Forums / Community / Matchmaking Feedback & Discussion

[Locked] Matchmaking Feedback Update – June 11

OP ZaedynFel

  1. 1
  2. ...
  3. 36
  4. 37
  5. 38
  6. 39
  7. 40
  8. ...
  9. 50
QX wrote:
D M4N8 wrote:
Game

I'd like to know how the guy in my team ranks up, but I don't...
We did about the same...
He probably does better than you on average. Just because he had a bad game or you have a good one doesn't negate all of the previous games you two have played. You may have started to hit your mmr while he still might have a little csr to get to catch up.
<p></p>
We played a couple games together and won all games. Just doesn't make sense. Similar outcomes.
eLantern wrote:
What are up with these warzone games that I lose every game 1000 to 300? Every single game. Does not see like equal matchmaking to me 😡
Most likely population related. Not enough players available within a reasonable search time to give you ideally 50/50 balanced matches.
I think this game has really just gone to crap. Really disappointed in 343 to be honest. Ruined a good game back when they changed the BR and now with the matchmaking and ranking.
eLantern wrote:
What are up with these warzone games that I lose every game 1000 to 300? Every single game. Does not see like equal matchmaking to me 😡
Most likely population related. Not enough players available within a reasonable search time to give you ideally 50/50 balanced matches.
I think this game has really just gone to crap. Really disappointed in 343 to be honest. Ruined a good game back when they changed the BR and now with the matchmaking a ranking.
The game is coming up on 3 years old so matchmaking is going to suffer from some population problems. Not much can be done about that; however, this game's population isn't terribly bad considering its age.

I wouldn't say they ruined the game with the BR change or really any of the sandbox changes. While I was absolutely no fan of the initial BR changes and still share some disappointment with its current state I understand much of the logic that was used in targeting it for modification. The end result has been that the weapon still has it's uses and it does fill a role. Dare I say, the current default H5BR is fairly suitable for BTB, but for a simple non-intricate fix I do wish they'd consider using the Long Barrel H5BR with Recon or Classic scope variants as a pick-up option within Ranked Arena instead of the default H5BR because the standard Magnum pretty much outclasses it in just about every way so there's little reason to pick it up.

The individual CSRs within matchmaking are fine. Ignorance of the system is probably the biggest problem, but the amount of inflated ego is right up there too. CSRs are much more representative of our actual TrueSkills and that's a good thing. Our TrueSkill (aka MMR) is also much more accurate and that's thanks to TS2 which in turn helps the matchmaker create more balanced matches as long as the population is available to support it. This is also a good thing.

Now, would it be nice to have a playlist progression system similar to the military ranks that Halo 3 had on top of an accurate skill-based rank? Absolutely, but that type of update isn't likely to occur because most of 343i are working on Infinite or the MCC update. It'd also be neat to see fireteam registrations similar to Spartan Companies so that purely team-based CSRs could be earned and seen within their own leaderboard, but that too isn't going to happen because of the amount of work required to make it possible.
D M4N8 wrote:
QX wrote:
D M4N8 wrote:
Game

I'd like to know how the guy in my team ranks up, but I don't...
We did about the same...
He probably does better than you on average. Just because he had a bad game or you have a good one doesn't negate all of the previous games you two have played. You may have started to hit your mmr while he still might have a little csr to get to catch up.
<p></p>
We played a couple games together and won all games. Just doesn't make sense. Similar outcomes.
Hard to tell with such a small sample size. He could have been having a bad day while you might have had a slightly better one than average. Josh would be able to answer with MMR data but he seems to have gone MIA in the thread.

Just based on when I have played with you high diamond in doubles would seem to be your rank. But in all honesty I recommend not really thinking about rank and just enjoying good games. Everyone in this thread is too worried about ranking up and ranking down and having their numbers change when matchmaking ranks mean very little. Not only are they based on the population that has ever played that playlist but with the d3 cap now there will always be people able to pass you in ranks.
QX wrote:
But in all honesty I recommend not really thinking about rank and just enjoying good games. Everyone in this thread is too worried about ranking up and ranking down and having their numbers change when matchmaking ranks mean very little.
I've been wanting to share this sentiment for a while now, although I would probably disagree to a point about ranks being meaningless.

I for one don't care about my rank, but I'm glad it's (mostly) accurate. It does feel good to see a "number" behind why a game had a certain outcome though. Like, oh, yep, those guys were high-diamond/Onyx; it sure felt like they were high-level players, it makes sense why we lost. Or, yay, I did really well against that particular team of Diamonds!
I have proof (sample size of 1 but whatever) that the new ranking system has clear mmr bias. Here are my stats in the elimination playlist:

kills: 1750
deaths: 560
Assist: 630
games: 160
win% : 80.6
accuracy: 55%

My CSR has converged with my mmr at platinum 6. Now I’ve been champ in this playlist multiple times, and I know that CSR was inflated before but I find it hard to believe that mine was inflated this much.

