Its not even about facing harder opponents, its about getting less awful teammates. Me and my friends are already matching top tier Onyx players, and pros. The game assumes we are good enough to compete with people of that caliber so your theory is flawed.
You don’t understand what anyone means when they say KD. You think KD is just people getting 1 kill and hiding. You’re very stubborn about this concept.
Lol no it doesn’t. At all. If proves me right in fact. You keep trying to justify a wildly mismatched win % by saying its not the matchmaker setting one team up to lose. But the fact is that’s exactly what’s happening.
Well that’s just a flat out lie.
Because of in-game variables. The fact remains the MATCHMAKER set up a game that one side was SUPPOSED to lose. You’re still being stubborn about this.
Yes, glad you finally agree
Yes, it averages the ranks in the ranked playlist. Seems fairly straightforward if you ask me.
Here’s a thought experiment: What about cheaters? The gold player in this match is a cheater I encountered tonight in social. I made a video about them using wallhacks and aimbots that proves it, (https://www.youtube.com/watch?v=rFrvCohGBh4) but that’s neither here nor there. Why is the matchmaker giving them matches that are so lopsided? Could it be the entire system is flawed? https://halotracker.com/halo-infinite/match/851bbc06-18cd-4091-8834-0a816ef63bb8
First off. It’s not a very accurate way of doing it.
Early in the season your CSR may be 2 to 3 tiers away from your MMR. Later in the season when they’re close they may be more relevant.
And you are completely ignoring weightings for squads etc. Which we know are very important in balancing the sides.
And finally you are relying on potentially dodgy stats from the get go. Up until a day or so ago HaloTracker was showing everyone’s CSR from the Season before. So their “guess” for win % for the last week or two are munted before we even start.
That’s the problem in a nutshell. MMR should not be the relevant factor in forming teams in a ranked playlist. The Playlist’s specific rank, I.e. the CSR, should be how the game matches people.
It’s the entire core of what’s wrong with ranked, and why you’ll see wildly spread out ranks sometimes, and mismatched battles.
MMR and SBMM in general pollute the system with frustrating circumstances. That’s why not one person who plays the game regularly has ever said anything positive about it
Your MMR is specifically designed to do the ranking.
Because it’s a probability curve it makes all the maths work for averaging teams, calculating the result probabilities, and working out how much rank to allocate against the result.
The system literally works as well as it does because it uses the extra information hidden in those curves.
For the player they are pretty much the same. Or at least behave the same.
Your CSR is hovering around the left hand edge of your MMR curve. Three standard deviations below the mean of the curve.
But your CSR is just a number. It doesn’t have all the other bits of information about your rank probabilities that make ranking, and matchmaking, work.
And for player “comfort” your CSR doesn’t vary with form as much as your MMR. It may swing up and down a bit more than we think. So, by limiting your CSR to 15 points per game you average out this oscillation and player’s get a smoother ranking journey. Mean while your MMR can match according to how you are going on that particular day.
The wildly spread out ranks are due to Squads and lower population. They are match maker issues. It has nothing to do with ranking or TrueSkill2.
The mis-matched battles are just what happens with players being put on teams. Sometimes they gel and sometimes they don’t. The mis-matches are amplified in objective games - and can blow out when you have a wider range of ranks. That’s frustrating, but again, not the fault of the ranking system.
The main frustration comes from lack of communication from 343.
They could be communicating more. Have better explanations both here and in-game. And there is no reason we shouldn’t be able to look up graphs of our MMR and/or KPM on Waypoint. That would go a long way to educating the masses.
Update: nothing actually changed. Still rigged and unfair. I just got lucky as suspected. They’re making me lose many matches in a row because I won a lot on that day. Sorry to get your hopes up guys.
If you are going on a bit of a losing streak it’s usually because;
a) bad luck
b) you are just having an off day - and probably making it worse by trying to over compensate (sometimes it’s better to just step away and calm down).
c) you have just spent some time playing with mates and the team-work advantage you had isn’t carrying over to solo and your MMR is taking it’s own sweet time to readjust.
d) you genuinely did rank up in that purple patch - but now, for whatever reason, you are struggling big time to hold that rank. The system is trying to give you a chance to prove yourself - but you aren’t quite holding it.
Or some combination of all of the above.
The algorithm treats every game the same way. You are not being set up to lose. But your form will fluctuate naturally. Game to game. Day to day. Week to week. And your MMR will lag a bit behind that. And if it’s over a run of games then the lag will be over a run of games.
