What is wrong about the post?
The only thing I’m seeing is that relative SR is awarded by MMR differences within the SR window.
Edit: For example, looking at Tesla’s Bronze to GM series on Moira, by day 5 he has 40 wins to 5 losses for a ~89% winrate. His 41st win, instead of getting the average 25 SR, he got 42 (1763 -> 1805) i.e.–an extra 17 SR. His average performance is much higher than his rank and the system is trying to quickly converge on his real SR.
I have known this since season 5.
That’s the problem with “performance” based rewards. The game doesn’t reward you for making decisive plays, such as:
Walling off enemies, or booping them away to end overtime
Eliminating the last enemy
Holding a point alone until your team gets back
Making enemies waste time by chasing you around
Instead you get rewarded for just spamming damage and kills, even if that damage is healed, and those kills don’t result in a captured point. If you’re a clever flanker and wait silently for the right moment without spamming damage, you lose SR.
Minor influence over time should not be 300 points on a positive win ratio.
The most important indicator is winning. Again NEGATIVE 300 does not show most important indicator. The most important factor not shown is the mmr winning when you should losing when you should. That’s not free will.
This topic … riles me up. Dunno how we went from “win more games” to stats stats stats
How are wins and losses calculated? If you play 5 different heroes in a win, does it count as a win on 5 different heroes? Hypothetically, if every win you played 5 heroes, and every loss you played 1 hero, but in reality had a 50% win rate, would your win loss record look like 50 - 10 even though it should be 30 -30?
Back on my first time ever climbing I used to keep track of my own games manually. I climbed from 2450 to 3000 with a 105-98 record. Max win streak of 5. Max loss streak of 6. Then from 3000 to 3500 at 287-269 record. Max win streak 9, max loss streak 7. Nothing really interesting other than it takes a long time to climb.
For the third time, OP is comparing their career high to their current seasonal SR. Right now, that difference is 195 SR (peak 2035, current 1840).
That’s a difference of -195 SR over 913 games with 474 won, 423 losses, and 17 draws. This season, they are up 135 SR from the previous season end.
If we pretend that 195 difference is solely due to PBSR and not any difference in MMR (a ridiculous assumption), that’s an average of 4.7 SR. That represents the maximum possible PBSR average–which again, is a completely absurd assumption and must necessarily be less than that after accounting for losing games where OP’s team was MMR favored.
So yes, an average 4.7 SR is a minor factor in one’s SR gain/losses.
The most important factor to climbing is winning against equal or better opponents.
It also means that ‘minor’ influence can negate a winrate of 50% like the OP is stating. In his case, he needs to have a higher than 50% winrate to climb whereas in Diamond and higher you only need a winrate above 50% to climb, which means players below Diamond are expected to do something more than the higher ranked players are.
You realize that having a winrate of above 50% is all you need to climb in Diamond or above? This means he is being asked to contribute (relatively) more than a Diamond player (in their respective matches) in order to climb. In other words, he would be better off with no PBSR, since the algorithm intended to help players climb is the very thing preventing him from doing so.
that isnt necessarily true and the OP proved that. if you dont get many pbsr points and win only a little over 50% then you may not climb.
PBSR doesn’t exist in Diamond, sooooo, ready my post again.
The thread isn’t about diamond. Read the op again and stay on topic at all times
Of course I’m cherry picking I’m the one playing and you’re the one questioning my own play so I’m watching my own plays. As far as didn’t do well I am pretty confident that you already have that in the bag I felt no need to point out times I didn’t do well when I knew you’d be glad to point them out but very few times do people actually point out good points in a replay review. And if you mean that I’m cherry picking wins no the first replay I sent you was of a loss and it just so happens that the next seven games I’ve played have to spend wins because I’m so terrible and matchmaking is so great. Amazing how you place of faith in either my lack of skill or the games matchmaking ability and ignore big things such as high streaks and oddly high SR gains. For someone who sucks apparently.