That’s why I created a smurf account named WearyFawn535895, here are his stats:

kills: 600
deaths: 130
assists: 130
Games: 40
win% : 85
Accuracy: 59.6%

I got champ 36 after 24 games and reached champ 12 when my mmr and CSR converged at 40 games.
now I know what you’re thinking, “man lgndofhalo3, you’re playing much better on your smurf account, your skill must have gone up!”
I have played halo for 12 years, and I can promise you that my skill wouldnt go from plat 6 to champ 12 in a day.

In fact, the only reason my stats are so much better on my smurf is because I am playing even worse people than usual. At least on my main account I played (and beat) the top ranked champs in the playlist (still didnt rank my mmr up though lol). But back to the point: I’m playing even worse people, and the game is rewarding me with more mmr than on my main account, just because i created a new account.

Now 343, I can’t pretend to know how your ranking system works, and I won’t bash on it for locking my main accounts rank around plat 6/ diamond 1 range, but I wanted to show you this test so that you can take a look at why this might be happening. I believe TS2 is far superior to TS1, but it definitely needs work. If anybody would like to discuss this or reason this, I would love to have a constructive conversation about this, and can provide more details if needed. Thank you for your time reading this.
...
I too would like to hear from Josh Menke regarding what you've presented. To me it seems like it might relate to the rare outlier cases that they were supposedly working on a fix for. I assumed that they put that fix in already (with the seasonal reset), but maybe it was for a specific playlist or perhaps they just didn't implement the fix completely yet... idk. Anyways, like I said I'm curious to hear from Menke regarding what he can gather from this example. Oh, and thanks for sharing this constructive feedback. It certain seems to indicate that something is not right in your particular case.

I would like to ask you about one thing that you stated though:
Quote:
At least on my main account I played (and beat) the top ranked champs in the playlist (still didnt rank my mmr up though lol).
How did you verify that your MMR didn't go up after that win? Was it because your CSR didn't go up +15?
eLantern wrote:
...
I too would like to hear from Josh Menke regarding what you've presented. To me it seems like it might relate to the rare outlier cases that they were supposedly working on a fix for. I assumed that they put that fix in already (with the seasonal reset), but maybe it was for a specific playlist or perhaps they just didn't implement the fix completely yet... idk. Anyways, like I said I'm curious to hear from Menke regarding what he can gather from this example. Oh, and thanks for sharing this constructive feedback. It certain seems to indicate that something is not right in your particular case.

I would like to ask you about one thing that you stated though:
Quote:
At least on my main account I played (and beat) the top ranked champs in the playlist (still didnt rank my mmr up though lol).
How did you verify that your MMR didn't go up after that win? Was it because your CSR didn't go up +15?
You can’t 100% be sure what your mmr did after a match, but if your mmr and CSR are converged and your CSR is going up +1 after games and then you win a game you “shouldn’t” have won and still get +1 CSR, In my opinion, that’s the most accurate way as a player to determine if your mmr went up. Now obviously I could have something wrong on who was “supposed” to win the match but i dont remember
eLantern wrote:
...
I too would like to hear from Josh Menke regarding what you've presented. To me it seems like it might relate to the rare outlier cases that they were supposedly working on a fix for. I assumed that they put that fix in already (with the seasonal reset), but maybe it was for a specific playlist or perhaps they just didn't implement the fix completely yet... idk. Anyways, like I said I'm curious to hear from Menke regarding what he can gather from this example. Oh, and thanks for sharing this constructive feedback. It certain seems to indicate that something is not right in your particular case.

I would like to ask you about one thing that you stated though:
Quote:
At least on my main account I played (and beat) the top ranked champs in the playlist (still didnt rank my mmr up though lol).
How did you verify that your MMR didn't go up after that win? Was it because your CSR didn't go up +15?
You can’t 100% be sure what your mmr did after a match, but if your mmr and CSR are converged and your CSR is going up +1 after games and then you win a game you “shouldn’t” have won and still get +1 CSR, In my opinion, that’s the most accurate way as a player to determine if your mmr went up. Now obviously I could have something wrong on who was “supposed” to win the match but i dont remember
Well, if you're winning at an 80%+ clip it's possible that your CSR was able to reach and possibly go beyond the 2-tier goal of having a so-called converged MMR and CSR. While MMR can swing pretty wildly depending on expectations it may not have swung enough to overlap your CSR post match; thus, the +1 instead of +15. Nevertheless, your case regarding two accounts with noticeably different CSRs remains intriguing and I look forward to what Josh can learn from it and explain about it.
I have proof (sample size of 1 but whatever) that the new ranking system has clear mmr bias. Here are my stats in the elimination playlist:

kills: 1750
deaths: 560
Assist: 630
games: 160
win% : 80.6
accuracy: 55%

My CSR has converged with my mmr at platinum 6. Now I’ve been champ in this playlist multiple times, and I know that CSR was inflated before but I find it hard to believe that mine was inflated this much.

