Yeah, haha. Hey, if you’re getting assists and not dying, honestly I think that says more than a high KD in a team game. Even if it feels bad to get assist after assist.
1 Like
@Darwi yeah I need to clarify what I meant because I’ve even confused myself now 
It should absolutely be team based, meaning win, objective, supporting objective and kd should all be taken into account however individual skill leading to that win (or even loss) should also so people being carried do not rank up as much as the ‘carrier’.
I personally think things such as a flag distance stat in the scoreboard, showing how far you carried a successful cap as a percentage, flag carrier protection, oddball carrier protection, etc should all be taken into account but imagine the work involved in that
Although H5 wasn’t my favourite, I did play it a lot (forge kept me there), and never felt the rank caused anywhere near the issues as I see with infinite.
One thing I don’t like about personal performance though, is it doesn’t count assists nearly as much as kills. And if you’re running around with a disruptor or a plasma pistol and a teammate with a BR, the two of you can clean house as a pair, but the BR player will show much higher performance. Even though you both depended on each other for each kill.
Your point about objective performance I absolutely agree with though. And if objective performance matter more, maybe you’d have less people afraid to pickup the ball.
2 Likes
I think the problem is simple. The ranking system is designed for a Slayer playlist and we don’t have one.
3 Likes
Yeah I’ve always wondered if there’d be a different way with assists but may just complicate things.
One of my ideas was with the score.
200 points would be awarded for a kill but that would be broken down to 100 for the kill itself and 100 as a percentage of damage.
So if I killed someone with no help at all, I’d get 200 points.
If a teammate made someone absolute and I finished them with one tiny dot of a plasma pistol, they’d get 99 for 99 percent of the damage and I’d get 101, 100 for the kill, 1 for the 1%.
It may be complicated if the shield partly recharged then got stopped part way through etc. I don’t know, I gave it about a minutes thought
2 Likes
It’s definitely a good idea. I know it’s a hard thing to figure out.
1 Like
I’ve played a to a point where I figured out more how the system works. The 2 things that matter the most to rank up when winning is your placement within your own team (so 1st to 4th), and the number of kills. From 20+ kills and especially 30+, you can expect 1/4 of a bar almost. Deaths don’t matter as much but they do.
If you place 3rd or 4th in your team, your bar will barely move. That’s my experience so far. It’s pretty much the same thing in reverse for when you lose. I don’t think it’s an adequate system and you should rank up compared to your enemy team not your teammates.
1 Like
Getting a ton of kills is overrated in an objective based game. I’ve gotten 20+ quit a few times and yes my rank went up more than getting points for posting the objective
1 Like
Yup. This ranking system is straight up BS… just won three games in a row. Third game played oddball had the ball for 3 minutes 7 seconds, 6 kills 9 deaths, 3 carrier kills rank one up like 2 mm.
Looks like it’s a total waste of time to play the objective.
2 Likes
I get that but being realistic, unless you have a team already set up, the likelihood of everyone using their mics is slim therefore nullifying any teamwork.
I see what you’re saying but looking at the current ranked play, this chaos you speak doesn’t really happen. People are still playing the objective and some playing for themselves which hasn’t changed. This has happened with every Halo.
It’s fine to have a rank based on wins/losses. Groups can get together based on their ranks in a particular game type. But nonetheless, I want an additional rank that shows my skill and mine only. I can’t control how my teammates play at the end of the day micced up or not, but I can at least do my best.
I guess objective-based gameplay does matter on strongholds, cuz I had about the same stats and got ranked up a whole inch. Captured a lot of strongholds though
I agree, and in its current state, yes, time spent on the actual objective needs work in terms of how much rank you gain/lose.
Personally, I’d rather take a team full of good k/d’s rather than wins/losses. At least I know the player can handle themselves. I’ve been in groups where we carry the team and have the same guy always at the bottom but keeps pace with the groups rank. If that person were to queue alone, he wouldn’t have a good time since he’s getting paired with people well over his skill level.
They can keep the old state of rank, based on wins and losses. That will still make players put in enough teamwork to get that ranked up, but I’d like an additional rank showcasing my skill, not to be necessarily judged amongst my peers like the win/loss.
2 Likes
It’s definitely hard to gauge what the best way to get the most rank is. I feel like getting slot of kills and playing the objective gets you a lot but I’ve had times where I’ve gotten a small percentage.
Well that’s why KD and other stats would be nice. I mean, it wouldn’t solve all the problems, but if we had stats we could view and improve outside of wins and losses, I think I lot of people would care to work on it.
