Skip to main content

Forums / Community / Matchmaking Feedback & Discussion

[Locked] MATCHMAKING FEEDBACK UPDATE – November 5

OP ZaedynFel

  1. 1
  2. ...
  3. 10
  4. 11
  5. 12
  6. 13
  7. 14
  8. ...
  9. 27
Thorulfr wrote:
I'm just going to list all the ways you can play 4v4 Halo 5 at the moment...

Quick Play
Core Play
Slayer
Team Arena

Honourable mentions to Super Fiesta, Elimination and SWAT.

I get there were reasons behind each addition and slightly different use cases for each playlist... but come on. No wonder the population is so split between different playlists.
Team Arena is the competitive settings which are completely different than slayer. Core play is rotational so it will be gone in a week and a half or so.

So there is a competitive playlist, a 4v4 slayer one for ranked and social and then a rotational that will be switched out for something else. Combining all these into one would be terrible and drive more people off the game.
RzR J3ST3R wrote:
This Roy/Flamesword clinic on stream is why /min rates should not be used in Halo MMR. Slow and steady.
H5 4v4 plays NOTHING like H3 2v2. Different universe entirely.
Not a comparison I made, but whatever- Dubs is dubs. You can successfully and often should play slower than the MMR optimum, even if it's not H3 slow.

Point is: don't punish other ways to win for the sake of a predictor.
But you are comparing a system that ranks all level of players through a trackable method versus very top end players playing against each other. Yes at the highest levels slowing down the game is necessary and a good strategy. However, I bet that in normal matchmaking these players would play faster compared so how they are at a tournament.

Also yes you are comparing H3 doubles to H5. Which do in fact play differently because you can move around exposed areas of the map on H5 with less risk than H3 due to movement mechanics.

And finally the current MMR system is literally based on what predicts a team to WIN. So it turns out that higher KPM results in higher win percentages on average in the overall matchmaking. Would you rather have them utilize a worse prediction method just because you feel that your rank is being negatively affected by KPM?
RzR J3ST3R wrote:
This Roy/Flamesword clinic on stream is why /min rates should not be used in Halo MMR. Slow and steady.
H5 4v4 plays NOTHING like H3 2v2. Different universe entirely.
Not a comparison I made, but whatever- Dubs is dubs. You can successfully and often should play slower than the MMR optimum, even if it's not H3 slow.

Point is: don't punish other ways to win for the sake of a predictor.
/minute rates are entirely relative. It’s relative to the players themselves and what they achieve per the modes/game-types. Point is, players determine the /minute rates that reflect optimum MMR.
eLantern wrote:
RzR J3ST3R wrote:
This Roy/Flamesword clinic on stream is why /min rates should not be used in Halo MMR. Slow and steady.
/minute rates are entirely relative. It’s relative to the players themselves and what they achieve per the modes/game-types. Point is, players determine the /minute rates that reflect optimum MMR.
The point is it's not inclusive. If that was a matchmaking game, Roy/Flamesword (or whoever) would be better off rushing/trading and potentially losing then playing for the win.
I point to Roy/Flamesword because they are easily recognizable examples, but this certainly is not limited to mainstage pros.

Playing anything off the current meta does not necessarily mean you are less effective, regardless of what Halo you are playing.
QX wrote:
But you are comparing a system that ranks all level of players through a trackable method versus very top end players playing against each other. Yes at the highest levels slowing down the game is necessary and a good strategy. However, I bet that in normal matchmaking these players would play faster compared so how they are at a tournament.

Also yes you are comparing H3 doubles to H5. Which do in fact play differently because you can move around exposed areas of the map on H5 with less risk than H3 due to movement mechanics.

And finally the current MMR system is literally based on what predicts a team to WIN. So it turns out that higher KPM results in higher win percentages on average in the overall matchmaking. Would you rather have them utilize a worse prediction method just because you feel that your rank is being negatively affected by KPM?
TS2 isn't just a H5 thing, but if we want to talk about H5: I'm not playing main stage. I am in an H5 bracket where it is still a viable strategy to slow the game down. Doubles, regardless of what Halo title, should often be played slower than the MMR average. So would I sacrifice a few percentage points in global prediction accuracy to not have my gameplay dictated? Emphatically yes.