I’m sure I made mistakes I’m fallible too but like I said I’m trying to look at the bigger picture here while I have made mistakes? are they more costly than the good things that I’ve done. Well mistakes do matter everyone at every rank makes mistakes what could have made better place, pointing out mistakes doesn’t really mean anything. What’s important is the value I’m bringing to the game, This is an about me performing perfectly, This is about whether or not I am performing below average for the 1.7k SR range, and if out of all my teammates I am the one hindering them more often than not.
But I’ve been tracking my SR gains lately. And well just another one of OverWatch’s temporal glitches because while apparently I am hard stuck in 1700 and I suck so bad I belong here I’ve won like my past seven games and actually managed to get pretty high SR on the most recent 2, because I consistently play good until I suddenly don’t
1684 +26
1710 +23
1733 +23
1756 +24
1780 +27
1807 +33
1840 +33
1873
The odd thing is that you would even question me Who’s actually giving you statistics As much as I can actually. However you completely wholeheartedly trust Blizzard that gives you no information on how this is calculated and leaves you just to guess and assume that what they say is how it is that it’s perfect that they didn’t miss any details and what they said, that didn’t require any details, and you’ll defend it regardless of you knowing almost nothing about it beyond a very basic outline of how it’s done.
If you say you know statistics then how the heck have I been performing below average (If we believe PBSR and your claims regarding my skill) and maintaining a positive win ratio After 884 games? Statistically how is this done? The only way I see this accomplished is if you place all your bets on me either Never solo queuing always getting carried or insane luck.
And here’s a question why do you think PBSR isn’t calculated for the game you’re playing? Why does it need to get an average of games to calculate your PBSR or if PNSR was in fact accurate? You usually gather numbers to get an average like this usually to make up for a lack of accuracy. After all what’s the point of giving you higher SR because your last three games were great and this game sucked and awarding a PBSR That just sounds like it’s more prone to error. It would be more accurate to award it for that game right now.
Also when ratio is not redundant because, averages. You would have to automatically assume that for some reason I’m playing more lower level games and I am playing higher level games. Averages over time would average out any discrepancies.
Clearly I don’t play those characters that often on rare occasion I do because somebody asked me to switch or something dumb. I maybe have 11k most this season but it wasn’t not so long ago I had 20k in one season. You really can’t tell because you don’t know how long the matches I’m playing are neither do you know how long I was playing that certain character for
I’ll tell you this though I’ve just been playing Brig and I got m 7 wins in a row. My last two games I’ve been 33 SR gained for match. Does this prove that I’m awesome? probably not most likely I’ll just have A losing streak in the future As the game is always done to me. High moments filled with low moments oddly never progressing though getting better
It is on topic to point out that the OP is being asked to perform better (relatively) than Diamond in order to climb. In other words, it’s harder to climb below Diamond and OP is providing proof of that.
Sure. My goal was to get you thinking critically about your gameplay. Figure out why you’re dying and then fix it. That’s step 1.
I don’t have faith in the match maker, I’m looking at the data and drawing supported conclusions. Winning and losing has some stochasticity. Streaks are expected.
Yes, they are costly. Your team might only need 2 extra seconds on the clock to win and your gameplay can be that difference. Every action and decision you make during the game determines if you win.
Yes, you are gaining more SR because you are winning against higher MMR opponents and you’re likely performing better. When that “suddenly I don’t win anymore” kicks in, you are encountering MMR opponents that are playing better. If you want to keep that SR, you have to improve and maintain that improvement by actively curbing bad gameplay habbits.
I’m not questioning the statistics you’ve provided. I am refuting the reasons you’ve provided for those statistics. Blizzard hasn’t provided the minutia of their match maker, but they have given us some details and there are publicly available conceptual frameworks for match makers.
We know how these match makers work, we just don’t know every single variable that Blizzard uses to calculate MMR.