I think the key is to appreciate that you at least earnt a run of harder games. That’s a good sign that you are on the precipice of ranking up. It’s up to you know to work out what you need to do to take that next step.
Nobody is blaming the ranking system itself, they’re blaming the matchmaking system, which is terrible and directly contributed to killing the game. Taking a system designed to force you to lose games to meet a win/loss quota and putting it into a ranking setting is backwards in every way.
MMR is dumb and doesn’t mean anything. Nothing of value is gained or lost from it existing or not existing. They don’t even let you see it so it’s all built on a lie. The game has never been good at forming even matches, as almost none of the games wind up being close to a 50:50 win chance.
Using a broken system to then influence your rank is just twisting the knife.
You may be the only person on earth who likes the system and thinks it works well, and it’s literally just because you sit around making spreadsheets instead of playing competitive matches so it has nothing to do with the actual game itself.
Most people stopped playing this game because it’s bad, not because they’re ignorant.
The game is dying because of how bad the matchmaking is, among a host of other problems.
You adamantly defending a system nobody likes is just baffling.
None of what you mentioned included the game being a broken buggy dumpster fire.
“Nice winning streak you got there, would be a shame if someone were to desync it”
The ranked match making system and the way the individual CSR (Competitive Skill Ranking) is calculated works against a player.
The biggest factor of how many CSR points are added or subtracted from the individual CSR depends on the Team performance - win / lose the match.
That itself is a big contradiction. It doesnt help if you got positive k/d and played also the objective. If the team lost the match, you will be punished. Even if you carried the match.
The last 25 ranked matches i had a lot team mates that had really bad match k/d between 0.2 - 0.5. One third of my matches were on ping above 120ms. Still managed to end with positive k/d.
My last 25 ranked matches 9 win / 16 loss. Is a win rate of 36%. I got mostly positve k/d each match and played the objective. Still went down from Gold 5 to Gold 4. Last season i was platin 5.
I have read all posts in this topic and a lot of other topics discussing this subject. And im not the only one that is experiencig such “bad luck” situations. Does it really needs more evidence that this match making system sound great on paper but in reality, it doesnt work. If this system would work how it was meant to work, then losing streaks like 9 win / 16 loss shouldnt happen.
Because of the 16 lost matches, i lost also a lot of CSR points, that i cant regain when im lucky enough to get good team mates and win 2 matches.
Im absolute sure, that this contributes alot, that a lot player left ranked and in the end, halo infinite.
Yep. But it’s the same for both teams. So it’s not a factor for rankings.
I imagine someone like Lucid would beat me by similar margins despite the ping.
Yeah, but you’d have to have an idiot to believe that. I suspect you are trolling.
It’s not a contradiction. It’s a team game. The team performance is what leads to the win/loss.
If you got a positive KD and played the objective - that is good team play. The team is rewarded.
If you lose - the team is punished. Not just you.
The best player on the losing team is still the loser. That’s how team games work.
If you genuinely carried though, in terms of KPM and DPM, you can minimise the amount of MMR you lose. In rare cases you may even raise it despite your CSR going down.
Ouch.
But I assume the other team was dealing with the same problem.
You are doing better than me.
Well played. But to make sure you are “pulling your weight” you need to compare your expected kills and deaths on Waypoint. Sometimes your KD gives you a false sense of security.
It’s a tricky stat. One more win = 10 win / 15 loss and a win rate of 40% - which is “acceptable”. A case of too small a sample size to get upset about.
Again, your contribution is defined by KPM and DPM. You may have come out with a KD above one - but you may have under (or over) performed based on your skill vs your team-mates skill. Check Waypoint to get your bearings.
Need context. That could by a drop of 1 point!
Your MMR may still be higher. Especially if you have been carrying hard in the losses. How much does your CSR go up with a win. If it’s by 10 to 15 then your MMR may still be Platinum.
And this season the reality is going to be wider ranges of ranks. You will need to adjust the way you play to account for lower skilled team mates. Those who do will have a happier time ranking up.
A bit biased - if you are having a bit of bad luck or having trouble adjusting then you are more likely to be vocal about it.
The match making is obviously struggling with the population. It would much rather games with tighter ranks - but what can you do?
As for the ranking - it will still do it’s job and rank everyone relative to each other.
And that may be a key to why people are struggling to match their ranks from previous seasons. If you only let Onyx players from one season through to the next the ranking system would start to resort them Bronze to Onyx. It takes the population at hand, sorts them into an order, and spreads them across the x-scale (which is arbitrary).
