Virtual 5K course normalization

Hi, Have you guys thought about taking elevation gain and other factors into account to allow people running a Virtual 5K to be compared against each other performance-wise?

In other words, the results from races on various courses are available going back years. If you take the huge amount of available data for racecourses around the world, you should be able to make statement like "a 15:30 on course X is like a 17:00 on course Y"... you could even start to take custom values from people running courses anywhere and compare that to all the known values for the known racecourses and start to make comparisons.

My kid in WA state could run against his uncle back in NY and they should be able to compare performance taking into account all of the course variables. Does this make sense?  Thanks for listening!