This is the problem with trying to using proper statistical logic in a discussion with folks to haven’t studied it / don’t remember / etc. Your stats are fine, your logic is fine, and I believe you are revealing a detail about PBSR calculations, that is, you have an edge case that shows how, when Blizzard says PBSR has a limited affect, you likely have measured that limit. Your 800+ games are plenty representative of your skill vs the general population to make them a reasonable pool to use for your conclusion.
1 Like
I always thought that the end-of-match stat screen showed your numbers for the just ended match along with per-match averages but IT DOESN’T. When it says career average, if means average per 10 minutes, so you have to take into account the match duration to understand if you just performed better / worse than your average. It might be a small thing, but I really fault Bliz for not making that clearer on the screen.
Too bad Blizzard makes it hard to track this data over hundreds of games. They do it intentionally. It would literally take 5-10 lines of code to save your SR gain/loss history into some file or log somewhere; it’s a feature that has been requested for years, the fact it still hasn’t been implemented is evidence that Bliz doesn’t want to make it easy to collect these very statistics, which leads us to guess why.
2 Likes
Makes sense to me. You’re loosing more SR on losses and gaining less on wins because your performance is below that of the average player on that hero and map at your level.
As a support player I would recommend to try and die less. Even if it means leaving your team to feed on their own. Don’t get sucked in to lost fights. Try to fall back and regroup instead.
In games where I die once or twice I gain a ton more SR than in games where I die more often. Even if I get more healing or damage boost done in those games.
My stats on Mercy are average at best (9k heals / 1.5k boost / 4 deaths over 10). She’s my main and I’ve been steadily climbing from silver into mid gold this season with a 55% WR. Just by actively trying to die less often.
I agree with some of the other posts. I think it would be super useful to have proper stats somewhere, and an easy way to track your performance on a per hero basis compared to the average. I do suspect though that the “average” is actually way above the true average player at that rank because of the amount of alt accounts and smurfs in the game. The system probably worked quite well back in the day, but now, not so much.
2 Likes
Yes, you can absolutely have a wining percentage and still lose SR – this has been addressed in previous developer statements (and it remains true if those developers statements remain true). It just means, from Blizzard’s perspective, that you’re losing games you should be winning, or not performing “well” in games that are weighted in your favor.
3 Likes
I dont think they do it intentionaly since you can all write it down easily yourself. What would be the point then? They dont hide it.
But you are right, they should implement it just like many other features years ago.
1 Like
Are you suggesting that PBSR inaccurately captures the value that you bring to the team or not?
Yes, you can. It’s been demonstrated ad nauseam by numerous independent players over several years.
It has nothing to do with me. Either own your gameplay, take responsibility, and improve as a player or don’t. The only thing keeping you “stuck” is you. It’s not the system. It’s not your teammates. It’s not a conspiracy by Blizzard to single you out.
It is you.
As I’ve explained, you are losing games that you should be winning and when you do win games, your performance is average or below average. This is causing you to lose more SR than you gain. This means you are being handed some easy 40-60 games where your team is favored and you’re losing. Most often, when you do win a favored or even game, you aren’t performing above average.
It doesn’t show error, it shows you are where you belong. Your SR has floated around 1700-1800 for 7 seasons. As of this month, your win percentage on XBL for Brig is lower than the average for Silver (58.57%).
It does and I’ve clearly explained it to you. You get different amounts of SR per game depending on the enemy MMR and your performance. Again, the analogy is:
A chess grandmaster at 2750 elo claps an untitled opponent at 1300 elo. The GM gets 1 point. The GM could play against this person for 100 games and have a 99% winrate. However, if the untitled opponent at 1300 beats the 2750 GM, the untitled gets 100 points while the GM loses 150 points. The GM has lost elo despite having a 99% winrate.
In some of your games, you are supposed to be the GM but you are losing. In other games where you win against an equal opponent, let’s say two untitled 1300 players, you are showing average performance for that elo. That means you get rewarded for an average performance.
1 Like
Yes, it does even out. The number of asymmetric games that you play at 60-40 is a small fraction of the games that you will play at or near 50-50. This is in comparison to an “open elo” system where you would be randomly paired with highly skilled players more often and get absolutely stomped.
If you think that’s fair or fun, that is certainly your prerogative.