That’s why I created a smurf account named WearyFawn535895, here are his stats:

kills: 600
deaths: 130
assists: 130
Games: 40
win% : 85
Accuracy: 59.6%

I got champ 36 after 24 games and reached champ 12 when my mmr and CSR converged at 40 games.
now I know what you’re thinking, “man lgndofhalo3, you’re playing much better on your smurf account, your skill must have gone up!”
I have played halo for 12 years, and I can promise you that my skill wouldnt go from plat 6 to champ 12 in a day.

In fact, the only reason my stats are so much better on my smurf is because I am playing even worse people than usual. At least on my main account I played (and beat) the top ranked champs in the playlist (still didnt rank my mmr up though lol). But back to the point: I’m playing even worse people, and the game is rewarding me with more mmr than on my main account, just because i created a new account.

Now 343, I can’t pretend to know how your ranking system works, and I won’t bash on it for locking my main accounts rank around plat 6/ diamond 1 range, but I wanted to show you this test so that you can take a look at why this might be happening. I believe TS2 is far superior to TS1, but it definitely needs work. If anybody would like to discuss this or reason this, I would love to have a constructive conversation about this, and can provide more details if needed. Thank you for your time reading this.
This is exactly the type of feedback that I love to see in this thread. Not only did you propose something might not be right with the csr system in a respectable way, but you also included the data that is relevant to the problem.

Considering how many games you have played the difference in ranks between the two accounts is very weird especially since you say that you play worse players on the Smurf on average. It would be wonderful if you or someone could go through the recent games to see if that is actually true or just you remembering the games you played bad people.

The only possible explaination that I can think of besides the system actually being broken is that on your Smurf you played considerably more people who are unranked or diamond, but have very high mmrs. Maybe people who don't actually play the playlist often to get champ but are extremely good at the game to warrant a very high mmr. If you are matching those people more on your Smurf and playing well against them then that account's mmr would be higher.

Also as a side note I have always wondered if the kpm/DPM values are as accurate at calculating win/loss in breakout especially compared to more standard gametypes where you are constantly in combat.
QX wrote:
I have proof (sample size of 1 but whatever) that the new ranking system has clear mmr bias. Here are my stats in the elimination playlist:

kills: 1750
deaths: 560
Assist: 630
games: 160
win% : 80.6
accuracy: 55%

My CSR has converged with my mmr at platinum 6. Now I’ve been champ in this playlist multiple times, and I know that CSR was inflated before but I find it hard to believe that mine was inflated this much.

That’s why I created a smurf account named WearyFawn535895, here are his stats:

kills: 600
deaths: 130
assists: 130
Games: 40
win% : 85
Accuracy: 59.6%

I got champ 36 after 24 games and reached champ 12 when my mmr and CSR converged at 40 games.
now I know what you’re thinking, “man lgndofhalo3, you’re playing much better on your smurf account, your skill must have gone up!”
I have played halo for 12 years, and I can promise you that my skill wouldnt go from plat 6 to champ 12 in a day.

In fact, the only reason my stats are so much better on my smurf is because I am playing even worse people than usual. At least on my main account I played (and beat) the top ranked champs in the playlist (still didnt rank my mmr up though lol). But back to the point: I’m playing even worse people, and the game is rewarding me with more mmr than on my main account, just because i created a new account.

Now 343, I can’t pretend to know how your ranking system works, and I won’t bash on it for locking my main accounts rank around plat 6/ diamond 1 range, but I wanted to show you this test so that you can take a look at why this might be happening. I believe TS2 is far superior to TS1, but it definitely needs work. If anybody would like to discuss this or reason this, I would love to have a constructive conversation about this, and can provide more details if needed. Thank you for your time reading this.
This is exactly the type of feedback that I love to see in this thread. Not only did you propose something might not be right with the csr system in a respectable way, but you also included the data that is relevant to the problem.

Considering how many games you have played the difference in ranks between the two accounts is very weird especially since you say that you play worse players on the Smurf on average. It would be wonderful if you or someone could go through the recent games to see if that is actually true or just you remembering the games you played bad people.

The only possible explaination that I can think of besides the system actually being broken is that on your Smurf you played considerably more people who are unranked or diamond, but have very high mmrs. Maybe people who don't actually play the playlist often to get champ but are extremely good at the game to warrant a very high mmr. If you are matching those people more on your Smurf and playing well against them then that account's mmr would be higher.

