Originally Posted by BigTallUgly
Missed the logic leading up to that argument. In fact, making a statement out of the blue and then saying it wouldn't be fair is really hard to argue against.
Alright, I'll put this in baby terms so your small mind can understand.
There is something called a normal distribution. ELO is not normally distributed, but the actual distribution for elo is probably too hard for the little baby to understand, so I'm still going to use it as a model because ELO comes close to being normally distribution. (it's only slightly more skewed.)
At the center of a normal is something called the arithmetic mean. If you extend to the left or right of the mean, you are extending in units of deviations. According to the empirical rule, 95.54% of all values fall between 2 standard deviations of the mean.
This means that if the true elo distribution mean was 1200 and 1 deviation is 150 elo, you can expect 95.54% of all people in the world to fall between 900 and 1500 elo.
If you start people at 0 elo, there is a 95.54% chance that any given player picked to be in that game is somewhere between 900 and 1500 elo.
This means that people who really do have a true elo of 0 would constantly have people between 900 and 1500 elo in their games as they climb up into upper elo.
Does the little baby understand?