Context. Smartphones contain advanced sensors called microelectromechanical systems(MEMS). By connecting a smartphone to a computer these sensors can be used to test new interaction techniques for games. Objective. This study aims to investigate an interaction technique implemented with a gyroscope that utilises the leaning of a user’s torso and compare it in terms of precision and enjoyment to using a joystick. Method. The custom interaction technique was implemented by using the gyroscope of a Samsung Galaxy s6 Edge and attaching it to to the torso of the user. The joystick technique was implementation by using the left joystick of an Xbox One controller. A user study was conducted and 19 people participated by playing a custom-made obstacle course game that tested the precision of the interaction techniques. After testing each technique participants took part in a survey consisting of questions regarding their enjoyment using the technique. Result. The results showed that the leaning technique was not as precise as the joystick implementation. The participants found the learning technique to be more fun to use and also more immersive compared to the joystick implementation. The leaning technique was however also more uncomfortable and difficult to use, and players felt less competent in their ability to control the player character with it. Conclusion. The performance difference might have been due to the lack of familiarity with the leaning technique compared to the joystick implementation. The leaning technique was more difficult to use and more uncomfortable than the joystick method. However, the leaning technique was also more fun to use and more immersive. This offers up the opportunity to keep exploring possibilities with this technique.