343 Please Explain Placement Game Rank System

Played 10 placement matches in Team Doubles with a friend and this is what happened.

We won 6 of 10.
Friends KDA 2.3, Mine 6.9.
Friend had 113 total kills, I had 131.
He had more kills than me in 3 of 10 games.
Friends rank: Platinum 3
My rank: Gold 6

There seems to be a discrepancy here. I would like to understand a little better why it worked out this way. Thanks.

I wonder if they also used ur invusible rank for doubles before this season to calculate ur rank as well as ur placement games. I know mmr carries over across seasons so maybe it happened here

What were you rankings on the other playlists before this? And your friends? I’m curious as well. I know they do carry weight so people don’t smurf. But sounds to me like you should have ranked higher regardless.

Basically when you start a match (ranked or qualifying) the system randomly picks X amount of players and throws you all together. It doesn’t matter what CSR or level you are. It’s easier to not try and understand it because it’s broken as hell and not even 343 knows how it works.

Don’t remember my previous rank. The playlist was gone since January. Now I placed in Team Slayer Onyx 1700 and finished with a 0.0 KDA. Had a bad night between matching champions and people with ridiculous connections and don’t feel I deserve that rank at all. I’d like to keep this conversation going so 343 takes a hard look at the ranking system for Halo 6 and/or perhaps a classic map playlist in Halo 5.

TripleTap117 I too feel the system is broken. Rank almost feels predetermined. What are some possible alternatives for a ranking system?

I don’t know if the pool is big enough but keeping solos and teams separate would help a lot. I usually play solo and seems like more times than not I get stomped by a team and of course no one uses a mic on my team.

> 2757114414746464;5:
> Don’t remember my previous rank. The playlist was gone since January. Now I placed in Team Slayer Onyx 1700 and finished with a 0.0 KDA. Had a bad night between matching champions and people with ridiculous connections and don’t feel I deserve that rank at all. I’d like to keep this conversation going so 343 takes a hard look at the ranking system for Halo 6 and/or perhaps a classic map playlist in Halo 5.
>
> TripleTap117 I too feel the system is broken. Rank almost feels predetermined. What are some possible alternatives for a ranking system?

I’d be ok with CSR setup and maybe even seasons…if it worked. I can’t stress how irritating it is to match against unranked players, regardless of whether or not previous rank is taken into account. My time is spent in FFA 99% so I’m not sure about other playlists. There is no shortage of qualifiers so I don’t think it’s far fetched to separate ranked and qualifying matches. I still think a better alternative is to go back to the system used in H3.

I definitely understood 1-50 better. This current system needs to be more clearly defined to work.

the rank system is just broken AF.

http://halotracker.com/h5/csrbreakdown Onyx is supposed to be for the top 1-2% of the population lol

Well I came 5th in most of my placements in ffa with one win and a few 3rds and 4ths and came onyx 1581. But i was onyx 2059 at the end of last season with an MMR presumably much higher as i would match top 10 ffa champion players last season like jimbo, shystie, spxrrrkkky, rippin etc. most games and other high champs (so my mmr was prob closer to onyx 2300 or so). And in turn in my placements i was playing slightly lower champs, but still champs. Thats a huge mmr they have so my mmr was always going to put me higher.

I think csr doesnt mean as much as your hidden mmr. I mean diamond placements prob means your hidden mmr is high onyx (1800-2000 or so), and your csr just needs to catch up. In fact your csr always starts off way lower than your mmr- thats why initially you gain car from 5th place, as the system needs to catch your csr up to your mmr quickly. Look at those qualifying onyx 1700- most have had csrs in excess of 2500 in ffa so of course there is a disconnect between mmr and csr and thr system has to close that gap. Someone closer to their true mmr thrn needs 3rds and so on to increase not 5ths-hence why some higher ranks can gain more csr than the lower ranked player finishing above them (if the lower ranked player is close csr wise to their mmr), which is mad but its the mmr causing the problem.

I always think its worth looking at 2 things when it comes to looking at ranks; number of games played and your opponents.

1). No of games- a diamond 3 player after 20 games is likely to have a higher mmr than a diamond 3 player having played 200 since the first player still needs to play loads more games for their csr to catcg up with their mmr. The second player has already done this. This is why comparing csr isnt always as meaningful.

2). Opponents- a diamond 1 player matching high onyxes most games will probs have a higher mmr than a diamond 1 matching plats. Seeing as mmr matches players more than csr it makes sense.

Take home message is dont just rely on csr! It doesnt tell the whole story!

