If someone decides to join a comp match by themselves the matchmaking system will place them with five other players of similar competitive rank. If one player plays their hero (let’s say Reaper, for example) extremely well, but the team loses, the Reaper’s rank drops. However, Reaper’s solid play goes unrewarded. So now the reaper player enters another comp match and is matched up with another five players of slightly lower rank. The Reaper plays extremely well, however three or four of his teammates stink and once again the team loses. Reaper gains four medals, but once again his ranking drops. WHY IS THERE NOT A SYSTEM IN PLACE TO REWARD PLAYERS THAT PERFORM IN A SKILLED MANNER DURING COMP MATCHES? If a Reaper does well, but the team loses, the Reaper should be rewarded with being matched up against other skilled players that perform well with their hero’s. So if a Reaper performs well, but loses a match he should NOT receive an increased SR, however he should be matched up against higher skilled players for the next match. What this does is force the medal system to be overhauled to more accurately reflect the performance of any particular hero during a comp match. If that were to happen then great players that are matched up with under-performing teammates would still be recognized for solid play. These top performers are the people that deserve to be rewarded for their good play and continue to be matched up with other skilled players to guarantee high quality matches for everyone going forward. Right now a great player can be matched up with a crappy team and lose a handful of matches in one afternoon and see their ranking drop a few hundred points. What am I missing? Please enlighten me.
Performance based SR exists below Diamond. If you’re actually performing well every match, you’ll gain more than you lose. So even if you keep a 50% winrate, you should be averaging like +2 sr/game. And of course if you’re consistently overperforming, your winrate will be above 50% which will make the gains even larger.
So,the overperformers are pushed up the ladder. Once you hit Diamond, it’s all about making the plays that win games. So it’s more an impact game than stats game, although there does tend to be a correlation.
pbsr doesn’t work though, there’s no way to judge if the ways some one played contributed to winning or losing except to watch if they win or lose over a lot of games
I think the larger concern is the diminishing quality of the teammates you are matched with as you lose matches. (even though you play well)
The system is surprisingly good at knowing when someone is actually overperforming. On my gold alt I constantly gain 35-45 SR for wins and lose 15-20 for losses, meanwhile my true gold friend consistently gains and loses around 25.
Stats do affect it below diamond. But it’s also important to know that we are not sure what stats these are and they have nothing to do with golds. So for example if a team sucks and a reaper has for gold’s with 4elims, 2 obj elims, 2k dmg, and 1 min obj time. Those stats still suck, so he will get no benefit.
If the metric that determines how medals are rewarded needs to be changed then that should be discussed.
Blizzard has access to all these stats so I’m sure they see a correlation between a players stats and their win probability given those stats. So if a reaper for example had like 20k damage and 50+ elims, they probably see that more often than not, the reaper should win that game. If he loses, it was probably because the rest of the team under-performed and the system gives less of an sr penalty.
If someone is in the top 1% in almost all categories for a certain hero in that rank and are consistently losing many games, I’d think there was some strange phenomena taking place or possibly that hero is terrible in the current meta or they play a very niche hero. That’s why torb/sym one-tricks with nice looking stats but low winrates are able to climb out of low ranks but once they hit diamond and stats don’t matter they start to fall back down.
Yes, the current medals are horrible, just like the hero select used to be (and mostly still is). They are problems blizzard know exist but are low enough priority to basically never be addressed.
SR = grind, the more you play for SR rank the more you stay playing the game.
BLIZZ is hoping you will buy items due to them not having a mth to mth fee like WOW. SO they release items etc(mostly sh*t, but enough to keep people happy).
They are not interested in what SR you are or get to, or how hard or not hard it is to grind on whatever toon you play.
Its set up so you HAVE to grind, up and down through wins and loses, doesn’t matter how good you are at a certain toon, its all about grinding for that next SR level.
SR grinding as well as level grinding is BS, BLIZZ need to make it one or the other, not both.
MM is a broken system, blizz knows this and are still not willing to do anything about it. Getting placed in matches with bronze and diamonds is BS, supposed to be 1k sr between toon SR, not close to 2k. If they bought in SR ranked only matches if would be much better.
If a player is top 1% in all stats and still losing, that player needs to take a hard look at the plays he or she is making. If you are constantly losing yet have a good e/d, accuracy, aim, damage/min, etc. then you are not making the correct plays at crucial times, and you do not deserve wins.
That’s all thee is to it.
Allow me to tell you a tale of two smurfs.
I am not a Widow player, but I wanted to learn her so I made an account in plat on which I decided I would play Widow every game unless she was instalocked by someone else. A friend of mine is a 3.8k Widow/McCree main, and he too has a plat smurf account.
After about 50 games in plat, I was keeping a 53% winrate. I had nutty stats. 5.0+ e/d, 57% scoped acc, 25% scoped crit acc, etc, but my winrate was relatively low. Then I looked at my friend’s (now diamond) smurf - 79% winrate yet only a 3.5 k/d, 44% scoped acc, and 20% scoped crit acc, and all other stats were the same.
It really confirmed to me that stats can be misleading. It seemed like I was doing an amazing job on Widow, but when it came time for me to make the right plays, I didn’t have to skill to do it.
I could get on board with this. What you’re describing is the difference between a good and great player. Either way, you’re much more effective in a match than most of your teammates. And I’ll bet those teammates that let you down in a losing effort aren’t anywhere as effective as you were. Therefore you deserve to be recognized (based on your stats, whether the system is broken or not) for your positive effort and paired up with other effective players. Simple as that.
it probably works if you are paying exactly how its designed for, but theres lots of people who dont, so you get people way above or bellow there really rank, and people try to manipulate it, multiple of people have stated they ignore the objective and concentrate on farming elms if they think the game is lost for example
personally i think the experience of competitive competitive would be improved by removing incentives to ignore the objective, and being more honest about SR ie showing margin of error eg ±200 and marking SR’s with low certainty as provisional, like if someone only plays their placements it doesn’t mean much
I’m not saying someone who is top 1% will not climb. I’m not arguing for some voodoo forced winrate that is claimed. If someone has 90% accuracy with McCree, that tells me they are either the unluckiest person in the history of the world (very unlikely) or they only aim at tanks and never get kills. Or have terrible positioning. But if we were to plot a trend of accuracy vs wins for aim heroes like McCree, Widow or Soldier, I’m sure we’d see some kind of correlation there. It’s hard to imagine a GM widow player with 15% accuracy lol. Of coarse you also have to look at player positioning and target priority etc, but I think the general trend will show some pattern.I should have mentioned that but I forgot. I thought it was obvious.
what happens with people who can aim jsut fine but fire of most of their rounds while board waiting for something to happen? how are the algorithms going to account for that?
If for every GOLD medal obtained you potentially reduced the -sr from a loss by 5 pts with the max being it becoming neutral (0) then I think you would get more people try harding to make sure they maintain medals so they protect their SR in the event of a loss.
JMO and honestly wouldn’t hurt…
And yes supports like Moira/Brig and even Lucio/Zen can get 3-4 gold medals on a whack team, and when they do…they CERTAINLY shouldn’t be punished by their teammates performance…
You just defined my existence with the current competitive season.