Support NeoGAF

Crucio · Feb 23, 2011

A good example of unrespectable K/D: http://halo.xbox.com/en-us/stats/Gamesager

Devin Olsen · Feb 23, 2011

Cocopjojo said:
Right, but there are two problems with doing it by playlist:

1) It starts getting really complex, and from what I've read (and how I feel personally), Halo players want simplicity in a ranking system. We want to be able to understand how it works.

2) What playlists would we weight it by? SWAT? Team Slayer? What if there's someone who's really good but they don't like those playlists? Now we're hurting them by not picking their playlist.

I think weighting it by your global KD ratio is a pretty fair way to do it when looking at all of your options.

Thoughts?

I have literally no idea what goes on behind the scenes as far as true skill goes... But couldn't you fetch the true skill data from all of the play-lists and the take an average of the top three?

I guess this goes against what you're saying when you say you want simplicity.

My brutally honest opinion is giving a ranking based off of Global K/D only will simply cause the ranking to be irrelevant.

Edit - Sorry for posting so much, just trying to give constructive criticism.

Edit Again - If your top used weapon is a vehicle or a pro pipe (Dax) your rating should remain forever at 0.

ElRenoRaven · Feb 23, 2011

blamite said:
Looks like the Halo Waypoint website got a pretty big upgrade. It no longer uses silverlight, and it has it's own page for tracking your reach stats.
http://halo.xbox.com/en-us/stats/GANNONSMASH
Looks like the "Battle Proficiency Rating" is a 1-100 number meant to act like your overall skill level?

Wow. Looks like the transition is coming faster then we thought. I know Bungie said that eventually 343 would be handling all that stuff. Looks like this is the first step in that.

Forums are still unusable on Firefox though.

bobs...onGaf · Feb 23, 2011

Cocopjojo said:
Right, but there are two problems with doing it by playlist:

1) It starts getting really complex, and from what I've read (and how I feel personally), Halo players want simplicity in a ranking system. We want to be able to understand how it works.

2) What playlists would we weight it by? SWAT? Team Slayer? What if there's someone who's really good but they don't like those playlists? Now we're hurting them by not picking their playlist.

I think weighting it by your global KD ratio is a pretty fair way to do it when looking at all of your options.

Thoughts?

1) I think this is in response to complaints about the Arena system? Rather than simplicity I think its transparency which is what people want. If the system makes sense and seems like something you can consistently work towards then its good. If you can literally get benchmarks out of it and work towards them its brilliant, the problem with the Arena system was that it was insanely difficult to understand what it was doing with the data. It wasn't a linear system which involved doing well and ranking up... it was garbled, you could do well but still not rank up based on if your opponents where low (and you couldn't see what ranks your opponent where to know this) so no matter how you did you you never knew if that was going to translate into a good rating. Playing too many games even if you did well could potentially lower your rank... other players playing when you couldn't get on could lower your rank, it was just too hard to read. Trueskill is just too fickle.

If you could explain the system in a way that MLG takes your K/D and multiplys it by a higher value than BTB does then that would make more sense and would allow players to know how to achieve a good result.

2) I think you can generally tell which playlists people take seriously, why not just remove Zombies/ Grifball from the equation?

EDIT: Sorry if I sound like im coming off too strong, im just offering feedback. Im probably missing the bigger picture! I just really wanted to iterate that The Arena was partly underpopulated due to the fact there was no way you could really judge yourself against a benchmark, not because it was too complicated (A complicated system that you can understand and work towards is fine).

GhaleonEB · Feb 23, 2011

Cocopjojo said:
Right, but there are two problems with doing it by playlist:

1) It starts getting really complex, and from what I've read (and how I feel personally), Halo players want simplicity in a ranking system. We want to be able to understand how it works.

2) What playlists would we weight it by? SWAT? Team Slayer? What if there's someone who's really good but they don't like those playlists? Now we're hurting them by not picking their playlist.

I think weighting it by your global KD ratio is a pretty fair way to do it when looking at all of your options.

Thoughts?

Simple solution: provide a global average K/D as now, and a per-playlist one as well in the drill-down stats page.