Also as a side note I have always wondered if the kpm/DPM values are as accurate at calculating win/loss in breakout especially compared to more standard gametypes where you are constantly in combat.
I haven't been saying much up to this point but I have to agree there is something not right.. Something I just want to share also.. For some odd reason when I create a alternate account the first thing I notice right off the start is that the game performs much better. Im talking performance not skill of course but the outcome definitely has a great impact on ones skill outcome.. When I made my first alt account I didn't open anything up.. no reqs no nothing.. Also you don't want to tie your account to waypoint either.. Now over time I have always used my main more.. Because I always liked my gamer tag ZippySkippy.. And that tag has been with me for years!! What I am saying here is this.. I don't think the issue is with TrueSkill2.0.. I think what has happened is TrueSkill2.0 has exposed hidden issues this game has.. Its easy to point fingers at TS2 because of this.. But I mentioned to Josh way back when he started talking about this new TS2.. The first thing I said to him was this! This will really expose the issues this game has in performance consistency.. And It has... There is some underlying issue this game has and I think TS2 has irritated the heck out of it and upset a lot of players that are still trying to be loyal.. There was something else Josh had mentioned and it shocked me when he did.. But not one person here caught it.. Which greatly impacts skill.. I will save that for another time..
I haven't been saying much up to this point but I have to agree there is something not right.. Something I just want to share also.. For some odd reason when I create a alternate account the first thing I notice right off the start is that the game performs much better. Im talking performance not skill of course but the outcome definitely has a great impact on ones skill outcome.. When I made my first alt account I didn't open anything up.. no reqs no nothing.. Also you don't want to tie your account to waypoint either..
Any XBL account has a Waypoint account already, you can't avoid linking the two.

Quote:
Now over time I have always used my main more.. Because I always liked my gamer tag ZippySkippy.. And that tag has been with me for years!! What I am saying here is this.. I don't think the issue is with TrueSkill2.0.. I think what has happened is TrueSkill2.0 has exposed hidden issues this game has.. Its easy to point fingers at TS2 because of this.. But I mentioned to Josh way back when he started talking about this new TS2.. The first thing I said to him was this! This will really expose the issues this game has in performance consistency.. And It has... There is some underlying issue this game has and I think TS2 has irritated the heck out of it and upset a lot of players that are still trying to be loyal..
Dude stop with the tin foil. The REQ system has NOTHING to do with TS2. They aren't even related to each other. You've convinced yourself there's some connection but there really isn't.
There was something else Josh had mentioned and it shocked me when he did.. But not one person here caught it.. Which greatly impacts skill.. I will save that for another time..
Then why even mention this if you have no intention of discussing it further?
Playing the "I know something you don't know" card isn't helping anyone. Leave the bait to Buzzfeed please
QX wrote:
I have proof (sample size of 1 but whatever) that the new ranking system has clear mmr bias. Here are my stats in the elimination playlist:

kills: 1750
deaths: 560
Assist: 630
games: 160
win% : 80.6
accuracy: 55%

My CSR has converged with my mmr at platinum 6. Now I’ve been champ in this playlist multiple times, and I know that CSR was inflated before but I find it hard to believe that mine was inflated this much.

That’s why I created a smurf account named WearyFawn535895, here are his stats:

kills: 600
deaths: 130
assists: 130
Games: 40
win% : 85
Accuracy: 59.6%

I got champ 36 after 24 games and reached champ 12 when my mmr and CSR converged at 40 games.
now I know what you’re thinking, “man lgndofhalo3, you’re playing much better on your smurf account, your skill must have gone up!”
I have played halo for 12 years, and I can promise you that my skill wouldnt go from plat 6 to champ 12 in a day.

In fact, the only reason my stats are so much better on my smurf is because I am playing even worse people than usual. At least on my main account I played (and beat) the top ranked champs in the playlist (still didnt rank my mmr up though lol). But back to the point: I’m playing even worse people, and the game is rewarding me with more mmr than on my main account, just because i created a new account.

Now 343, I can’t pretend to know how your ranking system works, and I won’t bash on it for locking my main accounts rank around plat 6/ diamond 1 range, but I wanted to show you this test so that you can take a look at why this might be happening. I believe TS2 is far superior to TS1, but it definitely needs work. If anybody would like to discuss this or reason this, I would love to have a constructive conversation about this, and can provide more details if needed. Thank you for your time reading this.
This is exactly the type of feedback that I love to see in this thread. Not only did you propose something might not be right with the csr system in a respectable way, but you also included the data that is relevant to the problem.

Considering how many games you have played the difference in ranks between the two accounts is very weird especially since you say that you play worse players on the Smurf on average. It would be wonderful if you or someone could go through the recent games to see if that is actually true or just you remembering the games you played bad people.

The only possible explaination that I can think of besides the system actually being broken is that on your Smurf you played considerably more people who are unranked or diamond, but have very high mmrs. Maybe people who don't actually play the playlist often to get champ but are extremely good at the game to warrant a very high mmr. If you are matching those people more on your Smurf and playing well against them then that account's mmr would be higher.

