I found those numbers some time back from a source outside the forums, and I believe they were an average for several different rigging ropes.
Rope construction is a big factor in how much core fiber damage occurs for various ratios, so obviously a 16-strand climbing rope (cover dependent) will not be as affected as a kernmantle (more core dependent) and the double braid ropes will be somewhere in between.
Still, the accuracy of the chart or the ropes tested doesn't invalidate the figures entirely, because it gives us a good feel for the price we pay for bad ratios. For any given rope design, we're probably going to see similar jumps in fiber degradation for each jump in bend ratio. In other words, the ballpark figures given will probably be plenty close enough for us to decide how important the bend ratio is to our particular application.
For me, personally, it's important. I'm not worried about the rope breaking, because I've always got enough fudge factor in there... but I try to get as much life out of a rope as I can, so I try to avoid the three things I've found that are the hardest on the rope... excess friction, excess ultraviolet light exposure and horrible bend radius configurations. Not using natural crotch methods, not leaving ropes laying out in the sun, and using pulleys with large sheaves has made a huge difference in usable rope life, in my own experiences.