"Smart," "disciplined" play, as Bravo described it, when used appropriately, is not going to be rewarded. KPM would make a great advanced stat, but using it for rank remains a mistake. That I can game TS2 to rank up more searching solo (vs. with my buddy) by stealing kills from often lesser teammates is a problem.

I bring it up because rates seem to be so heavily baked into TS2, and it's not unreasonable to expect TS2 beyond H5. It is not a H5 vs H3 thing. As is, we might as well replace the score counter with a KPM tracker- it would at least simplify a convoluted feedback system. Wins are just a trigger event right now. Some might say that sounds ridiculous (I'd agree), but it's a much better representation of what matters now.
Not ranking up in doubles despite great performances
Sturbz wrote:
Not ranking up in doubles despite great performances
Great performances tend to show their adjustment influence on the following match if it’s a win and especially when it’s against a more difficult opponent.
Sturbz wrote:
Not ranking up in doubles despite great performances
Just as a reminder CSR change is now based on two aspects. First is how well you perform compared to the expectations set up for you on a game by game basis. So if you are expected to have those great performances and your CSR is near your MMR then your CSR isn't going to go up very far considering the game is correctly identifying your skill level.

The second aspect is defined on what types of opponents you beat. Basically the game takes into account your MMR vs the opponents and uses that to determine about 50% of CSR change.

So the first part drives your CSR to your MMR while the second drives your CSR to your opponents MMR. If you are mainly only facing players of worse skill then you then you will be expected to have great performances while also getting diminishing CSR returns since you are playing people below your skill level. This is going to cause a plataeu in your CSR progression unless you consistently increase your MMR which is impossible over time.
QX wrote:
Sturbz wrote:
Not ranking up in doubles despite great performances
Just as a reminder CSR change is now based on two aspects. First is how well you perform compared to the expectations set up for you on a game by game basis. So if you are expected to have those great performances and your CSR is near your MMR then your CSR isn't going to go up very far considering the game is correctly identifying your skill level.

The second aspect is defined on what types of opponents you beat. Basically the game takes into account your MMR vs the opponents and uses that to determine about 50% of CSR change.

So the first part drives your CSR to your MMR while the second drives your CSR to your opponents MMR. If you are mainly only facing players of worse skill then you then you will be expected to have great performances while also getting diminishing CSR returns since you are playing people below your skill level. This is going to cause a plataeu in your CSR progression unless you consistently increase your MMR which is impossible over time.
I think in Sturbz case the CSR is accurate.

Those great performances are shown as the spikes in the graph below, but in the context of the overall trend, you can see why CSR is going down.
sturbz
eLantern wrote:
RzR J3ST3R wrote:
This Roy/Flamesword clinic on stream is why /min rates should not be used in Halo MMR. Slow and steady.
/minute rates are entirely relative. It’s relative to the players themselves and what they achieve per the modes/game-types. Point is, players determine the /minute rates that reflect optimum MMR.
The point is it's not inclusive. If that was a matchmaking game, Roy/Flamesword (or whoever) would be better off rushing/trading and potentially losing then playing for the win.
I point to Roy/Flamesword because they are easily recognizable examples, but this certainly is not limited to mainstage pros.

Playing anything off the current meta does not necessarily mean you are less effective, regardless of what Halo you are playing.
QX wrote:
But you are comparing a system that ranks all level of players through a trackable method versus very top end players playing against each other. Yes at the highest levels slowing down the game is necessary and a good strategy. However, I bet that in normal matchmaking these players would play faster compared so how they are at a tournament.

Also yes you are comparing H3 doubles to H5. Which do in fact play differently because you can move around exposed areas of the map on H5 with less risk than H3 due to movement mechanics.

