Why was Performance-Based SR

Removed for Diamond and above? :gem:

Is there a specific issue with PBSR in the higher ranks? :thinking:

2 Likes

PBSR is largely to help get smurfs out of low ranks asap because their stats make their true rank so obvious to the system.

In diamond+ that’s less of a thing so it goes back to whether or not a player can consistently win. Stats are (mostly) meaningless when it comes to measuring actual performance unless it becomes extreme.

9 Likes

Yeah except it equally deducts points from you if it thinks your doing badly so the smurf can just soft throw for a moot point. Worse is that the presence of smurfs means we’re automatically being compared to them. If it were truly anti smurf then they would only make it capable of boosting SR and only do so on extreme cases and after all if it vy chance boosted someone up accidentally they would rank down i imagine.

It hasn’t fixed anything and they refuse to remove it, it was largely not even wanted but they forced it upon the community. So it seems to me like its not really all about smurfing or getting people to their rank quicker. Im willing to bet on average it does more harm than good.

Nothing against you personally btw, rant is against pbsr

Stats are mostly meaningless which is why it should be removed overall. If Blizzard had any sense of competitive integrity.

Not true, spend a decent amount of time on the competitive forums and you’ll see how many people whinging that SR isn’t “based on performance”

There were plenty of people who wanted PBSR

I was going according to this post

Written back then.

Back in the day players could abuse non-meta picks like Torb and Symm, deliberately throw matches and abuse their teammates, have a win rating as low as 30% and still maintain their skill rating. That is why they removed PBSRAs starting in season 8.

There was even one player that did this across multiple accounts who eventually had to be hardware and IP banned.

1 Like

Yes but what about discrepancies for today where you can have a 60% win ratio over 100 games on a hero an still gain les than average SR while losing average SR (22 gain 25 loss avg) Still seems broken to me

If this is in Platinum and lower ranks that is possible. Also don’t forget that other factors such as strength of schedule and frequency of play also apply at all ranks which can adversely affect your SR adjustments.

Oh good we just accept that its broken ok

Basically Performance Based Skill Rating Adjustments still work in Platinum and lower ranks because at those ranks, individual play is likely to carry more than team coordination.

I would also emphasize that good Performance Based Skill Rating Adjustments are NOT based on getting “4 gold medals.” PBSRAs are based on your performance in that match compared to community averages of the same hero(es) used. In other words, your skills on Tracer are compared to others who also use Tracer in their own matches. PBSRAs do not compare your performance to that of your opponents or teammates in the same match.

One other thing, even at the lower ranks, any PBSRAs are only a small influence on your overall skill rating adjustment. Winning the match is always going to be a bigger priority to gaining skill rating.

2 Likes

Here’s the tricky part about that argument however, if you argue this then you’re saying that wins don’t actually mean anything that they’re not reflective of the player’s skill because it’s always going to be one character carrying over the other. If a 60% win ratio over 100 games isn’t enough that a person is still in that shady area, that means that a person can can still always rank to lucky games after 100 and 100 lucky games is worth 2,500 SR which would make for a very inaccurate win-loss system.

Also in the same way the gold medals are not a accurate measure night is it with this system because gold medals measure you against your team members but this measures you against the average it’s the same thing, not at all indicative of what’s needed in that match or what you’re doing and not everything is stats largely there is strategy and team cohesion.

because we get things like hide and mass rez boosters going from silver to masters in a matter of matches which was the main point of its controversy and it general being a really dumb mechanic for any future abuse of it system as it takes into account both yuore average and community average i believe. so you could intentionally do bad to reduce your average slightly and when you increase it you get large bonuses

I edited my post above. PBSRAs are a small influence and your win/loss record will have more of a say to gaining rank. Its the PBSRA that is likely influencing that one or two points of SR that can feel like you need to win more than 50% of your matches.

One other thing, I am explaining things exactly how they work in the game (all of which is documented from dev posts), and whether you think that system is best for the game or not is up for you to express any opinions here on the forums.

Which is really deceptive because the last time I didn’t like calculation on support I had a 57 over 69 games win ratio if I included all supports played. I gained 111 and lost 112 to pbsr average gain for match 22.05 average loss per match 25.06. imagine that on average doing better than the competition yet not getting better or average pbsr compensation and actually just losing more than a gain.

Realistically anyone doing above 50% should technically be doing above average (especially when they play many games) because the average is 50% win ratio which should be indicative of the average player.

And don’t even get me started on my overall career loss with pbsr which has been -1,400 SR while maintaining a win ratio of 53% (since it includes my newbie days) may not be the highest but still it should be better than average or at least average. Statistically this is wrong.

See it all adds up in the end to a lot actually

I do appreciate you saying that we could express our opinions though.

So, the intent was to remove it entirely but after some pushback Blizz decided to keep it in for lower ranks. The reason it was being removed at all was because Blizz felt like the system was discouraging hero switching.

Yes but if you are “soft throwing” the system is going to notice these drastic swings, flag you , and you will end up banned. This is not a maybe, this is literally how they end up catching smurfs and why all of them are low level accounts because as they repeat this action they end up banned and have to buy a new account.

A severe issue i have with comparing me to others who played the hero at my rank is the millions of variables between our games…

As dps i have been getting horrible tanks that just want to dps as of late and lost a ton of sr due to not being able to move forward untill i started playing aggro doom… comms and chat off and just focusing on picks and getting out to health packs then boom… climbed mid bronze to mid silver with ease and still climbing often gaining 35 to 45 sr a win and 20 a loss…

If i play hitscan with these trashbag tanks i lose a metric ton of sr and it’s not worth my time…

Obvious mean competitive is a joke and dumb as hell

So even if you carried a freaking team in low ranks you can still only get 0-4 SR what a big dumb system.

Reason i stopped playing competitive all together not going to waste my time and hard work for just 0-4 SR…

it should just be removed entirely imo.

if Blizz can pretend they understand your performance on a game by game basis and intervene on regular SR adjustments then they can also give players a scorecard that more accurately represents players’ impact, right?

essentially it’s a tacit admission that a ‘good’ performance according to the metrics this system tracks isn’t necessarily a winning performance.
if competitive solely used MMR this wouldn’t be an issue, but when you have a ladder/ranking system and you’re intervening in that process and attempting to account for a performance that isn’t necessarily going to help you climb the ladder it’s going to be detrimental to the quality of the matchmaking over the long term.

the ladder system should naturally take care of this, but won’t when it’s being undermined/overruled.

It is not like it matters. Few more sr barely makes any difference compared to the sr gain for winning the match.