I'm used to bungie's "all of the above" solution to the Service Record, you guys got a tall bar to meet. Don't envy ya.

The Librarian · Feb 23, 2011

For new page:

Cocopjojo said:
Right, but there are two problems with doing it by playlist:

1) It starts getting really complex, and from what I've read (and how I feel personally), Halo players want simplicity in a ranking system. We want to be able to understand how it works.

2) What playlists would we weight it by? SWAT? Team Slayer? What if there's someone who's really good but they don't like those playlists? Now we're hurting them by not picking their playlist.

I think weighting it by your global KD ratio is a pretty fair way to do it when looking at all of your options.

Thoughts?

I like global, but being able to see my K/D ratio for each playlist would be awesome.

I think a reason why global is good for your main K/D ratio is because people play the playlists they enjoy the most or they're good at. At least it's that way for me. I don't play SWAT or MLG because I'm bad at those gametypes. My K/D ratio shows that I'm a pretty reliable player in matchmaking for Invasion, Team Slayer, BTB, and Multi-team.

I hope that makes sense.

vhfive · Feb 23, 2011

Got an idea:
Have a playlist (or playlists) called "ranked" (or something similar) where you are given a number. Your number then goes up or down depending on if you win or lose. You are matched up against people who have the same (or near the same) number as you do. The theory being that you will eventually reach a number where you wont really go any higher or lower and you will be constantly matched up against people your skill level.

This may sound crazy but I think it may work.

Devin Olsen · Feb 23, 2011

vhfive said:
Got an idea:
Have a playlist (or playlists) called "ranked" (or something similar) where you are given a number. Your number then goes up or down depending on if you win or lose. You are matched up against people who have the same (or near the same) number as you do. The theory being that you will eventually reach a number where you wont really go any higher or lower and you will be constantly matched up against people your skill level.

This may sound crazy but I think it may work.

Sound stupid.

Give random (literally random, crazy, I know) amounts of numbers (we will call them credits) to people for finishing a game. Make sure we don't reward them for winning (that's not what this is about)...

Also make sure there are bars that slide.

--- I miss ranks, not entirely sure why they completely abandoned the idea.

Deleted member 21120 · Feb 23, 2011

vhfive said:
Got an idea:
Have a playlist (or playlists) called "ranked" (or something similar) where you are given a number. Your number then goes up or down depending on if you win or lose. You are matched up against people who have the same (or near the same) number as you do. The theory being that you will eventually reach a number where you wont really go any higher or lower and you will be constantly matched up against people your skill level.

This may sound crazy but I think it may work.

lol

Remember, this isn't Truskill or anything that's being used to match you in-game. I honestly don't have a problem with the way Bungie did it for Reach, I think it works as intended and I've never noticed any issues with getting matched up regularly against people who are much better or much worse than I am. It works great IMO. What we're talking about here is just a quick way to get a good idea of how good someone is at a glance.

I should also point out that I don't have control over how this formula works, but I can talk to the people who manage it.

I think it would be nice for everyone to have the KD that they feel is appropriate for them but like Dax said, he plays the playlists that he's good at. Those players are going to be upset if we weight their "official" KD ratio by the playlists that we decide are the "important" ones.

I agree that seeing KD ratio by playlist (like how B.net does) is good, and I don't see any reason why we couldn't put that in eventually. I honestly think that might be the best solution since that way you can always drill down and see how good someone is in "your" playlist.

bobs99 said:
1) I think this is in response to complaints about the Arena system? Rather than simplicity I think its transparency which is what people want. If the system makes sense and seems like something you can consistently work towards then its good. If you can literally get benchmarks out of it and work towards them its brilliant, the problem with the Arena system was that it was insanely difficult to understand what it was doing with the data. It wasn't a linear system which involved doing well and ranking up... it was garbled, you could do well but still not rank up based on if your opponents where low (and you couldn't see what ranks your opponent where to know this) so no matter how you did you you never knew if that was going to translate into a good rating. Playing too many games even if you did well could potentially lower your rank... other players playing when you couldn't get on could lower your rank, it was just too hard to read. Trueskill is just too fickle.