And finally the current MMR system is literally based on what predicts a team to WIN. So it turns out that higher KPM results in higher win percentages on average in the overall matchmaking. Would you rather have them utilize a worse prediction method just because you feel that your rank is being negatively affected by KPM?
TS2 isn't just a H5 thing, but if we want to talk about H5: I'm not playing main stage. I am in an H5 bracket where it is still a viable strategy to slow the game down. Doubles, regardless of what Halo title, should often be played slower than the MMR average. So would I sacrifice a few percentage points in global prediction accuracy to not have my gameplay dictated? Emphatically yes.

"Smart," "disciplined" play, as Bravo described it, when used appropriately, is not going to be rewarded. KPM would make a great advanced stat, but using it for rank remains a mistake. That I can game TS2 to rank up more searching solo (vs. with my buddy) by stealing kills from often lesser teammates is a problem.

I bring it up because rates seem to be so heavily baked into TS2, and it's not unreasonable to expect TS2 beyond H5. It is not a H5 vs H3 thing. As is, we might as well replace the score counter with a KPM tracker- it would at least simplify a convoluted feedback system. Wins are just a trigger event right now. Some might say that sounds ridiculous (I'd agree), but it's a much better representation of what matters now.
We recently did a TS2 batch run which significantly reduced the influence of kpm on rank, hence the reason tons of doubles players have moved up from mid-Diamond to Onyx and Champ.

The batch run also resulted in a massive increase in Doubles accuracy, pushing it over 70% compared to the under 60% it was at before.

Also, no, we won't change how we estimate skill in matchmaking since that's what players experience there, regardless of how things may play differently at event. Though the overall Ranks line up pretty well with who the top Pros are, at least within a tolerable delta in MMR. So even there, it's not far enough off for us to worry.

So high MMR players belong at events, and if they want to show who really deserves the top, they should go.
Now that HCS is over with, are there any plans with the Team Arena playlist or will things stay the same?

Once Thanksgiving break is done, could we get a popularity list to see if BTB improved at all or not?
ZaedynFel wrote:
QX wrote:
Sturbz wrote:
Not ranking up in doubles despite great performances
Just as a reminder CSR change is now based on two aspects. First is how well you perform compared to the expectations set up for you on a game by game basis. So if you are expected to have those great performances and your CSR is near your MMR then your CSR isn't going to go up very far considering the game is correctly identifying your skill level.

The second aspect is defined on what types of opponents you beat. Basically the game takes into account your MMR vs the opponents and uses that to determine about 50% of CSR change.

So the first part drives your CSR to your MMR while the second drives your CSR to your opponents MMR. If you are mainly only facing players of worse skill then you then you will be expected to have great performances while also getting diminishing CSR returns since you are playing people below your skill level. This is going to cause a plataeu in your CSR progression unless you consistently increase your MMR which is impossible over time.
I think in Sturbz case the CSR is accurate.

Those great performances are shown as the spikes in the graph below, but in the context of the overall trend, you can see why CSR is going down.
I just want to confirm that that graph is for doubles right? I know it probably is but since there is no label people might get confused.

Thanks.
LUKEPOWA wrote:
Now that HCS is over with, are there any plans with the Team Arena playlist or will things stay the same?

Once Thanksgiving break is done, could we get a popularity list to see if BTB improved at all or not?
I think waiting until after the refresh for a population update would be for the best since then people won't change how they play based on where BTB is at the moment, for good or for bad.
For the Infection playlist can’t you make it less laggy, it is pissing a lot of people off and causing them to leave the game for good. (Talking about Infection sweats and other noobs
The Match Composer is great. One changed I think should be made for CE is with BTB KOTH being 5 minutes. That's way too long, I'd suggest 3 mins or less.
The Match Composer is great. One changed I think should be made for CE is with BTB KOTH being 5 minutes. That's way too long, I'd suggest 3 mins or less.
This is for Halo 5. Please leave MCC feedback in the MCC section.

