I was watching stylosa (sorry if I missspelled ) and he was coaching a D.VA and he said plat is actually above average because the average rank is low gold how accurate is this ? and what do you think id above average ?
given that the rating scale is 0-5000, if the ranks arent being pushed up/down unevenly, the standard distribution would have most players right at the border of gold/plat. meaning 2500 (high gold/low plat) should be near the average rank
dont think i used the right term but one of the stats undergrads can correct me on it for ya 
Pretty accurate? Take lowest player and highest player. Average would be somewhere in between, meaning around 2300 SR
given that the rating scale is 0-5000, if the ranks arent being pushed up/down unevenly, the standard distribution would have most players right at the border of gold/plat
Yeah, but we dont have 5000sr players
the same force that prevents people from climbing much further than 4.7k is the same that prevents people from falling further than (what should be) around 300sr
sure, people can desperately throw EVERY game to force themselves as low as possible below 500, but i wonder if that amount of players is enough to actually offset the whole ladder.
If I’m not mistaken, the average player is high gold/low plat, so above average would be mid/high plat.
queue times?
nah theres no shortage of time to play games, some of the people in 4.6/4.7 play hundreds of games per season and there’s no time based sr decay anymore. limiting factor in peoples ability to climb past 4.7 is mostly the tapering of sr rewards once you approach the extremes. for example if you look at people with high sample sizes (>100-200 games) in t500, the people at the top have very high winrates like near 60%+. they need those winrates otherwise they can’t maintain that elo, they lose more sr on a loss than they gain on a win
i could be wrong on this, but it seems reasonable to expect the same taper to exist for sr punishments when one is losing constantly in <500
Gold is where most people are in, so high gold is above average
2 Likes
You are right. If you watch GM streamers they are often put with Master players, so the team SR will average out lower than their SR, this means they’ll lose more than they gain.
The same happens when you queue with someone of a lower rank.
In Bronze to Plat you do have performance based SR gains which help mitigate this.
If you go by straight percentage, 50% of the playerbase is in gold and below, according to the last stats Blizzard gave us.
If you go by community standards, anything below 4.5k is trash.
4 Likes
players != accounts which breaks most of these normal dist assumptions
that, and rigging + alts + no-resets → ranks are fake pixels
1 Like
this right here 100 percent facts
I’m sorry, you had an answer to OP’s question?
2 Likes
i don’t. just pointing out that players != accounts so it’s hard to assert the median of active pop etc.
if i had to guess? it would be positive skew almost lognormal with the build up of duplicates attenuating things to the left (a priori bias that ‘alts ending up lower than mains’) which bloats up the lower ranks. so now you have different mode and mean. you might even have multi-modal dist with a clump of players anchored above the clouds, sparsity in some of the inflection points of the pdf, and another mode in the catch-all of bronze.
shout-out to the sponsor of today’s video: siraj
we gonna nft our accnts for ow2
1 Like
Then you won’t mind if I flag your reply to me for being extraneous?
3 Likes
Oh in that case I do have an answer, see what I posted:
What I don’t have is a certifiable answer because i don’t see the recent data.
Is extremely relevant as it explains why it is hard to consider ranks AT ALL, let alone what is “considered average” for the OP. Next time please read the entire post instead of taking the reply out of context. Upsetting.
1 Like
I’m sorry, but this was the post I was referring to:
…which is on brand for you, but certainly not pertinent to the question.
In the future, I’d respectfully request that you keep these kinds of unfounded theories unrelated to the posts you reply to and leave them in your own discussion threads, which are too numerous to keep track of and easily accessible.
3 Likes
It is pertinent to the question because “rank” is losely defined and can’t be appropirately defined given the current ladder conditions. It’s important to frame the question with context so the answer is on point. The context here of current ladder is that:
is that a personnal attack?
False. We have dev statements explaining how the use mmr to rig matches, we have patents that show an attempt to rig OW, we know there hasn’t been a full ladder reset, and we know there are alts (galore).
All facts that make players != accounts which makes the OP question:
Hi OP. See above. I have tried to answer these using informed guess.
Sorry for people trying to derail.
1 Like
i take it you have very good reason to assume that the majority of alt accounts belong to high ranked players
it stands to reason that, in your world with a matchmaker forcing 50/50, anybody below average would see their alt accounts above their mains
what’s the source for that assumption
3 Likes
This is completely unrelated to the question, however much you’d like to shoehorn it into every discussion. They asked if Stylosa’s statement was accurate, and according to the last data we have from Blizzard in 2019, it is.
MMR is not SR, the question did not involve matchmaking at all, all accounts that participate in Competitive have a measurable SR rating, and your argument (weak as it is) applies to none of it.
Please stop appending this extraneous argument to non related threads. It deletes the quality of these boards.
3 Likes