Skip to content
@applied-probability-institute

Applied Probability Institute

Independent quantitative research institute specializing in discrete probability, margin decomposition, stochastic modeling, and mathematical game theory.

Applied Probability Institute (API)

Research Portal License: MIT Open Science: Datasets Datasets: 10K EPL Lines

The Applied Probability Institute is an independent quantitative research collective dedicated to open-access mathematical modeling, empirical market benchmarks, and reproducible stochastic algorithms across sports wagering markets and competitive information economics.


🔬 Core Open-Source Repositories

Repository Focus & Domain Language / Stack License
no-vig-fair-odds Margin stripping algorithms (Multiplicative, Additive, Power, Shin's insider fraction $z$) Python 3.11 / SciPy MIT
kelly-criterion-sim Log-wealth maximization, fractional Kelly scaling & Monte Carlo risk-of-ruin engine Python 3.11 / NumPy MIT
poisson-match-models Independent & bivariate Poisson goal expectancy distributions for 1X2 and over/under markets Python 3.11 / Poisson MIT
research-papers 10,000 EPL fixture closing line benchmark dataset, reproducibility notebooks & BibTeX citations Open Access / CSV / Py CC-BY-4.0

📊 Research Pillars

  1. Margin Decomposition & Market Clearing: Quantitative extraction of true implied probability densities from multi-way bookmaker price vectors utilizing Multiplicative, Power, and Shin's informed trader parameter ($z$) equilibrium models.
  2. Optimal Capital Growth & Drawdown Control: Theoretical parameterization of John L. Kelly Jr.'s log-optimal betting criterion ($f^* = \frac{bp - q}{b}$), fractional volatility reduction ($f_{\text{half}} = 0.50 \cdot f^*$), and asymptotic Brownian ruin boundaries under finite bankroll horizons.
  3. Discrete Goal Expectancy & Count Distributions: Stochastic modeling of association football and ice hockey scorelines via bivariate Poisson processes ($\lambda_H, \lambda_A$) accounting for low-scoring covariance and draw deflation.
  4. Empirical Closing Line Efficiency: Longitudinal econometric tracking of Closing Line Value (CLV), measuring pricing efficiency, liquidity formation, and the favorite-longshot bias across Tier-1 European football competitions.

🌐 Publications & Live Interactive Calculators

All research papers, empirical benchmark datasets, and interactive zero-runtime JavaScript calculators are accessible open-access at sportsbettingmath.org:


Citation

If you utilize our algorithms, simulation suites, or benchmark datasets in your academic research or financial modeling, please cite:

@article{api2026overround,
  title={Empirical Margin Decomposition, Closing Line Efficiency, and Information Fraction Estimation in European Football Wagering: A 10,000-Match Study},
  author={{Applied Probability Institute}},
  journal={Applied Probability Institute Research Hub},
  year={2026},
  url={https://sportsbettingmath.org/research/overround-decomposition-study/}
}

Popular repositories Loading

  1. no-vig-fair-odds no-vig-fair-odds Public

    Mathematical margin decomposition algorithms (Multiplicative, Additive, Power, Shin's Method) for extracting true implied probabilities from betting lines.

    Python

  2. kelly-criterion-sim kelly-criterion-sim Public

    Optimal geometric compounding bet sizing and Monte Carlo risk-of-ruin simulator for fractional Kelly staking systems.

    Python

  3. poisson-match-models poisson-match-models Public

    Bivariate Poisson goal expectancy engine for soccer and hockey outcome modeling (1X2, Over/Under, correct score matrices).

    Python

  4. research-papers research-papers Public

    Academic preprints, benchmark datasets (10K EPL closing lines), and reproducible empirical analysis pipelines from the Applied Probability Institute.

    Python

  5. .github .github Public

    Institutional profile and quantitative research overview for Applied Probability Institute (sportsbettingmath.org)

Repositories

Showing 5 of 5 repositories

People

This organization has no public members. You must be a member to see who’s a part of this organization.

Top languages

Loading…

Most used topics

Loading…