A Projected Steepest-Descent Algorithm for the Nonlinear Constrained Optimization of Football Team Performance

Coaching staff must repeatedly decide how to divide a finite pool of training and match-day effort across tactical roles, and evidence from load-monitoring research indicates that the resulting gains in synergy and the resulting build-up of fatigue do not scale in proportion to that effort (Mandorino et al., 2025; Harris et al., 2025). Yet the great majority of published squad-composition and lineup-selection models still rely on binary, integer or purely linear-programming formulations (Zhao et al., 2021; Deb and Das, 2023), which cannot represent such curvature. This paper instead formulates team-performance maximisation as a continuously differentiable nonlinear programme whose objective combines power-law direct contributions, pairwise synergistic interactions between roles, and quadratic fatigue penalties, subject to individual workload ceilings, a global squad-load budget, a nonlinear fatigue-interaction inequality and an effort-sum equality constraint. A projected steepest-descent algorithm equipped with an Armijo backtracking line search and an exact Euclidean projection is developed, analysed and implemented in Python using NumPy, SciPy and automatic differentiation via JAX. On calibrated professional-season data the algorithm converges in 47 iterations to a stationary allocation that improves performance by 9.9% relative to a uniform allocation and by 6.2% relative to the club's historical average. A Karush–Kuhn–Tucker multiplier analysis shows that only two constraints bind at the optimum — the global load budget and the pairwise fatigue-interaction inequality — while all five individual role ceilings remain slack. Scenario and parametric sensitivity experiments confirm the robustness of this qualitative structure and show that a linearised counterpart of the same problem underperforms the nonlinear optimum by 3.6% while misidentifying the operative bottleneck. The resulting Lagrange multipliers translate directly into prioritised, quantitatively grounded recommendations for rotation policy and midfield-freshness preservation, illustrating the practical value of a transparent first-order method as decision support for coaching staff.

nonlinear constrained optimisation; projected steepest descent; Karush–Kuhn–Tucker conditions; binding constraints; sports analytics; workload management; football performance modeling

1. Agu, C.M., Unaegbu, E.N., & Chikwendu, C.R. (2024). Harnessing Karush–Kuhn–Tucker (KKT) as optimality conditions for solving nonlinear constrained optimization problems. International Journal of Research and Innovation in Applied Science, 9(7), 503–508. [Google Scholar] [Crossref]

2. https://doi.org/10.51584/IJRIAS.2024.907043 [Google Scholar] [Crossref]

3. Andrei, N. (2013). Nonlinear Optimization Applications Using the GAMS Technology. Springer Optimization and Its Applications, Vol. 81. New York: Springer. https://doi.org/10.1007/978-1-4614-6797-7 [Google Scholar] [Crossref]

4. Bache-Mathiesen, L.K., Andersen, T.E., Dalen-Lorentsen, T., Clarsen, B., & Fagerland, M.W. (2021). Not straightforward: modelling non-linearity in training load and injury research. BMJ Open Sport & Exercise Medicine, 7(3), e001119. https://doi.org/10.1136/bmjsem-2021-001119 [Google Scholar] [Crossref]

5. Bertsekas, D.P. (2016). Nonlinear Programming (3rd ed.). Belmont, MA: Athena Scientific. [Google Scholar] [Crossref]

6. Deb, S., & Das, S. (2023). Optimal selection of the starting lineup for a football team. arXiv:2303.12385. https://doi.org/10.48550/arXiv.2303.12385 [Google Scholar] [Crossref]

7. Harris, A., Gabbett, T.J., King, R., Bird, S.P., & Terry, P. (2025). Influence of acute and chronic load on perceived wellbeing, neuromuscular performance and immune function in male professional football players. Sports, 13(6), 176. https://doi.org/10.3390/sports13060176 [Google Scholar] [Crossref]

8. Lasek, J., & Gagolewski, M. (2021). Interpretable sports team rating models based on the gradient descent algorithm. International Journal of Forecasting, 37(3), 1061–1071. [Google Scholar] [Crossref]

9. https://doi.org/10.1016/j.ijforecast.2020.11.008 [Google Scholar] [Crossref]

10. Mahrudinda, S., Supian, S., Subiyanto, & Chaerani, D. (2020). Optimization of the best line-up in football using binary integer programming model. International Journal of Global Operations Research, 1(3), 114–122. [Google Scholar] [Crossref]

11. Mandorino, M., Gabbett, T.J., Tessitore, A., Leduc, C., Persichetti, V., & Lacome, M. (2025). The interaction of fitness and fatigue on physical and tactical performance in football. Applied Sciences, 15(7), 3574. https://doi.org/10.3390/app15073574 [Google Scholar] [Crossref]

12. Nocedal, J., & Wright, S.J. (2006). Numerical Optimization (2nd ed.). New York: Springer. [Google Scholar] [Crossref]

13. Pantuso, G., & Hvattum, L.M. (2021). Maximizing performance with an eye on the finances: a chance-constrained model for football transfer market decisions. TOP, 29(2), 583–611. [Google Scholar] [Crossref]

14. https://doi.org/10.1007/s11750-020-00584-9 [Google Scholar] [Crossref]

15. Zhao, H., Chen, H., Yu, S., & Chen, B. (2021). Multi-objective optimization for football team selection. IEEE Access, 9, 90475–90487. https://doi.org/10.1109/ACCESS.2021.3091185 [Google Scholar] [Crossref]

• Applied Partial Differential Equation Modeling of Grating-Based Devices for Enhanced Sensing and Monitoring in the Petroleum and Gas Industry
• Analysis of Critical Path Method (CPM) To Optimized Project Scheduling
• Boussinesq Navier Stokes to A Modification on the Lorenz-63 Model
• Evaluating Time-Series Models for Dengue Haemorrhagic Fever Prediction: Arima vs. Sarima and Exponential Smoothing