Thanks!
Hey senpai Menke, I wanna know my FFA & Slayer MMRS! Ontop of my highest social MMR

Also, a graph for Slayer if posdible too! Thanks in advance, dr.
I really need someone from 343 to take the things I have to say seriously!! I have a host of topics that I've observed over recent months that I'd like to address, but foremost right now, I wanna talk about achievements! Specifically the MCC achievements that are attached to Halo 2 Anniversary. I doubt it if anyone has realized at 343, but I have noticed that because of a Playlist update, territories on H2A maps are virtually non - existent. Which makes holding 3 territories next to impossible to get and juggernaut has been COMPLETELY removed from all H2A maps making the acquisition of the achievement for staying alive 30 seconds as the Jug impossible to accuire! This might not seem like an important issue, but it's not fun to be excluded from being able to achieve something in your favorite game! If I'm mistaken, please let me know someone, but I've been sitting here combing through menu data and juggernaut is completely gone and I've only hit on one territory game in my last 100 H2A matches
I really need someone from 343 to take the things I have to say seriously!! I have a host of topics that I've observed over recent months that I'd like to address, but foremost right now, I wanna talk about achievements! Specifically the MCC achievements that are attached to Halo 2 Anniversary. I doubt it if anyone has realized at 343, but I have noticed that because of a Playlist update, territories on H2A maps are virtually non - existent. Which makes holding 3 territories next to impossible to get and juggernaut has been COMPLETELY removed from all H2A maps making the acquisition of the achievement for staying alive 30 seconds as the Jug impossible to accuire! This might not seem like an important issue, but it's not fun to be excluded from being able to achieve something in your favorite game! If I'm mistaken, please let me know someone, but I've been sitting here combing through menu data and juggernaut is completely gone and I've only hit on one territory game in my last 100 H2A matches
1. This is the wrong section for this post. Try HERE.

2. I too have made some comments regarding modes missing within Matchmaking. You can read it HERE.

3. I personally think that the Juggernaut mode should be made available through the MCC's Action Sack option which ought to combine 3 different 8 player options: 4v4, FFA, and Multi-Team (4 teams of 2). Hopefully 343i can make Juggernaut available as an Action Sack FFA mode as that's where I'd argue it best belongs.
BTB isn't popular any more because 343 clearly treated it as a second rate game type. It's clear to see pretty much all effort went into warzone for H5. That is why it's maps look great and are diverse from one another. When BTB first launched you could tell the maps were forged up quickly and slammed into the playlist. BTB got a bit better after the refresh graphics wise but maps still weren't great. Doesn't help that vehicles are pretty weak, (probably nerfed down because of req varients in warzone). You can pick gunners out of warthogs easily. In previous Halo warthogs where worrisome and now they are hardly a factor. Not to mention not many maps are even vehicle friendly anyways. I also have to disagree with the H5BR. It has never ever been good IMO. Worse feeling BR in all Halos. Doesn't seem like an awesome burst fired rifle, seems more like a nail gun. We need bigger slightly more open maps pretty much all around. Why make Spartans faster and more agile but maps smaller and more cramped?
I don't know if anyone else is experiencing this issue or doing anything about it but I waited 15 mins just to find a game and when I did it restarted then I waited 10 more minutes and I shut off my xbox this is definetly a matchmaking problem and this needs to be fixed ASAP I mean cmon it takes me not even 30 seconds just to find a game in halo reach and the mcc has been out for 4 years and its still a mess so 343 if you can hear me please please please fix the matchmaking its not fun waiting this long just to find one game and fyi its not a population issue this is a matchmaking bug that needs to be dealt with before anything else
I don't know if anyone else is experiencing this issue or doing anything about it but I waited 15 mins just to find a game and when I did it restarted then I waited 10 more minutes and I shut off my xbox this is definetly a matchmaking problem and this needs to be fixed ASAP I mean cmon it takes me not even 30 seconds just to find a game in halo reach and the mcc has been out for 4 years and its still a mess so 343 if you can hear me please please please fix the matchmaking its not fun waiting this long just to find one game and fyi its not a population issue this is a matchmaking bug that needs to be dealt with before anything else
MCC is hardly a mess anymore. You may want to go back and check it out . Especially matched composer. It is awesome.
ZaedynFel wrote:
Could you tell me my new MMR/graph in Slayer, please?

Happy Thanksgiving!
  1. 1
  2. ...
  3. 10
  4. 11
  5. 12
  6. 13
  7. 14
  8. ...
  9. 27