Also as a side note I have always wondered if the kpm/DPM values are as accurate at calculating win/loss in breakout especially compared to more standard gametypes where you are constantly in combat.
I haven't been saying much up to this point but I have to agree there is something not right.. Something I just want to share also.. For some odd reason when I create a alternate account the first thing I notice right off the start is that the game performs much better. Im talking performance not skill of course but the outcome definitely has a great impact on ones skill outcome.. When I made my first alt account I didn't open anything up.. no reqs no nothing.. Also you don't want to tie your account to waypoint either.. Now over time I have always used my main more.. Because I always liked my gamer tag ZippySkippy.. And that tag has been with me for years!! What I am saying here is this.. I don't think the issue is with TrueSkill2.0.. I think what has happened is TrueSkill2.0 has exposed hidden issues this game has.. Its easy to point fingers at TS2 because of this.. But I mentioned to Josh way back when he started talking about this new TS2.. The first thing I said to him was this! This will really expose the issues this game has in performance consistency.. And It has... There is some underlying issue this game has and I think TS2 has irritated the heck out of it and upset a lot of players that are still trying to be loyal.. There was something else Josh had mentioned and it shocked me when he did.. But not one person here caught it.. Which greatly impacts skill.. I will save that for another time..
I mean there has always been whispers and rumors that a Smurf makes the game easier for various reasons. It makes the game play better, less aim assist is applied to you. And while I have found that I generally play better on a Smurf and the rec menus work better when you have very few recs I think that there is no way to point out an association between playing better and being on a smurf account.

We have no way to actually test those ideas unlike where we have ranking data for the previous poster.

Overall I think that it is just placebo that causes me to feel better on a Smurf.
QX wrote:
I have proof (sample size of 1 but whatever) that the new ranking system has clear mmr bias. Here are my stats in the elimination playlist:

kills: 1750
deaths: 560
Assist: 630
games: 160
win% : 80.6
accuracy: 55%

My CSR has converged with my mmr at platinum 6. Now I’ve been champ in this playlist multiple times, and I know that CSR was inflated before but I find it hard to believe that mine was inflated this much.

That’s why I created a smurf account named WearyFawn535895, here are his stats:

kills: 600
deaths: 130
assists: 130
Games: 40
win% : 85
Accuracy: 59.6%

I got champ 36 after 24 games and reached champ 12 when my mmr and CSR converged at 40 games.
now I know what you’re thinking, “man lgndofhalo3, you’re playing much better on your smurf account, your skill must have gone up!”
I have played halo for 12 years, and I can promise you that my skill wouldnt go from plat 6 to champ 12 in a day.

In fact, the only reason my stats are so much better on my smurf is because I am playing even worse people than usual. At least on my main account I played (and beat) the top ranked champs in the playlist (still didnt rank my mmr up though lol). But back to the point: I’m playing even worse people, and the game is rewarding me with more mmr than on my main account, just because i created a new account.

Now 343, I can’t pretend to know how your ranking system works, and I won’t bash on it for locking my main accounts rank around plat 6/ diamond 1 range, but I wanted to show you this test so that you can take a look at why this might be happening. I believe TS2 is far superior to TS1, but it definitely needs work. If anybody would like to discuss this or reason this, I would love to have a constructive conversation about this, and can provide more details if needed. Thank you for your time reading this.
This is exactly the type of feedback that I love to see in this thread. Not only did you propose something might not be right with the csr system in a respectable way, but you also included the data that is relevant to the problem.

Considering how many games you have played the difference in ranks between the two accounts is very weird especially since you say that you play worse players on the Smurf on average. It would be wonderful if you or someone could go through the recent games to see if that is actually true or just you remembering the games you played bad people.

The only possible explaination that I can think of besides the system actually being broken is that on your Smurf you played considerably more people who are unranked or diamond, but have very high mmrs. Maybe people who don't actually play the playlist often to get champ but are extremely good at the game to warrant a very high mmr. If you are matching those people more on your Smurf and playing well against them then that account's mmr would be higher.

Also as a side note I have always wondered if the kpm/DPM values are as accurate at calculating win/loss in breakout especially compared to more standard gametypes where you are constantly in combat.
Well the elimination mmr is seperated into 3 gametypes: breakout, breakout 2.0 and extinction. The game only looks at your mmr in the gamemode you played when deciding if you ‘improved’ or not, and from there it adjusts that gametypes mmr only. And your total mmr is the average of the 3 gamemodes.

So the kpm/DPM factor is based off of whatever gamemode you’re playing compared to the average ‘onyx’ or ‘diamond’ player to determine your mmr change. I would assume that 343 would account for the elimination playlist having weird statlines, but this playlist along with swat seem to have issues.