> 2533274823886169;11:
> Well I came 5th in most of my placements in ffa with one win and a few 3rds and 4ths and came onyx 1581. But i was onyx 2059 at the end of last season with an MMR presumably much higher as i would match top 10 ffa champion players last season like jimbo, shystie, spxrrrkkky, rippin etc. most games and other high champs (so my mmr was prob closer to onyx 2300 or so). And in turn in my placements i was playing slightly lower champs, but still champs. Thats a huge mmr they have so my mmr was always going to put me higher.
>
> I think csr doesnt mean as much as your hidden mmr. I mean diamond placements prob means your hidden mmr is high onyx (1800-2000 or so), and your csr just needs to catch up. In fact your csr always starts off way lower than your mmr- thats why initially you gain car from 5th place, as the system needs to catch your csr up to your mmr quickly. Look at those qualifying onyx 1700- most have had csrs in excess of 2500 in ffa so of course there is a disconnect between mmr and csr and thr system has to close that gap. Someone closer to their true mmr thrn needs 3rds and so on to increase not 5ths-hence why some higher ranks can gain more csr than the lower ranked player finishing above them (if the lower ranked player is close csr wise to their mmr), which is mad but its the mmr causing the problem.
>
> I always think its worth looking at 2 things when it comes to looking at ranks; number of games played and your opponents.
>
> 1). No of games- a diamond 3 player after 20 games is likely to have a higher mmr than a diamond 3 player having played 200 since the first player still needs to play loads more games for their csr to catcg up with their mmr. The second player has already done this. This is why comparing csr isnt always as meaningful.
>
> 2). Opponents- a diamond 1 player matching high onyxes most games will probs have a higher mmr than a diamond 1 matching plats. Seeing as mmr matches players more than csr it makes sense.
>
> Take home message is dont just rely on csr! It doesnt tell the whole story!

I can run with that. My issue is why hide stats and parameters that actually matter. They should showing us what we actually need to focus on to rank up!

> 2533274807278095;12:
> > 2533274823886169;11:
> > Well I came 5th in most of my placements in ffa with one win and a few 3rds and 4ths and came onyx 1581. But i was onyx 2059 at the end of last season with an MMR presumably much higher as i would match top 10 ffa champion players last season like jimbo, shystie, spxrrrkkky, rippin etc. most games and other high champs (so my mmr was prob closer to onyx 2300 or so). And in turn in my placements i was playing slightly lower champs, but still champs. Thats a huge mmr they have so my mmr was always going to put me higher.
> >
> > I think csr doesnt mean as much as your hidden mmr. I mean diamond placements prob means your hidden mmr is high onyx (1800-2000 or so), and your csr just needs to catch up. In fact your csr always starts off way lower than your mmr- thats why initially you gain car from 5th place, as the system needs to catch your csr up to your mmr quickly. Look at those qualifying onyx 1700- most have had csrs in excess of 2500 in ffa so of course there is a disconnect between mmr and csr and thr system has to close that gap. Someone closer to their true mmr thrn needs 3rds and so on to increase not 5ths-hence why some higher ranks can gain more csr than the lower ranked player finishing above them (if the lower ranked player is close csr wise to their mmr), which is mad but its the mmr causing the problem.
> >
> > I always think its worth looking at 2 things when it comes to looking at ranks; number of games played and your opponents.
> >
> > 1). No of games- a diamond 3 player after 20 games is likely to have a higher mmr than a diamond 3 player having played 200 since the first player still needs to play loads more games for their csr to catcg up with their mmr. The second player has already done this. This is why comparing csr isnt always as meaningful.
> >
> > 2). Opponents- a diamond 1 player matching high onyxes most games will probs have a higher mmr than a diamond 1 matching plats. Seeing as mmr matches players more than csr it makes sense.
> >
> > Take home message is dont just rely on csr! It doesnt tell the whole story!
>
>
> I can run with that. My issue is why hide stats and parameters that actually matter. They should showing us what we actually need to focus on to rank up!

I agree. The only time the rankings seemed to match mmr was the january season where you could place at say onyx 2400 straight away so it was clear what your mmr seemed to be. Now with ranking caps and needing to keep populations high mmr is more hiddrn than ever in ways that make csrs hard to compare. Now you need to.play 200 odd games a season to be sure what your true rank is, which is crazy! If mmrs were visible placements would make so much more sense and i cant see why they arent in use at all.

