I am learning Unity3D and now creating a trivial (useless) game as follows.

The ball rolls down the inclined floor and the camera must follows the ball with the following relationship

• x camera = x ball
• y camera = y ball + 3
• z camera = z - 10

There are two possible ways to control the camera position.

The ball controls the camera

In this scenario, I attach the following script to the ball.

public class Ball : MonoBehaviour
{
    [SerializeField]
    private Transform cameraTransform;

    void Start() { }

    void Update()
    {
        Vector3 newCameraPos = new Vector3
        {
            x = transform.position.x,
            y = transform.position.y + 3f,
            z = transform.position.z - 10f
        };
        cameraTransform.position = newCameraPos;
    }
}

The camera controls itself

In this scenario, I attach the following script to the camera.

public class Camera : MonoBehaviour
{

    [SerializeField]
    private Transform ballTransform;

    void Start() { }


    void Update()
    {
        Vector3 newCameraPos = new Vector3
        {
            x = ballTransform.position.x,
            y = ballTransform.position.y + 3f,
            z = ballTransform.position.z - 10f
        };
        this.transform.position = newCameraPos;
    }
}

Question

Even though both methods work as expected, I am wondering whether there are any pros and cons for each method. Which one should I use?