Dear OP;
I sincerely hope that you do not believe that what you have written here is a good idea in the slightest, for it is the case that it is not a good idea.
Not the idea of 'ranked games are for tryhards and people need to meet [reasonable standard] before playing them', but much more specifically:
Using an average needs a definition of time period. Since... ever? Since... the last 6 months? In your last 20 games?
But also your standard.
Kill participation of 4 or 100 creeps killed?
I have ended games with 2 kills 1 assist 90-something creeps and no deaths. Enemy team surrenders because I'm not dying and because the rest of my team is doing good.
So even though I personally am not a large contributor, for a 21-minute game this is just bad. I've had games with almost 30-minute laning phases, too. They tend to be low kill-participation, and I've made my opponents be zoned out of creeps for absurd periods of time-- but in doing so (since I'm not that good), I miss a bunch of CS myself-- and if the rest of my team makes the enemies surrender because they're winning by a large margin, then am I really doing badly? My direct opposition is doing worse than I, and therefore I conclude that I am not doing badly because it was my mission to fight that person for a large period of the early-game's duration.
So OP, I would like to raise the following boolean query:
#Did you think about this suggestion for at least one (1) minute [sixty (60) seconds]?