Multi-Horizon Career Longevity Prediction for NHL Skaters: Landmark Logistic Regression Reveals Stage-Specific Risk Factors

Authors

  • Elizabeth J. Goldstein ORRO AI GENIUS LLC, North Carolina, USA
  • Jessica A. Pearson ORRO AI GENIUS LLC, North Carolina, USA

DOI:

https://doi.org/10.3384/ecp221002

Keywords:

Hockey analytics, Career longevity, Landmark analysis, NHL, Kontinental Hockey League, Competing labor markets

Abstract

We construct a dataset of 5,754 National Hockey League skaters spanning 1979–2024 and apply landmark logistic regression at four career horizons to predict whether a player who has reached season N will play at least one game in season M. The central empirical finding is that the penalty for Russian-born players concentrates at the season 3-to-5 transition (OR = 0.314, p < 0.0001), consistent with mid-career migration to the Kontinental Hockey League after entry-level contracts expire.

Downloads

Published

2026-09-25

Issue

Proceedings of the 5th International Ice Hockey Analytics Conference LINHAC 2026 Research Track and Community Notes

Section

Research Papers

License

Copyright (c) 2026 Elizabeth J. Goldstein; Jessica A. Pearson

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.