I thought at first it was due to low player count or only poor players in the matchmaking pool, but I think there may be a flaw in these 2 playlists, as both my HCS playlist and slayer playlist are performing as I would expect, with champions being on both sides, onyx’ being on both sides, or diamonds being on both sides equally distributed for the most part.
I haven't been saying much up to this point but I have to agree there is something not right.. Something I just want to share also.. For some odd reason when I create a alternate account the first thing I notice right off the start is that the game performs much better. Im talking performance not skill of course but the outcome definitely has a great impact on ones skill outcome.. When I made my first alt account I didn't open anything up.. no reqs no nothing.. Also you don't want to tie your account to waypoint either..
Any XBL account has a Waypoint account already, you can't avoid linking the two.

Quote:
Now over time I have always used my main more.. Because I always liked my gamer tag ZippySkippy.. And that tag has been with me for years!! What I am saying here is this.. I don't think the issue is with TrueSkill2.0.. I think what has happened is TrueSkill2.0 has exposed hidden issues this game has.. Its easy to point fingers at TS2 because of this.. But I mentioned to Josh way back when he started talking about this new TS2.. The first thing I said to him was this! This will really expose the issues this game has in performance consistency.. And It has... There is some underlying issue this game has and I think TS2 has irritated the heck out of it and upset a lot of players that are still trying to be loyal..
Dude stop with the tin foil. The REQ system has NOTHING to do with TS2. They aren't even related to each other. You've convinced yourself there's some connection but there really isn't.
1) XBL and Waypoint are linked.. But they can be targeted separately.. They have separate ip addresses per say.. When one first creates an account and comes to waypoint for the first time that is when data seems to sync up. But the method I use is blocking it at my router level. Granted one has to have knowledge and know how to do this but your very high end routers have very extensive features like this.. The only reason I bring this up is sometimes waypoint seems to have a performance hitch to the actual in game performance. In other words its running in the background while physically playing the game.. It sometimes stays connected and sometimes it doesnt. It only needs to connect after a game to record stats to waypoint.. Ive block it while playing then unblocked it after. If I don't unblock it and come to waypoint the game wont appear.. Like it never existed.. Again just to be clear this has nothing to do with TS2. I just feel this has helped expose performances hitches like this indirectly..
2) As a moderator I would have expected a little better response then this.. At the very least you should have asked me questions to what made me come up with this determination.. Again this has nothing to do with TS2.. I didn't say that you did! If you would have took the time to clearly read my post completely you would have noticed im talking about performance issues/hitches.. The Reqs are just another example of what I mentioned above.. There are several servers/apps involved with H5 that do various things.. Theres not just one.. When I get sometime to play again I will post them here. That way maybe more might understand..
QX wrote:
I have proof (sample size of 1 but whatever) that the new ranking system has clear mmr bias. Here are my stats in the elimination playlist:

kills: 1750
deaths: 560
Assist: 630
games: 160
win% : 80.6
accuracy: 55%

My CSR has converged with my mmr at platinum 6. Now I’ve been champ in this playlist multiple times, and I know that CSR was inflated before but I find it hard to believe that mine was inflated this much.

That’s why I created a smurf account named WearyFawn535895, here are his stats:

kills: 600
deaths: 130
assists: 130
Games: 40
win% : 85
Accuracy: 59.6%

I got champ 36 after 24 games and reached champ 12 when my mmr and CSR converged at 40 games.
now I know what you’re thinking, “man lgndofhalo3, you’re playing much better on your smurf account, your skill must have gone up!”
I have played halo for 12 years, and I can promise you that my skill wouldnt go from plat 6 to champ 12 in a day.

In fact, the only reason my stats are so much better on my smurf is because I am playing even worse people than usual. At least on my main account I played (and beat) the top ranked champs in the playlist (still didnt rank my mmr up though lol). But back to the point: I’m playing even worse people, and the game is rewarding me with more mmr than on my main account, just because i created a new account.

Now 343, I can’t pretend to know how your ranking system works, and I won’t bash on it for locking my main accounts rank around plat 6/ diamond 1 range, but I wanted to show you this test so that you can take a look at why this might be happening. I believe TS2 is far superior to TS1, but it definitely needs work. If anybody would like to discuss this or reason this, I would love to have a constructive conversation about this, and can provide more details if needed. Thank you for your time reading this.
This is exactly the type of feedback that I love to see in this thread. Not only did you propose something might not be right with the csr system in a respectable way, but you also included the data that is relevant to the problem.

Considering how many games you have played the difference in ranks between the two accounts is very weird especially since you say that you play worse players on the Smurf on average. It would be wonderful if you or someone could go through the recent games to see if that is actually true or just you remembering the games you played bad people.

The only possible explaination that I can think of besides the system actually being broken is that on your Smurf you played considerably more people who are unranked or diamond, but have very high mmrs. Maybe people who don't actually play the playlist often to get champ but are extremely good at the game to warrant a very high mmr. If you are matching those people more on your Smurf and playing well against them then that account's mmr would be higher.

