revision we will add a new section that supplies the bifurcation equations for the constrained multibody system, the step-by-step derivation of It from the local analysis, and the precise nonlinear combination of geometric, dynamic, and coupling variables. revision: yes

  2. Referee: Abstract: the assertion that 'all throwing techniques evolve through one of two instability archetypes identified via local bifurcation analysis' is presented without any bifurcation equations, phase-space analysis, or concrete mapping of techniques to the archetypes, making the axiom untestable and central to the unified framework.

    Authors: We agree that the claim requires explicit support. The current text identifies the two archetypes (rotational collapse and gravitational lever collapse) from symmetry-breaking considerations but supplies neither the phase-space portraits nor the mapping of individual techniques. The revised manuscript will include the relevant bifurcation analysis, phase-space diagrams, and a table mapping representative throws to each archetype. revision: yes

  3. Referee: Abstract: the AI-based pipeline is stated to extract 'finite time Lyapunov exponents, attractor topology, and coupling stiffness' from competition video, yet the manuscript contains no methods description, algorithm details, parameter values, or comparison against any empirical dataset, which is required to support the 'objective measures' and 'predictive science' claims.

    Authors: We concur that the AI pipeline description is insufficiently detailed. The manuscript mentions the pipeline at a conceptual level but omits algorithms, parameter choices, and validation. In the revision we will expand the methods section with the video-processing pipeline, finite-time Lyapunov exponent computation procedure, attractor reconstruction steps, example parameter values, and at least one illustrative comparison against annotated competition footage. revision: yes

Circularity Check

0 steps flagged

No circularity exhibited; derivation chain not inspectable due to absent equations

full rationale

The abstract introduces the Functional Instability Index It as 'Derived from local bifurcation analysis' and 'integrates geometric, dynamic, and coupling related variables through a multiplicative nonlinear structure' but supplies no explicit equations, bifurcation conditions, or parameter derivations. Without any mathematical steps or self-citations in the provided text that reduce a claimed prediction back to fitted inputs by construction, no load-bearing circular step can be quoted or exhibited. The paper defers all validation and presents the index as a new order parameter without showing equivalence to its inputs. This is the most common honest finding when no derivations are available to inspect.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 2 invented entities

The central claims rest on several postulated mappings between judo actions and dynamical-systems objects whose independent support is not supplied in the abstract.

free parameters (1)
  • Functional Instability Index It
    Dimensionless order parameter whose multiplicative structure combines geometric, dynamic, and coupling variables; values appear selected to mark Kuzushi-Tsukuri-Kake transitions.
axioms (2)
  • domain assumption The Tori-Uke pair behaves as a constrained multibody system whose global motion is captured by fractional Brownian motion with local Hölder exponent encoding interaction information.
    Invoked in the abstract to justify displacement modeling without stated justification or prior validation for judo.
  • ad hoc to paper All throwing techniques evolve through one of two instability archetypes identified via local bifurcation analysis.
    Stated as core pathways without derivation or exhaustive enumeration of counter-examples.
invented entities (2)
  • Functional Instability Index It no independent evidence
    purpose: Quantify critical transitions (Kuzushi, Tsukuri, Kake) as a single dimensionless order parameter.
    New composite index introduced without independent falsifiable prediction outside the model itself.
  • Rotational collapse (Uchi mata type) and gravitational lever collapse (Seoi otoshi type) no independent evidence
    purpose: Classify all throwing mechanisms as instances of two archetypes.
    Two new named instability pathways asserted as exhaustive without supporting classification data.

pith-pipeline@v0.9.1-grok · 5771 in / 1687 out tokens · 28705 ms · 2026-06-28T07:21:57.291138+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

39 extracted references · 1 canonical work pages

  1. [1]

    2, revised per Correction 3): a dimensionless order parameter derived analytically from local bifurcation analysis

    Definition of the Functional Instability Index 𝐼(𝑡) (Eq. 2, revised per Correction 3): a dimensionless order parameter derived analytically from local bifurcation analysis. It integrates dynamical stability (μ_max, maximum real Jacobian eigenvalue), geometric stability (η, normalised COM-to-support- polygon distance), and grip impedance (K_grip) in a mult...

  2. [2]

    Presentation of three level pedagogical model derived from nonlinear dynamics, aimed at teaching athletes to perceive instability, navigate meta stable regions, and exploit critical transitions

  3. [3]

    Incorporation of the local Hölder exponent h_t 𝐻 as an indicator of the informational structure of Kumi-kata and its role in the stochastic-to-deterministic phase transition

  4. [4]

    5.2 Limitations The limitations of this work fall into three categories: foundational, technical, and domain-specific

    Specification of an AI pipeline (Appendix A) for extracting instability signatures from competition video, enabling operationalisation of the framework for coaches and analysts. 5.2 Limitations The limitations of this work fall into three categories: foundational, technical, and domain-specific. 5.2.1 Foundational Limitations Empirical validation. As a fo...

