Background: Playing with others, which in school occurs mainly during recess, might be an enabling factor to improve children’s motor performance, as behaviour is shaped by observing and imitating others. Objective: The aim of the current study was to examine whether children’s play activities and with whom 6- to 8-year-old children play during recess are related to their motor performance. Methods: The study sample consisted of 86 Dutch participants (mean age 6.92 ± 0.62 years, 58.1% boys). A modified version of the System for Observing Children’s Activity and Relationships during Play was used to observe the following play variables: sedentary or active behaviour, group size, activity type, and type of interaction. The Movement Assessment Battery for Children 2 was used to measure three components of motor performance: Manual Dexterity, Aiming and Catching, and Balance. Associations between the play variables and the motor components were investigated using compositional data analysis and logistic regressions. Results: The results showed that children who spent more time in sedentary behaviour rather than in active behaviour were less likely to have motor difficulties in the Aiming and Catching component. None of the other play variables were related to motor performance. Conclusions: One explanation might be that these play variables in itself are not related to motor performance, but that these variables should be investigated as an integrated whole rather than in isolation. Therefore, future research should investigate whether interactions between the play variables are related to motor performance.