Limited prediction of self-rated performance in elite female footballers
Aikeremu, Ailikute, Song, Xiangqin, Pan, Siyu, Xu, Shubo, Farrahi, Vahid, Moreau, David, Heath, Matthew, Williams, A. Mark, Raab, Markus, Quartiroli, Ale, Taylor, Alyx, Paas, Fred, Si, Gangyan and Zou, Liye (2026) Limited prediction of self-rated performance in elite female footballers. International Journal of Clinical and Health Psychology, 26. ISSN 1697-2600
|
Text
Taylor, A 2026 IJCHP full article.pdf - Published Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (1MB) | Preview |
Abstract
Background Routine athlete monitoring captures psychological and physical states across training and match cycles. Whether these indicators distinguish subsequent self-rated match performance below an athlete’s own prior ratings remains uncertain. Objective This retrospective exploratory prediction study examined whether match-day minus one (MD-1) monitoring indicators identified below-baseline self-ratings of overall team performance and of each player’s own offensive and defensive performance among elite female footballers competing in Norway. Methods SoccerMon provided 207 player-match records from 30 elite female footballers in two Toppserien clubs during 2020–2021. Requiring at least three strictly earlier match reports retained 124 records from 25 players. Psychological-behavioural, physical-status and joint predictor sets were evaluated using logistic regression and XGBoost. Below-baseline ratings were more than 0.50 points below the historical mean. Player-grouped validation used nested threshold selection. Average precision (AP) was the primary metric. Results Below-baseline prevalence was 31.5%, 26.6% and 32.3% for team, offensive and defensive ratings. The joint logistic reference model yielded AP values of 0.450 (95% CI 0.346–0.570), 0.308 (0.245–0.409) and 0.280 (0.241–0.363), respectively; ROC-AUCs were 0.590, 0.496 and 0.401. High false-discovery proportions limited classification utility. Conclusions Discrimination was limited and inconsistent. These findings do not support clinical assessment, athlete selection or applied decision use. Individual historical references provide a transparent exploratory outcome definition; objective performance validity and transportability require direct investigation.
| Item Type: | Article |
|---|---|
| Schools: | School of Health and Rehabilitation Sciences |
| Depositing User: | Bridget Roberts |
| Date Deposited: | 02 Nov 2026 11:04 |
| Last Modified: | 03 Nov 2026 10:38 |
| URI: | https://hsu.repository.guildhe.ac.uk/id/eprint/685 |
Actions (login required)
 |
Edit Item |