It’s exactly commensurate with what the devs have told us about MMR. If you lose games you should win, you lose more SR. If you win games you should win, you get slightly less SR. If you win games you should lose, you get more SR. If you win games that are a coin toss, you get an average amount of SR. The SR you gain can be increased by performing above average.
Having a high winrate does not mean you deserve a higher rank. This is very simple logic and it applies to all competitive sports. It matters who you are winning against, not how many times you are winning.
OP represents people that are overestimating their skill and think the system is holding them back. If you’re stuck in an elo, that’s where you belong until you improve. A Diamond player is never going to be “stuck” in Silver. A GM is never going to be stuck in Plat. They will certainly experience pepega teammates, but they can and do climb out.
Again, OP is comparing a single seasonal peak to his current SR. His current SR is nearly identical to where his past 7 seasons have ended at 1700-1800 SR. That is OP’s true rank even if they had a lucky winstreak that took them to 2k. If they belonged in Gold, they would have stayed there.
1 Like
that’s a mot lie you can ever belive in yourself
immagine ana sleeping nano genji on overtime
this is what a stat cant track.
rein or zarya block enemy ult
dva sigma eats gravs.
stats are BS and just part of rigged system
2 Likes
Worded a bit differently than your accusation of me believing myself to bring more value than high leveled players. But to answer you question. I more than believe it, statistically with the win ratio maintained is a bit above average meaning that an accurate PBSR would asses me as at least average, neither boosting nor gimping my Sr in then end. However PBSR does gimp means that PBSR is not accurately valueing my PBSR.
Yeah but the manner which you were using “carry” wasnt to argue that it “, potentially” can happen. You straight up use it as to attempt to invalidate all my higher rank ges. All you’ve done is waited to an opportunity to take something I said out of context to detract from your original accusation which is completely wrong and on assumption. you began with a sleezy accusation, so I guess I shouldn’t be surprised you’d also attempt to end it with a sleezy win.
It has everything to do with you. See here’s the ironic part, Is you assume that I’m playing bad but, but you have no actual proof for this nothing to back up your statement. You only go by an obvious logic that if you’re not ranking up then you’re not skilled enough to rank up. This logic is true only in an environment where win ratio is the only thing that determines rank. Of course this logic doesn’t take it to account that there’s potentially something influencing your rank progression. Not only do we know there’s an influencing factor (PBSR) but this very post shows that the influencing factor invalidates win ratio because statistically I’m not playing bad I’m winning more games than I’m losing. It’s literally not me. Ironically what makes logic redundant is what’s made my SR games redundant.
As you can see clearly that the PBSR systems keeping me down. A simple consideration for middle school level math and averages will tell you that there is something inherently wrong here. If this isn’t about you then Why do you continue defending a a logical contradiction (having low PBSR indicative of bad play while maintaining positive wind ratio indicative of good play) It makes no sense to attempt to defend that, what are the reason can there be aside from your own inability to let go because of some weird pride.
You realize this is an assumption right? You’re trying to get me to accept as a fact that it is me, by using an assumption as the basis of your argument? That’s delusional
I’m going to assume that you just don’t understand what I mean by they’re being an error or a flaw or a discrepancy. You need to take time and reevaluate what I’ve said. Of course it’s going to show where I “belong” according to a contradictory system for whatever that’s worth
Literally, Your inability or unwillingness to actually take time to understand what I’m saying is going to make this difficult. I’m going to have to re-explain everything in Just to readjustice weird perspective that you have on what exactly The point of this proof is and what it shows and it’s implications are. There is a clear contradiction that I pointed out and I’m not sure you’re actually putting any effort into thinking about what that means. That you don’t have to think about what it means you never had to in the first place but if you actually want to argue when you’re arguing you it’s pretty much necessary for you to do so or we’re going to waste a lot of time.
Except that you’re trying to explain the effects of PBSR By using win ratios & SR discrepancies when PBSR is not calculated that way. Most importantly you’re trying to argue an assumption is truth. I’ll give you an example of why it matters.
Here’s my assumption equally valid as assumptions go. A 2000SR team claps another 2000SR team. The team gets 25 points. They lose to team of equal SR opponents They lose 25 points. They news again to another team of equal SR, but the player only loses 18 SR. Why? Well the reason why is PBSR. Can we tell why just based off the SR and the wind ratio why his PBSR is protecting them, no. Does this guarantee that PBSR is accurate, no. Would it be strange if they continue to lose and PBS are continue to protect his players SR? Yes, because continue with losses show that the player is not good enough to win, yet continue protection because of PBSR are to protect them from losses they do not deserve. This is only okay and logically only supposed to happen for a very small window of games but after a thousand games this is ridiculous.