It’s very likely that we are working with a different population of players from last season.
Again. Small sample. 10 & 15 = 40% which is ok. 11 & 14 = 44% which is definitely acceptable.
1-2 results shouldn’t be the tipping point.
How much is a lot?
You said Gold 5 to Gold 4. That’s somewhere between 1 and 50 points. So not a lot.
It’s all relative. You’ve lost two ranks in Halo 3 and only one rank in Halo 2.
And again - the key is your MMR. If you are really “carrying” in those losses your MMR will have fallen at a much slower rate. You will rank back up quickly with a run of form.
I would have thought content and performance issues way more than the ranking system.
As people dropped off and the population dropped to the point that the match maker lost control of rank spread - then yes, that may have been the last straw for some people.
But that’s not the fault of the ranking system. Or the match maker.
It is a contradiction, because when i got positve k/d and played the objective like 17 base capture (for exampe strongholds), several base defends, but all other team mates went negative and/or had less than 10 base capture. We lose the match. I get punished, even if i tried to carry the team. But its like you said its a team game. But i get punished, i am losing CSR points because the team did not perform.
Now you say: ok, the team did not perform as expected. But, what if the team performed as expected? What if the other team was expected to win? Im still getting punished.
And whats the point behind MMR? I dont see it, i see the CSR ranking going down. Telling me, that i performed as bad as my team mates. But i performed better. Once more: the team lost the match. The team consist of 4 player with individual skills and match outcome. Its one thing to track the W/L statistic on my matches, but its a whole other thing to calculate the individual skill of a player an display it as personal CSR.
We can discuss it the whole day. By the way, its alway a pleasure to discuss this subject with you.
The team performance and an individual skill rating like CSR is not the same. Track W/L, track damaged dealt/got, k/d whatever. The individual performance of match should be the biggest factor to calculate my skill rating. Instead im depending on the team mates that the match making system has chosen.
I keep always one eye on the Waypoint statistics. In 80% of my last matches i performed better than expected. Got more kills than expected and a few time less deaths as expected (but still more kills as expected).
Lately i played a ranked match +130ms ping. Got 23 kills 17 deaths, lost the match, got t-bagged a lot in this match. I know why, Got 5 doubles and 1 triple kill. It was extreme difficult to kill me. The other team was constantly meleeing the air. There is only 1 explanation, i lagged so much that the other player struggled to see my real position. I even survived 2 rocket shot in a row, had 0 shield and still managed to escape the fight in this match.
Kind of frustating for the other team, even if we lost. I clearly overperformed in this match, not because of my skills, it was because of the “lag shield”.
The ranking system just does the best it can to infer your individual skill from the debris.
The best player in the losing team of the Superbowl doesn’t get a ring. But the worst player on the winning team does!
Sometimes it feels personal. But the ranking averages everything out.
Think of your ranking as a team performance rank as opposed to your individual skill rank. There are plenty of people who do much better in FFA than 4v4. There are those who do worse. There are those who excel alongside a certain team mate (complimentary). And I’ve actually seen examples of friends who actually have a poor team dynamic and tend to do worse playing together.
Your team lost.
It’s frustrating. But everyone is in the same boat.
You can pick out the games where you lost and other’s let you down.
But overall everyone gets the same deal.
I’m sure there have been plenty of games where you’ve won and thought to yourself that it was despite your efforts.
That’s an inherent part of the way the CSR behaves. To some degree that was just a design choice by 343. Personally I would lower the scale and just have it not change at all (unless you trip into the next bracket).
If the team performed as expected you would be very unlucky to lose.
If the other team was expected to win it’s taken into account. You will only lose a little bit of CSR (unless it’s way above your MMR). Your MMR won’t change at all. You shouldn’t be disappointed about the loss - it doesn’t mean much - what was lost was the opportunity to cause an upset and push your MMR up.
The point about the MMR is match making.
It needs to do it’s bell curve math’s in the background to work out your rank. It’s the basis of TrueSkill2. It’s literally why it ranks you so quickly and accurately. And accurate ranks are the keystone for the match maker.
This is on 343.
One the scale of the CSR is pointless. People just get upset about every point lost. In the days of 1-50 people happily played for days without seeing a rank change. Now we get hung up over a couple of points at the end of every match.
Secondly - poor communication. People still think personal performance is the lynch pin of the change. It’s not. It’s the skill gap to your opponent and the relative position of your MMR to your CSR. These are different for each player - hence the different change in CSR.
