Forums / Community / Matchmaking Feedback & Discussion

[Locked] Matchmaking Feedback Update – January 8, 2018

OP ZaedynFel

  1. 1
  2. ...
  3. 2
  4. 3
  5. ...
  6. 8
Happy New Year!
It’s good to be back! I have a few fun updates.

Playlists
We have a couple of playlist changes coming this week. First, as we announced last Friday, the HCS list will be updated to reflect the official HaloWC maps and modes. Head over to http://halo.gg for more info.

In addition, we’ll be adding a simple playlist called Quickplay at the top of our Social lists. The purpose of this list is largely so incoming players having an obvious starting point. The biggest feedback we get from new players is many if not the majority haven’t played any Halo in the past and have no idea what those playlist names mean. This should make it super clear where to start. The list will feature Slayer on Plaza (not Rig), Truth, and Coliseum. We’re keeping the maps to a minimum so folks aren’t overwhelmed, and using Slayer since it’s a straightforward game mode for both new and solo players. The playlist will also have a maximum party size of 2. If you’re looking for some light Slayer gameplay with the chance to mentor incoming folks, hop in!

Skill System
As some of you saw, I’ll be speaking about the new system we are in the process of integrating at this year’s GDC. We now have just enough of the plumbing built to see results coming back from matches. The accuracy is significantly higher than what we have today despite the fact there are a few key pieces we still haven’t hooked up. That accuracy will only grow as we finish.

To be clear, we aren’t yet using the new system, but it is able to analyze current matches. I’m able to see win predictions, kill predictions, and death predictions, and compare to them what really happened. I’ll see numbers like “Kill Prediction: 13.8, Actual: 14” often. It’s fun for a stats nerd like me to be honest. On average, in its current unfinished state, it’s able to predict how many kills each player will have per minute within less than 0.4 kills. It’s often right on the money.

I’ve already applied it to examining a few samples games. For example, one match included several smurfs who are Champs on their main accounts. Our current (older) system said that the smurfs would get destroyed because it couldn’t tell they were smurfs. The newer system correctly called the match even, and identified the abilities of the players. What this means is players will quickly find themselves matching where they should be, and opponents will get a more accurate CSR update.

On the accuracy side of things, I’m seeing increases of up to 10% in how well it predicts the winner. I’ve been in this business for a long time, and have never seen that kind of jump in predictive accuracy. Definitely not on data where we are already doing a solid job of creating close matches that should be hard to predict. When the system goes live, we could see drastic improvement in match quality.

I don’t have an ETA on when we’ll switch over to using the system, but just thought some of you would enjoy hearing how well it’s doing.


Playlist Popularity
Here’s an update on playlist popularity, using the last 7 days last of playtime. Things have shifted around a little bit.

Super Fiesta
Slayer
Heroic Warzone Firefight
Warzone
Castle Wars
Big Team Battle
Infection
HCS Fall 2017
Action Sack
SWAT
Legendary Warzone Firefight
Team Arena
Doubles
Warzone Turbo
Free-For-All
Breakout
Snipers
Mythic Warzone Firefight
Warzone Assault
Grifball
So it sounds like there won't be any restrictions on accounts that can play "Quickplay", similar to Halo 3's "Basic Training" which was only available to players with the Recruit rank. Do you think there may be any issues with this playlist being the only social playlist which offers a "classic halo 5" style of play (magnum/ar starts - if that's the format of the playlist) available? The others offer variants of Halo 5 play like snipers, fiesta, and griffball, which is fun. But I'm curious if players will want a more casual experience as opposed to playing their hardest in ranked games now that this playlist is available. I'm wondering if skilled players will flood it, making it difficult for new players to get their bearings. I'm sure matchmaking will try to avoid this situation based on skill, but I'm curious about your thoughts on this, as I know social matchmaking is somewhat looser.
ZaedynFel wrote:
To be clear, we aren’t yet using the new system, but it is able to analyze current matches. I’m able to see win predictions, kill predictions, and death predictions, and compare to them what really happened. I’ll see numbers like “Kill Prediction: 13.8, Actual: 14” often. It’s fun for a stats nerd like me to be honest. On average, in its current unfinished state, it’s able to predict how many kills each player will have per minute within less than 0.4 kills. It’s often right on the money.
This is really, really cool!

If you ever do a, "How this works" seminar thing that's local, I'ma be in the front row nerding out. :)
What are the next steps for the playlists constantly lingering at the bottom of the population list? Surely you can't justify keeping Grifball, Mythic Warzone Firefight, Snipers and Breakout around given the amount of playlists this gamer already has.
RhysWX wrote:
Mythic Warzone Firefight.
I mean, this only pops up on weekends so I'm not surprised it's not super populated over the period of a week.
ZaedynFel wrote:
Happy New Year!It’s good to be back! I have a few fun updates.
PlaylistsWe have a couple of playlist changes coming this week. First, as we announced last Friday, the HCS list will be updated to reflect the official HaloWC maps and modes. Head over to http://halo.gg for more info.