2 Likes
I’m referring to what you said here:
You’re arguing that your playstyle on Brig contributes more value. I’m telling you that high-elo players CC and assist while also getting better statistics. Objectively, they are getting more value by adding more to what you’re already doing.
You said you can win games in low Plat if you’re duo’d with your Plat/Diamond friend. You are literally saying that your friend carried you in a lower SR than he/she regularly plays in. That does not at all indicate that you are a Plat player. There’s is nothing " sleazy" about my comments. You are offended because you think your skill level is higher, but it isn’t.
I literally watched your replay. I have looked at your stats on your profile. I looked at your stats on Overbuff. I looked at the win/loss stats you posted in this thread.
I have demonstrable proof that you are playing at the SR you’ve been placed in.
You is that “something.”
Again, who you are winning against is the biggest determining factor. If you’re consistently winning against 800 SR Bronze players, you are not demonstrating that your rank should be higher. If you’re consistently winning against 2700 players, you are demonstrating that you deserve a high rank. It’s really that simple.
Yes, it is. If you performed like a Gold player, you would be in Gold. But you do not perform like a Gold player. You can perform like a Gold player if you’re willing to accept responsibility and work on improving.
Can you be self critical enough to point out your mistakes during that Hanamura game? Can you articulate what the better play would have been? Do you understand that every small action you execute improperly costs you games? Are you willing to fix your mistakes or do you want to continuously blame the system?
It’s not an assumption. If your SR is going down and your winrate is positive, you are necessarily losing games that you should be winning. If your SR is going up while your winrate is below 50%, you are necessarily winning games you shouldn’t be winning.
You can’t make this stuff up. This is how ALL elo systems work–with or without performance-based gains.
You would get stomped as a solo queue in GM, Masters, Diamond, Plat, and Gold. There is nothing contradictory about that. If you think that’s not true, feel free to 1v1 people at those elos in custom games or find some scrims you can piggy back on. You are severely overestimating your gameplay.
You can improve, but not if you deny what your current skill is.
PBSR is minor factor in your SR gains and losses. The most important factor is who you are winning against and your consistency. PBSR is designed, and works, to quickly acclimate players to their real rank rather than making it a frozen molasses grindfest.
-
Teams don’t lose or gain SR. Individuals gain or lose SR based on their individual MMR/skill in the matchup. SR does not display that skill matchup as it’s a proxy for average team MMR.
-
If the player loses 18 SR, it means they were playing against higher MMR opponents and they were expected to lose. If the MMR was equal and the player only loses 18 SR, it means that player’s PBSR is high for that SR bracket i.e.–that player should not be deranking due to RNG as their consistent performance indicates they should be higher.
Again, if you think PBSR is unfairly working against your “skill,” climb to Diamond and stop worrying about it.
3 Likes
PBSR isn’t gimping you at all, your stats are below average for the rank you are in, if they we’re average, or above, you would be breaking even/gaining SR, as I said above, I was running a negative win rate, but still gaining SR, due to my stats and PBSR trying to push me up.
This is perfect.
2 Likes
I kinda made a thread for this, but seriously. Is PBSR real? How does it work?
Okay, a 50% win ratio means that the players winning as much as he’s losing therefore he probably deserves to be at that rank, making him the average player of average skill for that SR range. A player with below 50% when ratio is below average because they are losing more games than they win a player above it is above average because they’re winning more than they lose. Because of the many other factors you have to consider like your teammates. Your win ratio is not a reliable means to determine your skill. It’s recommended you play at least a hundred games so that you have a larger sample size and your wind ratio starts to reflect your personal skill. The more games played the more accurate win ratio is at determining your skill, It only grows ever more accurate. Now I have played 884 games and maintained a positive win ratio, throughout every season except the last one which took a little bit of a dive but even if you add up all the games and every season overall I still have a positive win ratio. Now it doesn’t take much to just look at the number of games I played and think about when ratios and its relation to your personal skills to determine the likelihood of me not being at least average in skill is very low. It is basically almost guaranteed by my win ratio that I am at least average but of course there’s always that 1% chance but there are no guarantees in what we’re arguing here
PBSR will adjust your SR gains depending on whether it believes that you are overperforming or underperforming. If you had average PBSR then PBSR will neither adjust your SR anything above or below average. As we’ve already established based on the win ratio It’s almost certain that I’m not below average. However the only thing that can be causing this is PBSR
So here’s the very likely situation. I am at least average, but my PBSR is placing me at below average. We do not know how previous art determines skill, personally I’ve done video game programming I can pretty much guess on how they’re doing it but but for an AI system to grasp an abstract idea like value to a degree that we can would require tons of work and you can easily get away with creating a system that is good enough or mostly works enough so that you likely won’t notice any flaws if it did have them.
