Autodeleveraging as Online Learning

Paper: arXiv 2602.15182

Authors: Tarun Chitra, Nagu Thogiti, Mauricio Jean Pieer Trujillo Ramirez, Victor Xu

Abstract

Autodeleveraging (ADL) is a last-resort loss socialization mechanism used by perpetual futures venues when liquidation and insurance buffers are insufficient to restore solvency. Despite the scale of perpetual futures markets, ADL has received limited formal treatment as a sequential control problem. This paper provides a concise formalization of ADL as online learning on a PNL-haircut domain: at each round, the venue selects a solvency budget and a set of profitable trader accounts. The profitable accounts are liquidated to cover shortfalls up to the solvency budget, with the aim of recovering exchange-wide solvency. In this model, ADL haircuts apply to positive PNL (unrealized gains), not to posted collateral principal. Using our online learning model, we provide robustness results and theoretical upper bounds on how poorly a mechanism can perform at recovering solvency. We apply our model to the October 10, 2025 Hyperliquid stress episode. The regret caused by Hyperliquid’s production ADL queue is about 50% of an upper bound on regret, calibrated to this event, while our optimized algorithm achieves about 2.6% of the same bound. In dollar terms, the production ADL model over liquidates trader profits by up to $51.7M. We also counterfactually evaluated algorithms inspired by our online learning framework that perform better and found that the best algorithm reduces overshoot to $3M. Our results provide simple, implementable mechanisms for improving ADL in live perpetuals exchanges.

Complexity vs Empirical Score

• Math Complexity: 7.5/10
• Empirical Rigor: 8.5/10
• Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: This paper offers a highly original formalization of crypto perpetual futures autodeleveraging (ADL) as an online learning and sequential control problem. It combines rigorous online optimization bounds with a high-impact empirical replay of the major October 2025 Hyperliquid stress event, showing significant quantifiable improvements over live production mechanisms.

Research Flowchart

  flowchart TD
    A["Research Goal: Formalize & Improve Autodeleveraging (ADL) in Perpetual Futures"] --> B["Formalized ADL as Online Learning Problem (PNL-Haircut Domain)"]
    B --> C["Derived Theoretical Regret Bounds & Optimized Algorithm"]
    
    D["Input Data: Oct 10, 2025 Hyperliquid Stress Episode"] --> E["Counterfactual Simulation & Regret Evaluation"]
    C --> E
    
    E --> F["Findings: Production ADL Queue Causes ~$51.7M Over-Liquidation (50% Regret Bound)"]
    E --> G["Outcome: Optimized Online Learning ADL Reduces Overshoot to $3M (2.6% Regret Bound)"]

Cited References (27)

1. Autodeleveraging: Impossibilities and Optimization (2026)
2. What Is Auto-Deleveraging (ADL) and How Does It Work? (2019)
3. Important Changes: Auto-Deleveraging (2016)
4. HyperMultiAssetedADL: Hyperliquid liquidation replay (2025)
5. HyperReplay: Canonical 10/10 Event Data (2025)
6. ‘Largest Ever’ Crypto Liquidation Event Wipes Out 6 , 300 Wallets on Hyperliquid (2025)
7. Drift Protocol v2: Liquidation Module (2024)
8. ADL Is Broken. Socialized Loss Is the Future (2025)

1 of 27 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.