This wasn't a response to whatever complaints people have about the Arena system, that never even crossed my mind; the "simplicity" point is directed more toward how Halo players have felt about ranking in general over the years.

But again, we're not talking about matchmaking ranking, we're just talking about a number on a website that gives you a quick idea of a player's relative skill level.

Squall ASF · Feb 23, 2011

You can't really assess accurately how good someone is in Reach by just looking at their numbers. The fact that there isn't a clear cut 'Ranked' playlist where people play serious and the addition of challenges effect numbers in a negative manner. For instance, I *usually* kick serious ass in SWAT, but I go into Multi team to knock out challenges and sacrifice my k/d to do so quickly. Challenges like the jetpack ones aren't exactly conducive to your k/d ratio.

On the other side of the coin, someone could EASILY boost their k/d ratio by playing playlists like Multi Team and Invasion and just sitting back and killing without participating in the objective (Moreso the former). I can only imagine one's k/d ratio if you only played Multi-Team and went for objective types with no intention on playing the objective. You'd get between 30-50 kills everytime and should have less than 10 deaths. Then you have the rank titles that relate to how much you play, not how good you are. Really, you just have to play someone to determine how good they are.

GhaleonEB · Feb 23, 2011

Cocop, my big suggestion for you to pass on (or not) is this: don't try to create a new measure of player skill. The "battle proficiency rating" doesn't really add anything. We all speak the stats language we're used to after playing Halo over the years. Some new, merged and unclear method of parking a big number at the top of a service record is just going to add confusion. If you want to list a stat, list K/D. And for that matter, list the K/D for Firefight as well.

If you're going to migrate the b.net stats support to Waypoint, the team should obviously put their own stamp on it. But Reach wasn't made with some big number to show off ("I'm a 50") like Halo 3 was. Reach split that into stuff like Rank and Commendations and other supportive stats. No need to try and add one now.

Tawpgun · Feb 23, 2011

I sort of miss the ranks as well, but they too were broken. Being rank locked is dumb abd being shafted by terrible teammates was annoying. Not to mention the 1-50 system fostered a plague of boosters in all the playlists.

Team Double was unplayable due to the boosters.

But this system is terrible as well. Why the hell isn't there any sort of victory bonus? The MVP of the losing team should get a little bonus, but the winning team should get a huge bonus.

Tashi · Feb 23, 2011

Cocopjojo said:
Yeah I think that's what it is currently. What would you suggest? (This is a serious question, not rhetorical).

It's tough to say. I can see the appeal of putting all your stats into one easy number so that players and communities can compare their "Overall Skill" just like we're doing here. Obviously there's an issue. I don't want to sound like a douche bag but Dax has a higher rating than I do and everyone seems puzzled by it.

So according to the description it calculates, "Kills, Wins, Deaths, Assists and Betrayals across all matches played" That seems fine. I don't know how much weight there is for each stat in the calculation though. I don't think that's the real issue though. I think the real issue is that the playlists are so different that it's difficult to get an idea of how skilled a player is by just looking at it globally.

For example, when all of these new juniors started coming in and answering my questionnaire, I asked what their highest level achieved in Halo 3 was but more importantly, which playlist. I saw some Team Slayer 50's, Double Team 50. I'm not impressed by that at all. (No offense) I'd be more impressed by a 45 in the MLG playlist than I would be of a 50 in any other playlist. I know the competition is much greater in the MLG playlist.

I don't know if you could divide it up amongst each playlist either. If you do that then you might as well just add a number next to our gamertags in Reach and call it a day.

What if the players were put into divisions, much like Arena, that signifies what level skill they possess? Maybe this could be based on TrueSkill. The BPR could then be compared to players in the same division. I'm an 81 in Onyx, while Dax is an 84 in Gold or something. Basic stats don't work without context. It's far from accurate because it doesn't take into account the opponents. If I go to Dax's house (he's got a higher BPR rating) and play on his gamertag and hop in matchmaking, I'll destroy the opponents he's matching against (if True Skill is doing its job) And if Dax plays on my gamertag, he'll get wrecked on a regular basis.