Now I try my best to explain this to you with information that’s present and that we can actually use to educated assumptions or guesses and in some cases very likely scenarios. If you’re seriously trying to make the argument that I’m wrong, I would like to hear the reasoning behind what you say that isn’t just what you think or assume, and that would serve to rebuttal the points I’ve made in regards to just the wind ratio statistics and the incongruency with PBSR.
2 Likes
Your not being forced to win your just getting lucky because of the system and the people in it. Imagine this say you are on a team of diamonds and your a plat, and you play hard you do all you can but your getting wrecked and not doing good but your team still wins.
That is something that’s not forced you just benefited from the system and there was an equal chance you could have lost that game. There is no force win and loss only a forced situation the rest is up to other variables
1 Like
Nobody knows for sure but it’s based on things like Elims and deaths accuracy but it’s also character specific too from what I hear
Have you tried doing this? I’ve been taking screenshots of that screen and it’s ok but you can only play one hero a game, otherwise there are too many to easily capture without already recording the video. The screen for each hero goes by too quickly to write down the numbers. Play one hero, you get one screen; play two heros and you get 3 screens, one for each hero and then a third overall screen.
In addition to this, there is the small problem that in each box are two numbers, the value for the current game and underneath a ‘career average’ but what it doesn’t tell you is that the career average is per 10 minutes while the current game number is just the total so you can’t directly compare them to judge your performance.
It’s reasons like this that make me suspicious that Bliz actively prevents players from being able to garner enough information about their play to reverse-engineer the PBSR calculation. They have said they don’t want players to stat-hunt but I have news for them, players already are; in fact I’d be willing to bet that the boosters and smurfs have a good handle on how it ‘actually’ works and it explains why you’ll get games with a Mercy pocketing a Hanzo and won’t heal anyone else. While that could be a friend boosting a buddy, it could also be a sign that ranking up is more about maximizing some stat than winrate (until Diamond, of course).
While I appreciate your ‘by your own bootstraps’ attitude, you need to understand that people are willing to accept their own responsibility to an extent but get really put off when they feel like there is an artificial current they are working against.
When you say “it is not the system”, you can not know that for certain. You can have a personal belief, which is fine, but you can’t make a blanket statement. There exist a group of people who are adversely affected by the teammates generated by the MatchMaker™. If you have a carry potential of 1% but are expected to contribute 2%, your net results will be negative, simply because of the expectation of the MatchMaker™.
1 Like
Elo is an artificial system. The only barrier to climbing is the player and most people would rather defer to an external factor for not making personal improvements.
We do know that for certain. It has been demonstrated over and over by numerous independent players of all skill levels. If the system worked against the player, we would expect a large portion of players to get hard struck below their skill level. That doesn’t happen. The age of the account doesn’t matter–they always climb to their appropriate elo. That means the ability to climb isn’t being limited by the system, it is solely contingent on the player’s ability.
Everyone is adversely affected by some teammates some of the time. Not every game is winnable due to random factors and those games stand out in our minds. However, these stand-out games represent a small minority of the total games we play even though they are psychologically taxing.
Which, by definition, means the player does not posses the required skill level to advance in rank. If you can’t pull your weight at higher ranks, you don’t deserve to be there.
Edit: Adding to this-
Sort of, you can compare the total to your average–although I understand this requires some mental math if you only spent a 1 minute or 2 on the hero.
PBSR correlates roughly to the amount of time spent “on fire” during the match. Outside of that, the average per 10 stats are also a good metric. For a hero like Mercy that would entail:
-Healing
-Damage boosted
-Deaths
-Resurrections
-Number of assists
Doing your best to maximize those stats each match will intrinsically correlate to winning more games.
