Matchmaking skill matching

Does anyone else feel like this game’s idea of an even team is one really good person and three awful people? Like it’s not trying to find teams of players with even skill levels, it’s trying to find a proportionate amount of good and bad? It feels like 80 percent of the games I play in team slayer my team loses with the post game stats for my team looking like:

Me: +12

Random: -3

Random: -5

Random: -14

Anyone else experience this a lot?

Been getting this since even before the weapon tuning update. I played Halo 4 for a long time because there were playlists such as SWAT, Team Snipers, and Team Doubles (Doubles Pro) that don’t use Infinity settings, but the bad skill matches is what pushed me off the cliff. I can’t stand losing games and dying constantly because the game matches me with players who are worse than me against players who are better than me.

I could go through my Game History and show you game after game of blatantly imbalanced CSRs on the teams.

> Been getting this since even before the weapon tuning update. I played Halo 4 for a long time because there were playlists such as SWAT, Team Snipers, and Team Doubles (Doubles Pro) that don’t use Infinity settings, but the bad skill matches is what pushed me off the cliff. I can’t stand losing games and dying constantly because the game matches me with players who are worse than me against players who are better than me.
>
> I could go through my Game History and show you game after game of blatantly imbalanced CSRs on the teams.

Tell me about it. Since starting this thread I’ve played 2 team slayer games. The first we lost and the stats for my team were:

Me: +4
Random: -8
Random: -8
Random: -6

The most recent game my team won, but it was despite my teammates. That one looked like:

Me: +11
Random: -1
Random: +1
Random: -3

It feels like it’s almost every team slayer game.

I’m pretty sure the game does this intentionally to try and “balance” things out

so you have eight players yeah?

P1 - Good player
p2 - decent player
p3 - decent layer
p4 - less than average
p5 - decent player
p6 - less than average
p7 - less than average
p8 - less than average

The teams would then look a lot like this

Red team
p1,p6,p7,p8

Blue team
p2,p3,p4,p5

They do this because the matchmaking system believes that the “good” player should be able to beat a team of decent players because that player is by large much better statistically, it doesn’t take into account how bad the other members of the team are and how important team work really is

I played a game a few nights ago when I went 30 kills 8 assists and 14 deaths and lost the game

If a player on your team scores 30 kills your team should win that game no questions

Then again I could be completely wrong maybe the system doesn’t work that way

> I’m pretty sure the game does this intentionally to try and “balance” things out
>
> so you have eight players yeah?
>
> P1 - Good player
> p2 - decent player
> p3 - decent layer
> p4 - less than average
> p5 - decent player
> p6 - less than average
> p7 - less than average
> p8 - less than average
>
> The teams would then look a lot like this
>
> Red team
> p1,p6,p7,p8
>
> Blue team
> p2,p3,p4,p5
>
> They do this because the matchmaking system believes that the “good” player should be able to beat a team of decent players because that player is by large much better statistically, it doesn’t take into account how bad the other members of the team are and how important team work really is
>
> I played a game a few nights ago when I went 30 kills 8 assists and 14 deaths and lost the game
>
> If a player on your team scores 30 kills your team should win that game no questions
>
> Then again I could be completely wrong maybe the system doesn’t work that way

That’s absolutely how it feels. It’s been a while since I’ve played Halo 3 so I could be wrong, but it felt like that game attempted to match entire teams of similarly skilled people. Or at least it felt like that was the end result more often. But I wonder if Halo 4 attempts to match teams out of unevenly skilled people as a result of a lower population. That sort of makes sense at a basic level.

How does the system differentiate between a good player or an average player?
Is it based purely on CSR, or are there other factors taken into account?
Such as SR or K/D.
I do see some messed up games i.e. CSR 1 against a CSR 43.

In this game and reach, it felt really lopsided as far as teams went. One team would dominate, and the other team would get blown out. It was better in Reach, though, I’ll say that much. IDK if theMorningStar is right with what he’s saying, but it definitely feels that way.

> How does the system differentiate between a good player or an average player?
> Is it based purely on CSR, or are there other factors taken into account?
> Such as SR or K/D.
> I do see some messed up games i.e. CSR 1 against a CSR 43.

I was under the impression that (at least initially) CSR wasn’t/isn’t used for matchmaking. Everyone has always had a trueskill rank in Halo 4. It just wasn’t visible. CSR was added as a visual representation (though not exact, as it’s calculated differently) of everyone’s trueskill rank. I’m not sure if they have recently started using CSR for matchmaking, though.

> How does the system differentiate between a good player or an average player?
> Is it based purely on CSR, or are there other factors taken into account?
> Such as SR or K/D.
> I do see some messed up games i.e. CSR 1 against a CSR 43.

I’m not sure but I don’t think it’s CSR. It’s probably each person’s invisible trueskill rank. Either way, it’s still doing it to me. I just played my first team slayer game since last night and my team lost. Our post game stats were:

Me: 12
Random: -6
Random: 0
Random: -15
Random: -2

We had a guy who was going negative drop out of the game halfway through and it replaced him with someone who went even farther negative. I hate everything.

> How does the system differentiate between a good player or an average player?
> Is it based purely on CSR, or are there other factors taken into account?
> Such as SR or K/D.
> I do see some messed up games i.e. CSR 1 against a CSR 43.

The CSR system is meant to match people of a generally close ranking (within 5 ranks) however because of the games low population that is near impossible to do which is why you often see such large discrepancies

As for the invisible trueskill
I believe its calculated off w/l ratio and your K=A/D ratio (kills+assists/deaths), however I could be wrong

Unfortunately, the population is low, and it would take a long, long time to accumulate players in a narrow skill band and end up with players close to your skill.

I’d probably be the -3 guy in your scenario, but I can tell you that your situation is better that being on the wrong end of a game where the winning side is a party of CSR20-40, and the losing side is a bunch of randoms from CSR5-15.

The game is trying to take the available players during that time slice and average them out. Given the low numbers, it will probably continue.

My suggestion is find some (slightly) weaker players that you can stand and friend them so you know what you are getting into.