> 2533274823886169;13:
> > 2533274807278095;12:
> > > 2533274823886169;11:
> > > Well I came 5th in most of my placements in ffa with one win and a few 3rds and 4ths and came onyx 1581. But i was onyx 2059 at the end of last season with an MMR presumably much higher as i would match top 10 ffa champion players last season like jimbo, shystie, spxrrrkkky, rippin etc. most games and other high champs (so my mmr was prob closer to onyx 2300 or so). And in turn in my placements i was playing slightly lower champs, but still champs. Thats a huge mmr they have so my mmr was always going to put me higher.
> > >
> > > I think csr doesnt mean as much as your hidden mmr. I mean diamond placements prob means your hidden mmr is high onyx (1800-2000 or so), and your csr just needs to catch up. In fact your csr always starts off way lower than your mmr- thats why initially you gain car from 5th place, as the system needs to catch your csr up to your mmr quickly. Look at those qualifying onyx 1700- most have had csrs in excess of 2500 in ffa so of course there is a disconnect between mmr and csr and thr system has to close that gap. Someone closer to their true mmr thrn needs 3rds and so on to increase not 5ths-hence why some higher ranks can gain more csr than the lower ranked player finishing above them (if the lower ranked player is close csr wise to their mmr), which is mad but its the mmr causing the problem.
> > >
> > > I always think its worth looking at 2 things when it comes to looking at ranks; number of games played and your opponents.
> > >
> > > 1). No of games- a diamond 3 player after 20 games is likely to have a higher mmr than a diamond 3 player having played 200 since the first player still needs to play loads more games for their csr to catcg up with their mmr. The second player has already done this. This is why comparing csr isnt always as meaningful.
> > >
> > > 2). Opponents- a diamond 1 player matching high onyxes most games will probs have a higher mmr than a diamond 1 matching plats. Seeing as mmr matches players more than csr it makes sense.
> > >
> > > Take home message is dont just rely on csr! It doesnt tell the whole story!
> >
> >
> > I can run with that. My issue is why hide stats and parameters that actually matter. They should showing us what we actually need to focus on to rank up!
>
>
> I agree. The only time the rankings seemed to match mmr was the january season where you could place at say onyx 2400 straight away so it was clear what your mmr seemed to be. Now with ranking caps and needing to keep populations high mmr is more hiddrn than ever in ways that make csrs hard to compare. Now you need to.play 200 odd games a season to be sure what your true rank is, which is crazy! If mmrs were visible placements would make so much more sense and i cant see why they arent in use at all.

Just one of the many mysteries that is 343. 200 matches is pretty steep. Maybe in 2 months I could do that, but with 1 month seasons that will probably never happen for me

> 2533274807278095;14:
> > 2533274823886169;13:
> > > 2533274807278095;12:
> > > > 2533274823886169;11:
> > > > Well I came 5th in most of my placements in ffa with one win and a few 3rds and 4ths and came onyx 1581. But i was onyx 2059 at the end of last season with an MMR presumably much higher as i would match top 10 ffa champion players last season like jimbo, shystie, spxrrrkkky, rippin etc. most games and other high champs (so my mmr was prob closer to onyx 2300 or so). And in turn in my placements i was playing slightly lower champs, but still champs. Thats a huge mmr they have so my mmr was always going to put me higher.
> > > >
> > > > I think csr doesnt mean as much as your hidden mmr. I mean diamond placements prob means your hidden mmr is high onyx (1800-2000 or so), and your csr just needs to catch up. In fact your csr always starts off way lower than your mmr- thats why initially you gain car from 5th place, as the system needs to catch your csr up to your mmr quickly. Look at those qualifying onyx 1700- most have had csrs in excess of 2500 in ffa so of course there is a disconnect between mmr and csr and thr system has to close that gap. Someone closer to their true mmr thrn needs 3rds and so on to increase not 5ths-hence why some higher ranks can gain more csr than the lower ranked player finishing above them (if the lower ranked player is close csr wise to their mmr), which is mad but its the mmr causing the problem.
> > > >
> > > > I always think its worth looking at 2 things when it comes to looking at ranks; number of games played and your opponents.
> > > >
> > > > 1). No of games- a diamond 3 player after 20 games is likely to have a higher mmr than a diamond 3 player having played 200 since the first player still needs to play loads more games for their csr to catcg up with their mmr. The second player has already done this. This is why comparing csr isnt always as meaningful.
> > > >
> > > > 2). Opponents- a diamond 1 player matching high onyxes most games will probs have a higher mmr than a diamond 1 matching plats. Seeing as mmr matches players more than csr it makes sense.
> > > >
> > > > Take home message is dont just rely on csr! It doesnt tell the whole story!
> > >
> > >
> > > I can run with that. My issue is why hide stats and parameters that actually matter. They should showing us what we actually need to focus on to rank up!
> >
> >
> > I agree. The only time the rankings seemed to match mmr was the january season where you could place at say onyx 2400 straight away so it was clear what your mmr seemed to be. Now with ranking caps and needing to keep populations high mmr is more hiddrn than ever in ways that make csrs hard to compare. Now you need to.play 200 odd games a season to be sure what your true rank is, which is crazy! If mmrs were visible placements would make so much more sense and i cant see why they arent in use at all.
>
>
> Just one of the many mysteries that is 343. 200 matches is pretty steep. Maybe in 2 months I could do that, but with 1 month seasons that will probably never happen for me

I think thats why many like longer seasons. It allows players to catch up mmr and csr wise to those who got handed all of theirs more easily. Problem is there is inequality still from.the jan season-those who got all their mmr from the jan season and placed well only needed 10 games whilst people playing now need 200 games to get that mmr and have a rank to reflect it. Also players with higher mmrs gain csr far easier than those with lower mmrs due to the system trying to catch them up to their true rank.

