In early childhood, factors that contribute to motor‐skill engagement (MSE) are unknown. Our aim was to explore the relationships between actual and perceived motor competence and their influences on MSE on a balance bike (bike with no pedals). A secondary aim was to investigate whether MSE had an effect on ability on a balance bike. This study comprised of 45 children (29% female) aged 4.5 ± 0.5 years. MSE was assessed using distance travelled on a balance bike over an 8‐week period. Actual motor competence was assessed using the Movement Assessment Battery for Children, second edition. Perceived motor competence was assessed using the Pictorial Scale of Perceived Movement Skill Competence. Ability on a balance bike was measured using timed trials on a specifically designed track. Pearson product‐moment correlations were used to assess relationships between actual and perceived motor competence and ability on a balance bike. Linear regressions were used to examine whether actual or perceived motor competence or ability on a balance bike predicted MSE. Repeated measures ANOVA was used to examine whether there was a difference in ability on a balance bike between three MSE groups over 8 weeks. No relationships were found, and none of the variables predicted MSE. There was a significant difference between the MSE groups on ability on a balance bike over time (P = 0.019). Investigating the contributors to MSE on a novel cycling task during early childhood provides knowledge to ensure children are given the best opportunities for practice and acquisition of skills.