Like I’ve said a few times now, you have to beat higher rated opponents to climb. If, on average, you’re beating more lower rated opponents, you will have a high win percentage but a low SR. Think of it like this: If a professional soccer team beats a recreational soccer team, should the professional soccer team be rewarded with a higher ranking?
That’s one explanation, but the more likely scenario is that you’re, on average, losing games to lower MMR opponenets that you should be winning. Meaning, that when you lose against these people, you lose more SR than you would gain by beating them.
Not necessarily, especially with only a 52% average winrate.
The current game is added to the PBSR statistics, but you won’t be rewarded with extra SR until you consistently perform at that level. Just like you shouldn’t be punished for 1 or 2 bad games.
Why is it that in Diamond and above, all you need is a 52% winrate to climb but below that you need more? That is another way of saying you have to work harder than a Diamond player (relavtively) to get the same result.
tl;dr - reward players for trying, punish those that don’t try, get to the appropriate rank faster. Diamond+, don’t want people milking the system for SR at high levels of play
From Scott Mercer on December 8th, 2017 (emphasis mine):
Removing Personal Performance Skill Rating Adjustments for Diamond tier players and above
Ok, this explanation is going to be pretty lengthy, so everyone grab some popcorn and buckle up.
When you win a competitive match of Overwatch, you gain Skill Rating, and when you lose a match you also lose Skill Rating. The amount that you gain or lose is calculated based on many different factors, and here’s a quick list of some of the most important ones:
- If you have a higher than 50% chance to win a match, you gain less for a win and lose more for a defeat. Conversely, if you were an underdog in a match than you gain more SR when you win and lose less SR when defeated.
- New players experience both higher gains and higher losses than players who have completed a lot of matches.
- You gain less SR for a win than you lose for a defeat as you more closely approach the system’s mathematical upper limit 5000 SR. (So at very high SRs you do need a greater than 50% win rate to keep your SR stable.)
There is also another factor in determining the SR change after completing a match, and that’s a measurement of how well you personally performed during the match. If you perform well than you gain more SR when you win, and lose less SR when defeated. The reverse is also true, so if you perform very poorly you gain less SR for a win and lose additional SR when defeated. The personal performance adjustments have been controversial amongst the community for quite some time, especially since the calculations for these adjustments are not at all transparent.
The adjustment does create a lot of positive system wide effects including rewarding players who make the effort to play well, punishing inactive players, and more quickly providing fairer matches for new players or those who decide to play a new hero or role. So we spent quite a lot of time examining data over multiple seasons, checking a lot of math, reading a LOT of community feedback, and then doing some deep soul searching about this. Especially at the higher levels of online competition where every point of SR matters, we want players to not be distracted and worry about how to optimize around the personal performance adjustment. They should just be trying to WIN. So after we get back from the holidays on January 2nd we’re going to turn off the personal performance SR adjustments for players in the Diamond skill tier and above.
In other words, winning isn’t enough below Diamond. He is pretty much admitting that you have to stat hunt to climb.
Nope. It means it’s easier to climb below Diamond if you’re a good player.
Hmmmm, in Diamond all you need is a winrat greater than 50%, below that you need more than that, thus it is not ‘easier’.
Wrong.
That means it’s more difficult to climb once you hit Diamond. You must rely on your teammates more than in lower ranks. So everyone complaining about “bad teammates” holding you back is in for a treat.
Not wrong, your points are irrelevant. Your opponents have the same skill as you do so the tier can be ignored. You are assuming that PBSR always makes it easier to climb but the OP shows it isn’t always true. Winning more than 50% of your games against equally skilled opponents should result in climbing.
Whatever this player is doing, he is helping to win games, but some hidden stat isn’t high enough for the MatchMaker’s liking, so he has to do MORE than win 50% of his games. PBSR is supposed to help move people to their proper rank, assuming they win, and he IS winning. He has found an edge case where the PBSR is working against Blizzards stated goal for the system.