Basically the csr system is a joke and mmr should be made visible. Most csr ranks dont reflect their mmr and that means the visible ranks are a bit unsafe.

I can’t argue with that. I been thinking the CSRs have been a joke from the start. I feel like I’d still favor Trueskill 1-50 better, but who knows what’s best at this point. Maybe a combination of the 2?! Either way I think all anybody really care about is consistent, fair matches. How can anybody hope to improve and retain new skills and tactics when the skill gap from one match to the next is all over the board. I’m definitely not taking anything away from you hitting champ and me or anybody else making onyx. It would hold more weight and just be nice to know where actually playing against the upper 1 or 2% the system claims we are.

> 2533274807278095;16:
> I can’t argue with that. I been thinking the CSRs have been a joke from the start. I feel like I’d still favor Trueskill 1-50 better, but who knows what’s best at this point. Maybe a combination of the 2?! Either way I think all anybody really care about is consistent, fair matches. How can anybody hope to improve and retain new skills and tactics when the skill gap from one match to the next is all over the board. I’m definitely not taking anything away from you hitting champ and me or anybody else making onyx. It would hold more weight and just be nice to know where actually playing against the upper 1 or 2% the system claims we are.

I agree, the dominant skill determiner should be what is use to rank players (mmr), not something else on top of that which is prone to failure (csr). I liked 1-50 but not the losing to 30s who were really 50s also trying to rank up and losing skill to them. The placement system is a great concept that just needed a little bit more work as it could be used to skip 1-40 for the 50 skill player, saving everyone else below it some pain. I am definitely open to 1-50 working in this way.

I personally prefer the halo 5 system but how it worked on the january season where after 10 games you could place 2500 csr or more if you were good enough or in fact any rank, i.e. uncapped. That way your mmr and csr were one and the same. And then when you went up its cos you earnt it for real rather than being given what you were owed (even in 5th place in ffa) like it is at the moment-plus you have realistic elo ranks too rather than the astronomical scores that you would never find in a million years in say chess or football (soccer) rankings.

> 2533274823886169;17:
> > 2533274807278095;16:
> > I can’t argue with that. I been thinking the CSRs have been a joke from the start. I feel like I’d still favor Trueskill 1-50 better, but who knows what’s best at this point. Maybe a combination of the 2?! Either way I think all anybody really care about is consistent, fair matches. How can anybody hope to improve and retain new skills and tactics when the skill gap from one match to the next is all over the board. I’m definitely not taking anything away from you hitting champ and me or anybody else making onyx. It would hold more weight and just be nice to know where actually playing against the upper 1 or 2% the system claims we are.
>
>
> I agree, the dominant skill determiner should be what is use to rank players (mmr), not something else on top of that which is prone to failure (csr). I liked 1-50 but not the losing to 30s who were really 50s also trying to rank up and losing skill to them. The placement system is a great concept that just needed a little bit more work as it could be used to skip 1-40 for the 50 skill player, saving everyone else below it some pain. I am definitely open to 1-50 working in this way.
>
> I personally prefer the halo 5 system but how it worked on the january season where after 10 games you could place 2500 csr or more if you were good enough or in fact any rank, i.e. uncapped. That way your mmr and csr were one and the same. And then when you went up its cos you earnt it for real rather than being given what you were owed (even in 5th place in ffa) like it is at the moment-plus you have realistic elo ranks too rather than the astronomical scores that you would never find in a million years in say chess or football (soccer) rankings.

Playing smurf accounts was definitely frustrating and I definitely agree that qualifying could save everyone some pain and irritation, if it worked. As another FFA player I’m curious to know if you face unranked players frequently too, or if being in the EU ( which country btw?) changes that. I know I’ve bought it up plenty. I’m only reiterating it because I feel like it basically creates the smurf effect every match, all season long.

The way the January season was, I definitely felt it to be the most consistent. It almost sounds like how your describing it would be a complete reset of csr and mmr every season which would give the elo feel along with not skew the placement results.

> 2533274823886169;2:
> I wonder if they also used ur invusible rank for doubles before this season to calculate ur rank as well as ur placement games. I know mmr carries over across seasons so maybe it happened here

Then what is the point of resetting the ranks?! It’s nonsense!

If they reset csr and mmr every season then a longer season would make sense. Since onyx 1700 is the cap for ranking in, players who take the two months to get to that level will then be matched against champion skill players early the next season. This has been my experience. The major difference I see when playing champs is that they (almost) always have a superior connection(I live in a high latency area). Makes the game more frustrating than just being beaten by a superior player.