Board-to-Board: Evaluating Moonboard Grade Prediction Generalization

TL;DR

This paper develops and evaluates machine learning models for predicting bouldering route grades from Moonboard data, achieving high accuracy and demonstrating generalization across datasets, with potential applications in climbing progress tracking.

Contribution

It introduces a novel, bias-reducing feature set and a vision-based approach for grade prediction, advancing the state of the art and enabling practical climbing assessment tools.

Findings

Achieved 0.87 MAE and 1.12 RMSE in grade prediction.

Models generalize across different Moonboard editions.

Vision-based method offers a new approach to route grading.

Abstract

Bouldering is a sport where athletes aim to climb up an obstacle using a set of defined holds called a route. Typically routes are assigned a grade to inform climbers of its difficulty and allow them to more easily track their progression. However, the variation in individual climbers technical and physical attributes and many nuances of an individual route make grading a difficult and often biased task. In this work, we apply classical and deep-learning modelling techniques to the 2016, 2017 and 2019 Moonboard datasets, achieving state of the art grade prediction performance with 0.87 MAE and 1.12 RMSE. We achieve this performance on a feature-set that does not require decomposing routes into individual moves, which is a method common in literature and introduces bias. We also demonstrate the generalization capability of this model between editions and introduce a novel vision-based method of grade prediction. While the generalization performance of these techniques is below human level performance currently, we propose these methods as a basis for future work. Such a tool could be implemented in pre-existing mobile applications and would allow climbers to better track their progress and assess new routes with reduced bias.