Also as a side note I have always wondered if the kpm/DPM values are as accurate at calculating win/loss in breakout especially compared to more standard gametypes where you are constantly in combat.
Each game mode has a matrix of values that weight the correlation between kpm, dpm, etc., and how much those values affect the outcome predictions.

I'd imagine those values are lower for breakout vs. elimination, and really low when compared to standard slayer.

Also, it hasn't been confirmed if the MMR/CSR Update Glitch (where CSR updates based on the MMR of the last mode played in the playlist, rather than the aggregate MMR) has been fixed yet. @ZaedynFel, can you confirm the current status of this?
QX wrote:
I have proof (sample size of 1 but whatever) that the new ranking system has clear mmr bias. Here are my stats in the elimination playlist:

kills: 1750
deaths: 560
Assist: 630
games: 160
win% : 80.6
accuracy: 55%

My CSR has converged with my mmr at platinum 6. Now I’ve been champ in this playlist multiple times, and I know that CSR was inflated before but I find it hard to believe that mine was inflated this much.

That’s why I created a smurf account named WearyFawn535895, here are his stats:

kills: 600
deaths: 130
assists: 130
Games: 40
win% : 85
Accuracy: 59.6%

I got champ 36 after 24 games and reached champ 12 when my mmr and CSR converged at 40 games.
now I know what you’re thinking, “man lgndofhalo3, you’re playing much better on your smurf account, your skill must have gone up!”
I have played halo for 12 years, and I can promise you that my skill wouldnt go from plat 6 to champ 12 in a day.

In fact, the only reason my stats are so much better on my smurf is because I am playing even worse people than usual. At least on my main account I played (and beat) the top ranked champs in the playlist (still didnt rank my mmr up though lol). But back to the point: I’m playing even worse people, and the game is rewarding me with more mmr than on my main account, just because i created a new account.

Now 343, I can’t pretend to know how your ranking system works, and I won’t bash on it for locking my main accounts rank around plat 6/ diamond 1 range, but I wanted to show you this test so that you can take a look at why this might be happening. I believe TS2 is far superior to TS1, but it definitely needs work. If anybody would like to discuss this or reason this, I would love to have a constructive conversation about this, and can provide more details if needed. Thank you for your time reading this.
This is exactly the type of feedback that I love to see in this thread. Not only did you propose something might not be right with the csr system in a respectable way, but you also included the data that is relevant to the problem.

Considering how many games you have played the difference in ranks between the two accounts is very weird especially since you say that you play worse players on the Smurf on average. It would be wonderful if you or someone could go through the recent games to see if that is actually true or just you remembering the games you played bad people.

The only possible explaination that I can think of besides the system actually being broken is that on your Smurf you played considerably more people who are unranked or diamond, but have very high mmrs. Maybe people who don't actually play the playlist often to get champ but are extremely good at the game to warrant a very high mmr. If you are matching those people more on your Smurf and playing well against them then that account's mmr would be higher.

Also as a side note I have always wondered if the kpm/DPM values are as accurate at calculating win/loss in breakout especially compared to more standard gametypes where you are constantly in combat.
Each game mode has a matrix of values that weight the correlation between kpm, dpm, etc., and how much those values affect the outcome predictions.

I'd imagine those values are lower for breakout vs. elimination, and really low when compared to standard slayer.

Also, it hasn't been confirmed if the MMR/CSR Update Glitch (where CSR updates based on the MMR of the last mode played in the playlist, rather than the aggregate MMR) has been fixed yet. @ZaedynFel, can you confirm the current status of this?
Sorry I think I worded what I was trying to say wrong. I know that each playlist and then each game mode in each playlist is only calculated to themselves for KPM and that they have different MMRs. I even got the oddball MMR glitch for team arena.

What I meant in my earlier post was whether KPM/DPM are as accurate at determining whether a team would win in breakout games because of the 1 life and also flag cap round wins. I just find that my kills per match and therefore general KPM is more varied in breakout than other gamemodes. However, that's just me and also I haven't collected all my data even so idk.
QX wrote:
QX wrote:
I have proof (sample size of 1 but whatever) that the new ranking system has clear mmr bias. Here are my stats in the elimination playlist:

kills: 1750
deaths: 560
Assist: 630
games: 160
win% : 80.6
accuracy: 55%

My CSR has converged with my mmr at platinum 6. Now I’ve been champ in this playlist multiple times, and I know that CSR was inflated before but I find it hard to believe that mine was inflated this much.

That’s why I created a smurf account named WearyFawn535895, here are his stats:

kills: 600
deaths: 130
assists: 130
Games: 40
win% : 85
Accuracy: 59.6%

I got champ 36 after 24 games and reached champ 12 when my mmr and CSR converged at 40 games.
now I know what you’re thinking, “man lgndofhalo3, you’re playing much better on your smurf account, your skill must have gone up!”
I have played halo for 12 years, and I can promise you that my skill wouldnt go from plat 6 to champ 12 in a day.