In addition, we’ll be adding a simple playlist called Quickplay at the top of our Social lists. The purpose of this list is largely so incoming players having an obvious starting point. The biggest feedback we get from new players is many if not the majority haven’t played any Halo in the past and have no idea what those playlist names mean. This should make it super clear where to start. The list will feature Slayer on Rig, Truth, and Coliseum. We’re keeping the maps to a minimum so folks aren’t overwhelmed, and using Slayer since it’s a straightforward game mode for both new and solo players. The playlist will also have a maximum party size of 2. If you’re looking for some light Slayer gameplay with the chance to mentor incoming folks, hop in!
Skill SystemAs some of you saw, I’ll be speaking about the new system we are in the process of integrating at this year’s GDC. We now have just enough of the plumbing built to see results coming back from matches. The accuracy is significantly higher than what we have today despite the fact there are a few key pieces we still haven’t hooked up. That accuracy will only grow as we finish.

To be clear, we aren’t yet using the new system, but it is able to analyze current matches. I’m able to see win predictions, kill predictions, and death predictions, and compare to them what really happened. I’ll see numbers like “Kill Prediction: 13.8, Actual: 14” often. It’s fun for a stats nerd like me to be honest. On average, in its current unfinished state, it’s able to predict how many kills each player will have per minute within less than 0.4 kills. It’s often right on the money.

I’ve already applied it to examining a few samples games. For example, one match included several smurfs who are Champs on their main accounts. Our current (older) system said that the smurfs would get destroyed because it couldn’t tell they were smurfs. The newer system correctly called the match even, and identified the abilities of the players. What this means is players will quickly find themselves matching where they should be, and opponents will get a more accurate CSR update.

On the accuracy side of things, I’m seeing increases of up to 10% in how well it predicts the winner. I’ve been in this business for a long time, and have never seen that kind of jump in predictive accuracy. Definitely not on data where we are already doing a solid job of creating close matches that should be hard to predict. When the system goes live, we could see drastic improvement in match quality.

I don’t have an ETA on when we’ll switch over to using the system, but just thought some of you would enjoy hearing how well it’s doing.
Playlist PopularityHere’s an update on playlist popularity, using the last 7 days last of playtime. Things have shifted around a little bit.

Super Fiesta
Slayer
Heroic Warzone Firefight
Warzone
Castle Wars
Big Team Battle
Infection
HCS Fall 2017
Action Sack
SWAT
Legendary Warzone Firefight
Team Arena
Doubles
Warzone Turbo
Free-For-All
Breakout
Snipers
Mythic Warzone Firefight
Warzone Assault
Grifball
Still no word on a solution to the smurfing? It's by far the worst part of matchmaking for me.
Chogun Rua wrote:
ZaedynFel wrote:
I’ve already applied it to examining a few samples games. For example, one match included several smurfs who are Champs on their main accounts. Our current (older) system said that the smurfs would get destroyed because it couldn’t tell they were smurfs. The newer system correctly called the match even, and identified the abilities of the players. What this means is players will quickly find themselves matching where they should be, and opponents will get a more accurate CSR update.

Still no word on a solution to the smurfing? It's by far the worst part of matchmaking for me.
I've left the information related to smurfing.
Chogun Rua wrote:
ZaedynFel wrote:
Happy New Year!It’s good to be back! I have a few fun updates.
PlaylistsWe have a couple of playlist changes coming this week. First, as we announced last Friday, the HCS list will be updated to reflect the official HaloWC maps and modes. Head over to http://halo.gg for more info.

In addition, we’ll be adding a simple playlist called Quickplay at the top of our Social lists. The purpose of this list is largely so incoming players having an obvious starting point. The biggest feedback we get from new players is many if not the majority haven’t played any Halo in the past and have no idea what those playlist names mean. This should make it super clear where to start. The list will feature Slayer on Rig, Truth, and Coliseum. We’re keeping the maps to a minimum so folks aren’t overwhelmed, and using Slayer since it’s a straightforward game mode for both new and solo players. The playlist will also have a maximum party size of 2. If you’re looking for some light Slayer gameplay with the chance to mentor incoming folks, hop in!
Skill SystemAs some of you saw, I’ll be speaking about the new system we are in the process of integrating at this year’s GDC. We now have just enough of the plumbing built to see results coming back from matches. The accuracy is significantly higher than what we have today despite the fact there are a few key pieces we still haven’t hooked up. That accuracy will only grow as we finish.