Thirdly - given point two - why would you show everyone’s CSR change. Of course it’s going to -yoink- people off.
Remember the individual skill of the player is not reflected in the CSR change for that match.
Thank you. It’s a lonely job sometimes.
Microsoft tried that. Every metric you can think of is incorporated into the win. Except KPM and DPM. Those other things are important - but they are deeply ingrained into the result. Do them well (as a team) and you win. Do them badly (as a team) and you lose. The individual component isn’t helpful.
Which is the way it’s always been.
It’s just amplified at the moment because of the wider skill gaps (low population). And especially in Objective games, when it falls apart it does so spectacularly. You’ll notice that Slayer games don’t tend to have the same blow outs.
Context matters so much man. I’ve seen 1 player destroy his own teams set up over and over in a strong holds match by trying to force a tripple cap when the other team mates have the enemies locked to 1 spawn point…then of course they do bad when 1 guys making the enemies spawn right by them…or even right by power ups at times on some maps. That player can have great stats or bad stats…for the win it just doesn’t matter. The context of the match and whats going on matter so much. Theres a time and a place for a tripple cap…but man I’ve seen people run in circles to capture and lose capture points over and over like…wth are they doing?..you need 2 zones to get points…the points never start going up if people keep running from the zone to do the same exact thing they could have done from a position near a zone they can easily defend from the spawners…
In an actual team…where you have a party…all are talking and communicating what they’re about to try and do…it makes sense. Pushing a tripple cap, calling out spawns etc…
In a solo q game…no way…doubtful people are on their mics…doubtful that they have any idea what their team mate is doing and only aware of that 1 player messing the spawns up each time they try and get a set up going already after the fact that it happened. That players stats become meaningless to the game regardless how hiigh they are at that point.
Changing play style is basically key when you play alone. Match making is exacerbated (on your end usually) because its RNG as to how you will compliment each others play style. It will make the rankings seem much less accurate as well because the outcome of the matches are less in your control…because you’re starting and ending the match with this sensation…that its resting on your shoulders alone. Its also possible for each individual of the team to even feel that way.
“I had the better stats. I played the objective, everyone else played bad. Wth are my team mates doing?”
a sensation of feeling alone in the entire effort. That sounds extremely frustrating.
Its a nightmare. Get some friends if you can man. Make the experience more manageable and you’ll have way more enjoyment irrelevant of loss% when you’re at least playing with some friends. Or…at least the losses may not hurt as much or seem as frustrating.
Theres metrics used that can be played for to increase your awards for winning…but…imho, its much better to play for the win in mind over stats, because if you’re constantly losing you won’t see the fruits of your labor…it will be within the hidden stat (mmr going up) during a loss due to your weighting stats improving, but your visible stat going down (csr ranking) due to too many consecutive losses. A player naturally pulls better stats the more they win in most instances until you reach that plateau…then…the weightings can be played around at that point in a much more manageable way since winning the team effort has become stronger and now is basically only a point of increasing your own efficiency without sacrificing the team skill.
No it isn’t, not by a long shot. It’s completely random whether your bullets will work one game or not. Also there’s a oft unspoken handicap for playing on pc vs the Xbox series s/x. When the platform you’re on makes an active difference in how well the game is going to work, that’s important.
You don’t know what desync is?
That’s what we’re talking about. Getting anchor teammates to make you lose on purpose.
Why do you keep saying this when it’s either entirely random or actively organized to be a mismatched game? No aspect of this system is fair across the board, and I’m not sure that’s even possible at this point. I can count on one hand the number of games I’ve played of infinite where I felt the matchups were even and balanced. I’ve played halo for decades at this point and know the difference
Exactly. It’s random for you. It’s random for your opponent.
It’s frustrating as -yoink- when it happens.
But as long as it happens randomly to everyone your rank will be unaffected over time.
Weird, because there seems to be just as many threads talking about the handicap being for the other system.
The ranking system has got you.
As long as the person you are playing for or with is ranked relative to you it doesn’t matter what system they are on. Or what chair they are using, how many pixels their monitor has, how chilled their Mountain Dew is.
You just played someone who is performing at a certain rank. You either beat them or you didn’t. And your rank will change compared to their rank.
Does it really matter if that player would a bit better or a bit worse if they had a different set up?
We all know what desync is. And we all know that 343 don’t apply it to stop winning streaks.
You are getting team-mates - both better and worse - to anchor your team to be the same skill as the opposition.
