You've taken a good first look at space-saving optimizations. These are easy to measure by comparing the before and after sizes of the binary sketch. Measuring execution speed and power consumption is a bit more involved.
Power and speed have a close relationship. More speed generally requires more power. Finding the optimum balance between the two is the goal of overall performance optimization.
Let's explore the limits of both power and speed. First, you find out how fast you can make the Arduino run. Then you slow things down a bit. After that, it's time to find out how much this performance costs in terms of power. Before that, however, you need to equip yourself with the proper tools for the job.