But this solution I put out there sort of suggests that maybe it should be just based on Arena because it sounds perfect for it. But that can't happen, who really plays Arena any more? I also think Arena is flawed from the beginning. I think win % in a team game is crucial and should be heavily weighted.

What if each playlist had weighting towards the global BPR? You could have a standard playlist like Team Slayer and/or Team Objective and have it be 1.0. Then maybe SWAT could be .90. MLG could be 1.50. Snipers could be .95 and so on. That might still be a little janky though.

I'm not sure, it's very interesting though and I will definitely give it more thought.

Major Williams · Feb 23, 2011

:lol cocop your old post from 2007 is funny to read in hindsight.

A number for ranking is one thing, but I think your idea of a universal KD ratio as a measure of skill is great.

I know in CoD, it works. My friends who consistently pull a better K

over a long period of time truly are better than me at the game. Those who are lower are truly a little worse than me overall.

Count me on the plus side of K

as a measure of skill - even in objective games.

KidA Seven · Feb 23, 2011

Tashi0106 said:
It's tough to say. I can see the appeal of putting all your stats into one easy number so that players and communities can compare their "Overall Skill" just like we're doing here. Obviously there's an issue. I don't want to sound like a douche bag but Dax has a higher rating than I do and everyone seems puzzled by it.

So according to the description it calculates, "Kills, Wins, Deaths, Assists and Betrayals across all matches played" That seems fine. I don't know how much weight there is for each stat in the calculation though. I don't think that's the real issue though. I think the real issue is that the playlists are so different that it's difficult to get an idea of how skilled a player is by just looking at it globally.

For example, when all of these new juniors started coming in and answering my questionnaire, I asked what their highest level achieved in Halo 3 was but more importantly, which playlist. I saw some Team Slayer 50's, Double Team 50. I'm not impressed by that at all. (No offense) I'd be more impressed by a 45 in the MLG playlist than I would be of a 50 in any other playlist. I know the competition is much greater in the MLG playlist.

I don't know if you could divide it up amongst each playlist either. If you do that then you might as well just add a number next to our gamertags in Reach and call it a day.

What if the players were put into divisions, much like Arena, that signifies what level skill they possess? Maybe this could be based on TrueSkill. The BPR could then be compared to players in the same division. I'm an 81 in Onyx, while Dax is an 84 in Gold or something. Basic stats don't work without context. It's far from accurate because it doesn't take into account the opponents. If I go to Dax's house (he's got a higher BPR rating) and play on his gamertag and hop in matchmaking, I'll destroy the opponents he's matching against (if True Skill is doing its job) And if Dax plays on my gamertag, he'll get wrecked on a regular basis.

But this solution I put out there sort of suggests that maybe it should be just based on Arena because it sounds perfect for it. But that can't happen, who really plays Arena any more? I also think Arena is flawed from the beginning. I think win % in a team game is crucial and should be heavily weighted.

What if each playlist had weighting towards the global BPR? You could have a standard playlist like Team Slayer and/or Team Objective and have it be 1.0. Then maybe SWAT could be .90. MLG could be 1.50. Snipers could be .95 and so on. That might still be a little janky though.

I'm not sure, it's very interesting though and I will definitely give it more thought.

Yes, my 46 in MLG was harder to get to than my 50 in team slayer. MLG requires much more teamwork, and spans multiple game types. You had to be good at all in order to go anywhere in that playlist.

I agree that certain playlists should give different weightings to portray a player's accurate rating.

neoism · Feb 23, 2011

really loving the higher thread activity... Welcome juniors...

krakov · Feb 23, 2011

I think it's pretty safe to say that Halo reach would be a better game without any ability to track K/D.

Tashi · Feb 23, 2011

Major Williams said:
:lol cocop your old post from 2007 is funny to read in hindsight.

A number for ranking is one thing, but I think your idea of a universal KD ratio as a measure of skill is great.

I know in CoD, it works. My friends who consistently pull a better K over a long period of time truly are better than me at the game. Those who are lower are truly a little worse than me overall.