I could think about that, if i would play 95% of my matches with a stacked team. But is the other way. Im playing 98% of matches as solo player.
And thats the point where so see, that this logic is flawed. 3 ranked matches in a row where im starting solo the match making on peak playing hour europe. In this 3 matches, in a row, on each match i get matched with 3 new player. This way the CSR of my previous match has nothing to do with this new match and the 3 new team mates.
At this point, what remains is my individual skill that is not reflected in a accurate way with my CSR rank.
We can discuss it the whole day -the team won or lost the match. As soon that the match ends, the team of randoms doesnt exists anymore, but all 4 player get additinal CSR points if they won the match or lose CSR points if the lost the match.
I mean, after a few minutes i get a feeling if we will lose or win the match. Or if the outcome will be close because the team are really well balanced ( its absolute rare that this combination happens).
If i realize that we are going to lose anyway, why should i put more effort as what it needs, that guarantees me that i end the match with a positive k/d.
Im pretty sure, that i would lose exact the same CSR point as when i would end with a positive k/d and playing the objective, putting in as match effort as i can. Am i right or wrong?
It is the better way to think about it. For you own sanity. And to remind you that you do have some control over the way the team gels.
Ideally you would have multiple ranks - for each combination of player’s you played with. But obviously that is completely impractical.
Each new match is based off your current MMR.
The CSR gains or losses of your previous match don’t tell you a lot about what your MMR did - although by now you should be getting a feel for your own performance and how tough the team was to play.
It depends by what you mean as “accurate”.
If you are focusing on Diamond 3 1305 being significantly better than Diamond 3 1302 then you are going to be disappointed.
There just isn’t the precision in the system to a few points.
Your MMR is going up and down in the background on form (probably a bit more than we expect). But it’s doing it’s job assessing your current form (which fluctuates) and trying to give the match maker the best info possible.
Your CSR is trying to smooth that journey and, at the same time, provide a sense of reward for a win.
At times your MMR may hit a peak (well done you) but there isn’t time for your CSR to climb that high before your form drops off. That’s a frustrating thought. But we also know that at times your MMR drops below your CSR but you are saved that big drop by the slower moving CSR.
The lag works both ways.
So, what does it mean to say your rank is 1305? Does that mean you are currently performing every game as 1305. No, of course it doesn’t. We know individual performance can fluctuate by 100’s day to day. This is just your averaged performance over the last ‘x’ number of games / days.
And once again, this ridiculous grind for CSR points is the fault of 343. We need to go back to a smaller, simpler scale. One that changes and unfolds like your actual rank - over days to weeks.
It’s frustrating.
My only advice is to relax and concentrate on your KPM and DPM. If the match was genuinely expected to be a loss then your MMR is not going to be hurt. And you can minimise any change by at least meeting expectation in terms of K and D.
You are right. Your CSR gains or loss aren’t affected by your personal performance (or by how much your team lost by).
But by meeting (or exceeding) expectation in regards to KPM and DPM will minimise the MMR change (and in rare cases even raise it). This will affect how much CSR you get in the next game.
And if it affects everyone the same then you will still be sorted into the same ranking order.
If you had a ten pin bowling championship and every bowler had to bowl with a randomly selected ball weight - after enough games you would find the rankings come out pretty much the same. The more talented bowler will adjust better than the less talented bowler.
I have. But I don’t get your point.
You skill level is the total addition of all factors.
The ranking system is agnostic to all of that it. It just takes your performance and ranks you against the performance of everyone else.
Sure, a tennis pro is probably going to drop a few games if you give them a cheap department store tennis racquet. But the key is that if they only had that racquet in the first place they would have just ranked accordingly.
Because they don’t have to.
They simply have to give you harder games to the point where you aren’t good enough to win them and your win rate naturally falls back to 50:50.
Why would they put extra effort into rigging an outcome that you are going to get anyway?
It’s just paranoid -yoink- by people who are frustrated with their own skill ceiling.
And?
Everyone is wrestling with the same problem.
You will be sorted into a rank order like everyone else.
And then given an arbitrary label for that position.
If you don’t like the position you need to get better individually to carry the side better and/or learn how to work better as a team.
The factors are random. Nothing is deliberately assigned by 343. Except equally ranked teams.
Why? Because you don’t like the rank you have?
The MMR ranks the players into order. We know it does this well.
The CSR simplifies this journey. It’s just badly implemented and explained by 343. That justifies discussion on how to change it.
But I’m not sure how this results in getting rid of the MMR system?