In fact, the only reason my stats are so much better on my smurf is because I am playing even worse people than usual. At least on my main account I played (and beat) the top ranked champs in the playlist (still didnt rank my mmr up though lol). But back to the point: I’m playing even worse people, and the game is rewarding me with more mmr than on my main account, just because i created a new account.

Now 343, I can’t pretend to know how your ranking system works, and I won’t bash on it for locking my main accounts rank around plat 6/ diamond 1 range, but I wanted to show you this test so that you can take a look at why this might be happening. I believe TS2 is far superior to TS1, but it definitely needs work. If anybody would like to discuss this or reason this, I would love to have a constructive conversation about this, and can provide more details if needed. Thank you for your time reading this.
This is exactly the type of feedback that I love to see in this thread. Not only did you propose something might not be right with the csr system in a respectable way, but you also included the data that is relevant to the problem.

Considering how many games you have played the difference in ranks between the two accounts is very weird especially since you say that you play worse players on the Smurf on average. It would be wonderful if you or someone could go through the recent games to see if that is actually true or just you remembering the games you played bad people.

The only possible explaination that I can think of besides the system actually being broken is that on your Smurf you played considerably more people who are unranked or diamond, but have very high mmrs. Maybe people who don't actually play the playlist often to get champ but are extremely good at the game to warrant a very high mmr. If you are matching those people more on your Smurf and playing well against them then that account's mmr would be higher.

Also as a side note I have always wondered if the kpm/DPM values are as accurate at calculating win/loss in breakout especially compared to more standard gametypes where you are constantly in combat.
Each game mode has a matrix of values that weight the correlation between kpm, dpm, etc., and how much those values affect the outcome predictions.

I'd imagine those values are lower for breakout vs. elimination, and really low when compared to standard slayer.

Also, it hasn't been confirmed if the MMR/CSR Update Glitch (where CSR updates based on the MMR of the last mode played in the playlist, rather than the aggregate MMR) has been fixed yet. @ZaedynFel, can you confirm the current status of this?
Sorry I think I worded what I was trying to say wrong. I know that each playlist and then each game mode in each playlist is only calculated to themselves for KPM and that they have different MMRs. I even got the oddball MMR glitch for team arena.

What I meant in my earlier post was whether KPM/DPM are as accurate at determining whether a team would win in breakout games because of the 1 life and also flag cap round wins. I just find that my kills per match and therefore general KPM is more varied in breakout than other gamemodes. However, that's just me and also I haven't collected all my data even so idk.
No, you worded it fine. I was agreeing with you, and providing a little context to support the position. I find I have to do that with my posts on these forums. Not necessarily with you, but in general.

All good here. 👍
So I played over 250 games in this season already and I have some things to ask, that is not clear to me.

In the past couple of days, well, until a week ago, I was playing with the party of four. We all ranked P2-6 respectively but we gained our ranks quickly and that is D3-6. The problem started to occurs when we reached D4-5 tiers. We just stopped receiving CSR. No matter if we were winning like 5 in a row, one night ever 10 in a row. CSR just didn't want to move, not even a bit.

So I thought the system is broken again, then I changed the party, started playing with other people on my friends' list. And guess what? I moved from D4 to D6 in two hours, maybe less. Same matches, same enemies, people that I was playing with are the same skill level as my friends (d4-6). Again, we were winning the same teams, Onyxes, High Diamonds, even Champs in some matches. Then the problem started to occur again. I stopped getting any CSR after wins no matter who we play, and after a loss, I lose 33% or even more. I am not carried or whatsoever. Especially when I'm playing with my buddies. We're performing differently on different maps, but if someone does not lag out, we are almost all identical, swapping positions 1st, 2nd depending on the map.

But how it goes, you win some you lose some and I'm fluctuating between D5-6 now. but the problem still persists. No matter how matches I win with my party, I can not rank playing with them. Firstly, I was stuck at D3 with them, then D5 and now D6 not only with them but with everyone else (3 wins after high D or Onyxes mixed with high D, not moving an inch)

So I would like to know if the MMRs of the team, parties is still not... aligned or there's some other issue? Again, I mostly... almost never receive any CSR playing with my buddies, but I am receiving with other people. I can provide you a Gamertags of those people if needed.

And one more complaint if I may. You still can rank up playing in a party of golds. I got a couple of people on my friends' list that are using some -Yoink- players to rank up. Of course, they are truly very low D or even Plats, but they CSR is saying High Ds. I thought it is fixed if you try to be carried or carry some ultra low ranked players for the gains of CSR
  1. 1
  2. ...
  3. 36
  4. 37
  5. 38
  6. 39
  7. 40
  8. ...
  9. 50