UFC says Takedowns not important . . . except it doesn't

So there I was, just going about my morning routine when I saw several articles with titles suggesting recent research by the UFC Performance Institute that "[t]akedowns not as important to wins as traditionally thought.":

The UFC PI, which is now one year old, published an 80-page analysis last week and the first chapter is entitled “Winning in the UFC.” UFC Performance Institute vice president of performance Duncan French said at a recent media gathering on the promotion’s campus that the entire, lengthy study was founded on that question — how do fighters win fights in the UFC? Everything else, French said, was “reverse engineered” from that.

And the answers to that question might come as a surprise. At least some of them. UFC Hall of Famer Forrest Griffin, who is now the UFC vice president of athlete development, said the main thing that stunned him in the data is that takedown success is only the 19th-most important metric involved in winning fights in men’s divisions.

Interesting.  So what does that mean?  And as one "publication" suggests:

"In brazilian jiu jitsu having a well rounded game has long been a big point of interest. The major point of contention in many academies where most rolls start from the knees are takedowns."

So, is the implication that based on this study we no longer need to learn takedowns?  I am sure some are drawing that conclusion based on the number of headlines I have seen this morning seemingly declaring the death of takedowns in BJJ schools.

Okay, full disclosure.  I am a blue belt, have had exactly zero MMA fights, and have watched exactly zero UFC promotions from start to finish since UFC III.  Where I have some authority to speak on the issue is that I can read . . . and put sentences together . . . and stuff.

This notion that this study says takedowns are unimportant is just not true.  From the Performance Institute's study (which, by the way, none of the articles making the noted declarations linked to):

"It should be noted that these date do not show "how" fighters win, but rather which metrics are 'related' to winning." - Page 16

Okay, now everyone say it with me:  correlation does not equal causation.

Moreover, the metric used is "takedown success (%)", not whether a takedown led to a win.  So, if a fighter had a poor takedown success rate (meaning they only got one takedown out of twenty attempts), but they won the fight because of that one takedown, that fight would still, using the UFC PI's criteria, put takedowns low on the key performance indicators, and leave people to conclude that takedowns are unimportant . . . even though in that one instance it won the fight. 

I am not going to comment further on the roll of takedowns in MMA and self-defense due to my being ill-qualified to do so.  However, in terms of what this research actually shows, I am more than qualified to point out the fallacies in articles with click-bait headlines reporting on said study.