I still foundly remember Bad Company 2s stat tracker. I was proud that over 90% of my kills were with the defib. And it’s not like I was holding my team back by using it. Often times I’d be at the top of the scoreboard, or close to it. And that’s not counting the revives, just kills. It was fun caring about the stats. If we had stats tracked in game that’d be awesome.
1 Like
Yeah like maybe a win gives you the most XP and individual performance gives less but still stop you from losing overall XP.
It’s like you said, if it was calculated more precisely it could work.
1 Like
I feel like the KDA ratio is the most important factor in scoring how much you rank up if you win, and protects you from losing much rank if you lose. Which is ok for slayer-type matches, but not for CTF, stronghold or oddball.
I believe the score shown during game should have much more impact - if you get a sh*tton of slays, sure it may help the team, but holding the ball for whole match, or capturing many strongholds should give even more points to promote playing the objective and playing as a team.
The ranking system is pretty much based on wins and losses… so is really is designed for objective games.
There is a weighting for kills per minute… we don’t know how much… and hopefully not used outside of Slayer. But regardless, it is only a weighting, it won’t affect your ranking that much.
It shouldn’t be KDA - but kills per minute. You can read the paper published by Microsoft by literally googling Microsoft and TrueSkill2. The idea is that while 20:10 and 10:5 have the same K/D - the former is more likely to be a higher rank. It’s also based on the fact that K/D really just reflects the win.
No. The ranking works on the W/L. Playing the objective gives you the win.
How much you go up (or down) depends on the rank of the opposition team (you need to beat teams above you to rank up) and the shape of your skill curve (it narrows and locks you in over time as the system gains confidence in your rank).
Sounds like a job for ranked FFA.
But in a team game your skill is reflected in the team’s performance. You will get good teams (some will even carry you!) and bad teams. It will all wash out to an average team ranking that is your rank.
In the meantime… keep asking yourself what else could you have done to carry the team further.
Keep winning!
But realistically you will reach a ceiling that is hard to improve on.
Part of it will be your skill level. It’s hard to get better after the initial honeymoon period. Like, really hard. Otherwise we’d all be Onyx!
Part of it is the shape of your skill curve. It narrows as the system gets confident in your rank. This makes it harder to rank up but also makes it harder to rank down. It’s not as bad as Halo 3’s rank locking though. But it will slow the pace of any change.
And finally - the teams you play. To rank up you pretty much have to beat teams higher than you. If you are squadding up with pals you eventually become the big fish and the system struggles to genuinely find anyone significantly better than you. This leads to a pattern of small gains over a run of wins (the system doesn’t think you are improving but has to give you a token bit of CSR anyway) and then a chunky ranking down when you eventually lose (it brings you back to the rank you are).
We have to accept that the system falters at the high end. It just doesn’t get the playing data to accurately assign and sort everyone’s MMR. We all want a global rating to tell us where we stand… but unless there is a regular tournament style where all the best players play each other regularly - it just can’t do it. No system can.
You must be a dev or you don’t play the very game you’re talking about. The system is not fair and will make you lose after you win 2-3 games so you can’t rank up. And the number of kills is the only thing, or the biggest thing by a large margin, that affect your rank. It’s a known issue and it doesn’t work for objective games.
Edit: I played a couple games this afternoon (you can look it up) and I did relatively well. But after a few I got frustrated and got on another game because the system is playing me/you. As soon as I win 2-3 games they will inevitably put me on a team that play with their eyes closed so I return to my « expected rank ». I can’t do anything about it as a Solo Queue.
And of course when I lose, I lose about 4 times as much csr than when I win… I’m not the only one. If the system is working properly, then they should tweak it (which they said they will), because it’s unsatisfying and frustrating.
Love this.
I’ve been advocating for damage stats to replace K:D for a couple of years now.
If they broke damage up into four categories; damage given that led to a kill, damage given that was escaped, damage taken for kill, and damage taken but escaped. You could do so much with different ratios that reflect play above and beyond KD or KDA.
For example, you could concentrate on damage that led to a kill (which covers kills and assists) divided by your damage from deaths. You could even add in a small bonus for actually completing the kill (say 20%).
And you could do so much with end of game medals;
- MVP is the most damage done that led to deaths.
- Meet and Greet = most initial damage (that led to a kill).
- Finisher = most kills with the least amount of effective damage.
- Chef = most damage spread over different weapons.
- Escape artist = most damage taken for least number of deaths.
Hopefully the stats API will be made public soon and we can all have a play around with it.
2 Likes
These are all good ideas. It’s just outrageous how badly you go down when you lose compared to when you win, even with good personal results.
1 Like