    2015

  5. [5]

    No empirical data were collected or analysed; the methods described here concern the conceptual and mathematical development of the framework

    Methods This study employs a multi-layered analytical approach combining nonlinear dynamics, multibody formalism, stochastic modelling, and artificial intelligence to construct a unified theoretical framework for judo competition. No empirical data were collected or analysed; the methods described here concern the conceptual and mathematical development o...

  6. [6]

    3): Estimated via the Rosenstein algorithm over sliding windows to capture local divergence during the Kuzushi-to-Kake transition

    Finite-Time Lyapunov Exponents (FTLE, Eq. 3): Estimated via the Rosenstein algorithm over sliding windows to capture local divergence during the Kuzushi-to-Kake transition

  7. [7]

    4): Constructed at characteristic phases of dyadic motion (e.g., foot-plant events) to reveal attractor structures and detect interruptions in rotational or directional flow

    Poincaré Maps (Eq. 4): Constructed at characteristic phases of dyadic motion (e.g., foot-plant events) to reveal attractor structures and detect interruptions in rotational or directional flow

  8. [8]

    6): Computed from the variance of dyadic COM displacements across time scales to classify tactical regimes: stochastic exploration, persistent attack, anti-persistent defence

    Hurst Parameter Estimation (Eq. 6): Computed from the variance of dyadic COM displacements across time scales to classify tactical regimes: stochastic exploration, persistent attack, anti-persistent defence. 14

  9. [9]

    Phase Transition Detection: Critical thresholds in Kumi-kata are identified through abrupt changes in: grip topology, relative orientation, local Hölder exponent h_t 𝑑𝐻 𝑑𝑡 (Eq. 5). 6.4 AI Pipeline (Overview) The translation of the theoretical framework into practical coaching tools is mediated by a multistage AI pipeline (full details in Appendix A). The ...

  10. [10]

    Data Acquisition – High-frequency video capture (120-240 fps, single or multi-camera)

  11. [11]

    Pose Estimation – Deep-learning extraction of 2D/3D keypoints for both athletes, followed by segmental COM computation

  12. [12]

    Dyadic Feature Extraction – Computation of symmetry indices, micro-instability markers, relative displacement, and Kumi-kata transition metrics

  13. [13]

    5), and Hurst parameters H (Eq

    Nonlinear Metric Computation – Estimation of FTLE, Poincaré maps, 𝐼(𝑡), local Hölder exponents h_t (Eq. 5), and Hurst parameters H (Eq. 6) from extracted trajectories

  14. [14]

    o Supervised calibration of the logistic threshold χ_crit and steepness k of 𝐼(𝑡)using expert- labelled sequences

    Machine Learning Classification o Unsupervised clustering (HDBSCAN) for technique discovery. o Supervised calibration of the logistic threshold χ_crit and steepness k of 𝐼(𝑡)using expert- labelled sequences

  15. [15]

    The pipeline is designed to be modular: each stage can be refined independently as algorithms improve and as larger, higher-quality datasets become available

    Coaching Outputs – Generation of instability heat maps, early-warning indicators, critical-event summaries, and longitudinal athlete profiles. The pipeline is designed to be modular: each stage can be refined independently as algorithms improve and as larger, higher-quality datasets become available. The current specification represents a blueprint for im...

  16. [16]

    Traditional judo pedagogy is typically organised around techniques, biomechanical principles, or tactical scenarios

    Towards a New Advanced Didactic Program for High-Level Competition The nonlinear dynamical framework developed in this study not only reinterprets judo interactions but also provides the foundation for a structured, instability-centred teaching methodology. Traditional judo pedagogy is typically organised around techniques, biomechanical principles, or ta...

    2024

  17. [17]

    1 – The Tool: defines the constraint structure and the archetype to employ

    Eq. 1 – The Tool: defines the constraint structure and the archetype to employ

  18. [18]

    7 – The Map: dictates how to navigate the stochastic field (environment)

    Eq. 7 – The Map: dictates how to navigate the stochastic field (environment)

  19. [19]

    memory” of movement (persistence) to conserve energy and use the opponent’s stochastic noise to fuel projection. Scientific Core: fBm defines the “texture

    Eq. 2 – The Trigger: establishes when to initiate system collapse. Once collapse begins, Tori applies the mechanical tool (lever or couple). Artificial intelligence functions as a virtual sensor, enabling coaches and athletes to visualise these structures and transform intuitive mastery into a repeatable, quantifiable, scientifically validated training pr...

  20. [20]

    When does Athlete (X) most frequently enter the rotational instability archetype?

