this can be tested as long as copy / pasting stats from esea's site works. any specific questions you had in mind OP? I WILL SEE WHAT I CAN DO.
edit : I TRIED SOME STUFF.
FUN DISCLAIMER: I'M NOT A STATS GUY, ALL OF THIS IS PROBABLY VERY WRONG. ANYONE WHO IS A STATS GUY PLEASE CORRECT / TEACH ME. IF YOU DON'T GET THE PICTURE I'M SORRY, I'm BAD AT EXPLAINING STUFF.
http://i.imgur.com/LjhIFCB.png
I regressed some data from SEASON 13's TRIHARDS matches. The stats in question are differences between Tri Hards and their opponents DAMAGE per match, along with differences in UBERS and UBERDROPS.
Basically what this is says is that is ubers and drops and damage are identical between the two teams, Trihards have about a 55% chance of winning (cons_). For every single point of damage Trihards inflict more than their opponent, they gain an extra 0.0078% chance of winning. For every uber more than the enemy, Trihards gain an extra 0.035% chance of winning. For every drop more than the enemy, their odds DIMINISH by a huge 7.06%. Of course, the values are reversed for every unit less than the enemy.
Also notice that the t-scores of the ubers / drops are fairly small, indicating they're not very relevant (although dropping much moreso that ubering.,) while damage is very significant.
Again, this is probably very wrong, anyone who knows more about this that I do please feel free to chime in.
TL:DR - read the post sucka.