I chose to use natural breaks classification for this map because my data is pretty skewed and, on the histogram below, it shows to me like there are 4 different clear breaks that I can see, so it logically makes the most sense to break it in that way to me.