    Broader AI Applications in Judo and Beyond The integration of nonlinear dynamics with artificial intelligence opens a broad spectrum of applications extending far beyond the analysis of individual throwing actions. When judo is treated as a coupled dynamical system, AI becomes both a microscope for hidden instability patterns and a predictive tool for per...

  21. [21]

    High-frequency video acquisition (120-240 fps)

  22. [22]

    Dyadic feature extraction (symmetry indices, micro-instability markers, grip topology transitions)

  23. [23]

    Nonlinear metric computation (FTLE, Poincaré maps, Instability Index)

  24. [24]

    Machine learning clustering and classification

  25. [25]

    18 AI applied to judo must operate on properly detrended signals

    Coaching outputs (instability heat maps, early-warning indicators, longitudinal profiles). 18 AI applied to judo must operate on properly detrended signals. As highlighted by Bryce & Sprague (2012), DFA introduces artefacts in the presence of nonlinear trends, compromising feature quality. This confirms the need for AI pipelines based on physically interp...

    2012

  26. [26]

    Conclusion This foundational theoretical work applies nonlinear dynamics to competitive judo, enabled by recent advances in artificial intelligence that allow the evaluation of transient situations imperceptible to the human eye. Viewing competition as a dynamic continuum shifts the classical perspective from a catalogue of throwing techniques to the more...

    2024

  27. [27]

    Vertical Rotational Collapse (e.g., Uchi-mata),

  28. [28]

    naturally tends

    Gravitational Collapse (e.g., Seoi-otoshi, suwari version). These archetypes capture the fundamental pathways to projection. All throwing techniques can be viewed as points on a manifold stretched between these two poles of instability. Sutemi-waza techniques cluster toward the Seoi-otoshi pole (maximal gravitational instability), while Ashi-waza techniqu...

    work page doi:10.1007/3-540-23950-2 2006

  29. [29]

    Estimate 𝐼(𝑡) from pose data: compute COM, joint angles, FTLE over sliding windows Δ𝑡; rotational torque proxies from relative angular velocities and segment inertias; grip stiffness proxies from hand-displacement variance

  30. [30]

    Estimate 𝑄musc(𝑡) indirectly via acceleration magnitudes and segment inertias (inverse dynamics with estimated contact forces), or via EMG when available

  31. [31]

    Estimate 𝜆tact(𝑡) from grip geometry: relative hand positions, grip contact-area proxies, short-term stiffness from local compliance to micro-perturbations

  32. [32]

    Estimate 𝑆(𝑡) as the log-determinant of the short-window covariance matrix Σ of dyadic state variables (COMx, COMy, orientation, key joint angles): 𝑆 ∝ 1 2 log (det Σ)

  33. [33]

    2 via supervised learning on labelled windows (Kuzushi/Tsukuri/Kake), using cross-validation and regularisation

    Calibrate coefficients (𝛼, 𝛽, 𝛾, 𝛿, 𝜖, 𝜁) in Eq. 2 via supervised learning on labelled windows (Kuzushi/Tsukuri/Kake), using cross-validation and regularisation. Confidence intervals are obtained via bootstrap. 27 APPENDIX B Multibody Dynamics of the Tori–Uke Dyad and Formal Definition of the Functional Instability Index B.1. Lagrangian Formulation of the...

  34. [34]

    a gravitational collapse (rapid drop of Tori’s COM), and

  35. [35]

    The shoulder is a kinematic constraint, not a fixed pivot

    a lever action with a migrating fulcrum. The shoulder is a kinematic constraint, not a fixed pivot. B.6.3.1 Time-dependent fulcrum 𝐹(𝑡) = {𝐹1, 0 ≤ 𝑡 < 𝑡𝑐(feet–tatami friction dominates) 𝐹2, 𝑡 ≥ 𝑡𝑐(shins on Tori friction dominate) Torque about the moving fulcrum: 𝜏⃗𝐹(𝑡) = 𝑟⃗𝐹(𝑡)→𝐶𝑈(𝑡) × 𝐹⃗ (𝑡), 𝐈𝐹(𝑡) 𝜔⃗⃗⃗̇ 𝐹(𝑡) = 𝜏⃗𝐹(𝑡). B.6.3.2 Helical descent of the COM ...

    2025

  36. [36]

    Recognises static instability in the opponent,

  37. [37]

    Transforms it into dynamic instability,

  38. [38]

    Maximises μmax > 0 and FTLE to ensure total divergence of Uke’s trajectory,

  39. [39]

    This is the essence of the perfect Ippon

    Drives the system into irreversible gravitational collapse. This is the essence of the perfect Ippon. C.3 Application of the Instability Index 𝐼(𝑡) A Theoretical Example for Coaches The following example illustrates how the Functional Instability Index 𝐼(𝑡) can be applied to a real match sequence using the revised bifurcation-derived formula.𝜒(𝑡) = 𝜇max(𝑡...