You got to put the stats in their context though. If you're gonna compare a central midfielder with a box-to-box midfielder, then it's not a fair comparison.
If you are gonna compare Henderson and Yaya and say "Hey, they covered as much distance, they must have the same fitness levels", it's not always the same as we know Henderson runs limitless. The sprint stats also shows this.
One interesting stat there is for example Lallana vs Hazard. They both covered similar distance, they both have similar top speed, similar number of sprints. However, Lallana's avg speed is 7.71 while Hazard is 6.61. That could give Lallana the idea of a "headless chicken", or the idea of Hazard standing still a lot (since we know they're both fast when they actually sprint). Now, only when you compare the avg speed with other people on that list, you'd get context of what that number represent.
This kind of data mining could revolutionize Fifa ratings, Football Manager ratings, for players that are outside of Premier League and not as famous. It could possibly revolutionize club scouting, since you could find hidden gems all over the world without even watching them play.
However, if you chose to pick out certain stats and not put them in their context, you'd get pretty shitty stats. Football journalists aren't any statisticians, but it would be interesting if more people like the Whoscored guys would get more space on TV or in newspapers.