When people try to defend an overpowered champion's WR by saying "low sample size only Plat+"

It doesn't work like that at all. lol

https://upload.wikimedia.org/wikipedia/commons/thumb/f/f9/Largenumbers.svg/400px-Largenumbers.svg.png

That kind of supports what you are arguing against though. Overall, a 56% win rate isn't too far off from 50% when it can range from 0-100%, but it's considered a very good win rate in LoL. Within the first 4% of the trials of this graph at least, the number goes from 3.5 to about 3, which, when estimating (or whatever) from 1 to 6, is pretty huge. If you compare it to 0 to 100, this would be about a 10% win rate increase or decrease.

But the expected value is only plat+. There is no "universal winrate" that people use, because the difference between low and high rank is too great. You have to either choose the "expected value" for (insert high rank) or an "expected value" for (insert low rank)

When people try to pretend that winrate means much in terms of balance