3 Likes
High and mighty talk is cheap, assumptions are easy, bring you own statistics to prove that high ELO players are doing what they are doing plus bring me the proof that they are doing the exact same things I’m doing while getting those statistics that aren’t measured by statistics or PBSR (something I doubt you could actually prove and I would just have to take your word for it, which you wouldn’t do for me so why would I do for you) bring your own Brig replays. if you by chance have an alt in silver preferably bring that brig replay because brig play high ELO requires far less then low ELO
You still have not explaied to me the contradiction between the win ratio being above average which would mean that im not below average, and if I’m not below average then why/how can pbsr rate me below average? It would be stupid of you to try because you know as well as I do that the chances of you being right are slim to none. The only argument you could possibly make is that for some reason I specifically paired with people throughout my whole comp career that carried me and were willing to carry me and that that’s the only way I’m winning these games that I’m not solo queuing alone.
No way for going to talk in terms of integrity If you’re going to accuse me of having no integrity, what’s to say you have any. To say you even deserve the rank you are at how would I know You’re not a carried or boosted? A rank actually proves nothing considering you could sell and buy accounts and there’s boosters out there. So your position of power is questionable.
Unless you bring equal amounts of proof and replays in the same positions that I’ve been in that is the only way that I can know that you are truly correct. If you can’t then you can’t argue, I’m at a handicap because you don’t provide the same level of proof to back up your words, I just hide behind your words.
Clearly rank doesn’t mean you understand everything at the very least you don’t understand the statistical and improbability the PBSR conundrum.
I’m watching you try to circumvent your original point message which was a flat out assumption, completely wrong cannot be said in truth, and watching it turn into this some complex argument because you decided you want to go on your little side points to get your win somehow. It sleezy from the beginning, because of some believed position of power and ego, and it’s sleazy now although you’re arguing is more articulate the fact that you’re still trying to argue something which isn’t far from the original points is the sleazy part of it.
But you’re the pro, do You believe that I contributed below average value In the first two replays that I deserve to get gimped, and contributed above average value in the third that I deserve to get boosted? I
Today FPEBDM +23
1 day ago 00B6S4 + 23
1 day ago 3VVEYX + 26
Here’s my last 3 games on support, not sure where they are mistakes but I’m looking at the overall picture here.
The first one I think I did good on, I literally held the cart many times I was playing around the cart very well using cover, because apparently much of my team didn’t know how to deal with Sebastian to just play at an angle so you don’t get hit, I also harassed/killed their hog, apparently My performance was below average 23 SR which from what I know is below the 25 you’re supposed to get, So apparently I performed below average.
The second game with Moira I also think I performed well, which is surprising because I don’t win many games with more because I barely play her, but still I end up having silver damage that game and have the most pics and my healing wasn’t great but my offensive was and I think it was worth it and we did win. So clearly I brought value in the offensive is PBS are going to consider that? Or will it just see that I did very bad healing compared to other players in my rank and punish me for that? Either the way I gained 23 SR
The third game, I don’t actually think I did as well as the other two games, our team barely won that match and part of reason I don’t think I did well was because my team wasn’t doing good which in turn gimps my performance. Which brings up another good question, can the performance of your teammates affect your performance and cause you to potentially get higher or lower PBSR ratings making PBS or not completely about you either? If PBSR is calculated based on the averages of a rig player within my SR range, Anyway for this game I gained 26 SR which is a little bit above average.
One thing that is kind of clear is I don’t play the traditional support the last 2 games. In the most recent game with Brig I play a tanky brig and I get value out of it that way. However what did the in-game AI expect me to do? Does the game value my tanking as a support as much as it would value traditional support roles values? In this game I found it necessary to do so, and it did have value If I’m compared to the masses and the traditional role of support, by that standard I would look bad.
You seem to believe that PBSR is capable of understanding the game just like a human being, I’m telling you ask a game programmer, it cannot, You would need an AI capable of beating the turing test expect that kind advance analysis, to this day no AI has beat turing test and these are a eyes that were built specifically for that purpose, solely dedicated to the AI and not as simply part of the game.
1 Like
It’s at least role specific, I hope… Otherwise my 0 Elims on Mercy would be counted as a bad thing lol. I do find it strange that the easiest way to get her on fire is to pull out the blaster though 
This violates what myst says, simple as that.