To be clear, we aren’t yet using the new system, but it is able to analyze current matches. I’m able to see win predictions, kill predictions, and death predictions, and compare to them what really happened. I’ll see numbers like “Kill Prediction: 13.8, Actual: 14” often. It’s fun for a stats nerd like me to be honest. On average, in its current unfinished state, it’s able to predict how many kills each player will have per minute within less than 0.4 kills. It’s often right on the money.

I’ve already applied it to examining a few samples games. For example, one match included several smurfs who are Champs on their main accounts. Our current (older) system said that the smurfs would get destroyed because it couldn’t tell they were smurfs. The newer system correctly called the match even, and identified the abilities of the players. What this means is players will quickly find themselves matching where they should be, and opponents will get a more accurate CSR update.

On the accuracy side of things, I’m seeing increases of up to 10% in how well it predicts the winner. I’ve been in this business for a long time, and have never seen that kind of jump in predictive accuracy. Definitely not on data where we are already doing a solid job of creating close matches that should be hard to predict. When the system goes live, we could see drastic improvement in match quality.

I don’t have an ETA on when we’ll switch over to using the system, but just thought some of you would enjoy hearing how well it’s doing.
Playlist PopularityHere’s an update on playlist popularity, using the last 7 days last of playtime. Things have shifted around a little bit.

Super Fiesta
Slayer
Heroic Warzone Firefight
Warzone
Castle Wars
Big Team Battle
Infection
HCS Fall 2017
Action Sack
SWAT
Legendary Warzone Firefight
Team Arena
Doubles
Warzone Turbo
Free-For-All
Breakout
Snipers
Mythic Warzone Firefight
Warzone Assault
Grifball
Still no word on a solution to the smurfing? It's by far the worst part of matchmaking for me.
Did you not read the article? It says they are currently building a system that will address it.
Bring back the social team objective type playlists with oddball, ctf, and assault. Rotate in some more voted on forge maps too. #MakeHaloGreatAgain
When is the current season going to end? Apologies if this has been mentioned somewhere else already and I just missed it.
I am really interested on this “quick play” feature. Overall this seems like an improvement.
Why can't you apply the 2 player restrictions for HCS so we stop matching to4s?
I feel that WZ Assault is near the bottom of the list because when people can't find a match in it, they assume it's still broken and then they move onto another playlist which just further reduces the population of WZ Assault. It's a self-fulfilling prophecy of sorts.
I'm looking forward to the Quickplay playlist though. I've been wanting a social playlist featuring classic 4v4 Halo gameplay for a while now.
Very Nerdy wrote:
Why can't you apply the 2 player restrictions for HCS so we stop matching to4s?
And then where do you propose those TO4 play?
Vidiopro wrote:
Bring back the social team objective type playlists with oddball, ctf, and assault. Rotate in some more voted on forge maps too. #MakeHaloGreatAgain
Yeah, we need another playlist to sit under Grifball.
Very Nerdy wrote:
Why can't you apply the 2 player restrictions for HCS so we stop matching to4s?
And then where do you propose those TO4 play?
Halo is a team game though and HCS is the most team based playlist so why would you remove the team aspect?
DoubleSama wrote:
Very Nerdy wrote:
Why can't you apply the 2 player restrictions for HCS so we stop matching to4s?
And then where do you propose those TO4 play?
Halo is a team game though and HCS is the most team based playlist so why would you remove the team aspect?
....yes? And I was asking Very Nerdy where he expects the full teams to play if they were to restrict HCS to 2 players per lobby.

Did I miss something?
DoubleSama wrote:
Very Nerdy wrote:
Why can't you apply the 2 player restrictions for HCS so we stop matching to4s?
And then where do you propose those TO4 play?
Halo is a team game though and HCS is the most team based playlist so why would you remove the team aspect?
....yes? And I was asking Very Nerdy where he expects the full teams to play if they were to restrict HCS to 2 players per lobby.

Did I miss something?
No, I was agreeing with you. Sorry for the confusion. Waypoint wouldn't let me quote him when editing my previous post so I ended up quoting you as well instead.
Too many playlist period. That is the main problem and those who actually want a social experience of the game to play with their friends without tryharding in ranked are forced to the only comparable playlist BTB. Why not go back to having a social slayer and social obj and then having ranked versions. I don't enjoy having to hop into ranked when i only really play the game a few times a month. I want the core experience without having to jump into a ranked match.

With the number of playlist you seem to be just fragmenting the player base even further. Use some of those other playlist as like a weekend special here and there. With the way the game sits now, I just don't see myself playing the game more than a couple times a month as there seems to be no real focus on what can be done to build the playerbase back up and just throwing out playlist to appease small groups of players. Simplifying makes the game far more accessible rather than just a quickplay playlist. Go to 4-5 social with an additional rotating playlist and 4-5 ranked play list.
When will this quick play be in the game? Also, what’s the mmr thing gonna be like in this playlist? Really strict or loosened up?
  1. 1
  2. ...
  3. 2
  4. 3
  5. ...
  6. 8