Count me on the plus side of K as a measure of skill - even in objective games.

That's interesting. Do you think that it's because matchmaking doesn't really exist in those games? You never really know who you're going to match against. Good players with high ranks, bad players with high ranks or just bad players all together.

Would you agree though that your K/D in Team Deathmatch is higher than it is in Team Hardcore though? I think that seems fair and is sort of the same issue.

The Librarian · Feb 23, 2011

GhaleonEB said:
Cocop, my big suggestion for you to pass on (or not) is this: don't try to create a new measure of player skill. The "battle proficiency rating" doesn't really add anything. We all speak the stats language we're used to after playing Halo over the years. Some new, merged and unclear method of parking a big number at the top of a service record is just going to add confusion. If you want to list a stat, list K/D. And for that matter, list the K/D for Firefight as well.

If you're going to migrate the b.net stats support to Waypoint, the team should obviously put their own stamp on it. But Reach wasn't made with some big number to show off ("I'm a 50") like Halo 3 was. Reach split that into stuff like Rank and Commendations and other supportive stats. No need to try and add one now.

I agree with Ghaleon on this. I don't think BPR should be in there. The game wasn't designed around a numerical rating like Halo 3. Unless you can update the game like that.

Tashi does bring up some good points. I think the problems he presents could be mitigated by, again, showing your K/D spread for each playlist. My Bnet profile shows a higher K/D spread than Tashi, but we can see our K/D spread for each playlist, so the "overall" number doesn't matter as much.

Tashi0106 said:
It's tough to say. I can see the appeal of putting all your stats into one easy number so that players and communities can compare their "Overall Skill" just like we're doing here. Obviously there's an issue. I don't want to sound like a douche bag but Dax has a higher rating than I do and everyone seems puzzled by it.

What if the players were put into divisions, much like Arena, that signifies what level skill they possess? Maybe this could be based on TrueSkill. The BPR could then be compared to players in the same division. I'm an 81 in Onyx, while Dax is an 84 in Gold or something. Basic stats don't work without context. It's far from accurate because it doesn't take into account the opponents. If I go to Dax's house (he's got a higher BPR rating) and play on his gamertag and hop in matchmaking, I'll destroy the opponents he's matching against (if True Skill is doing its job) And if Dax plays on my gamertag, he'll get wrecked on a regular basis.

1. I don't have a rating in the Arena (at least I shouldn't). Haven't played a single game of it. Haha. And I'm determined to keep it that way!
2. I like how you say no offense when commenting on ranks, but say no offense to me after all those comparisons.

Squall ASF · Feb 23, 2011

Playlist rank is the unanimous suggestion and what word work the best. Even in the Halo 3, someone would have to ask you what your 'highest' rank was actually in but people were still satisfied with that.

Willeth · Feb 23, 2011

Weighting it by K/D is utterly pointless for someone who plays Objective. If someone is -6 every game, but is constantly throwing himself at the flag to keep his timer alive, is helping his team far more than the guy at the back of the map sniping.

I'd suggest something that tracks win/loss by gametype, and some combination of those for playlists that have more than one gametype. Of course, that's pie in the sky stuff.

EDIT: Whoops, I misunderstood and thought we were revamping matchmaking. Never mind! I'd like to see a global Win/Loss up top.

While we're on Waypoint, I'd love for there to be a comparison function for your friends' stats, and for this page to be on console Waypoint.

Tashi · Feb 23, 2011

Dax01 said:
Tashi does bring up some good points. I think the problems he presents could be mitigated by, again, showing your K/D spread for each playlist. My Bnet profile shows a higher K/D spread than Tashi, but we can see our K/D spread for each playlist, so the "overall" number doesn't matter as much.

1. I don't have a rating in the Arena (at least I shouldn't). Haven't played a single game of it. Haha. And I'm determined to keep it that way!
2. I like how you say no offense when commenting on ranks, but say no offense to me after all those comparisons.

It's not even all about which playlist either. It has to take into account the skill of the opponents. If you start playing MLG, your k/d will surely decrease because the competition is better in there but it will level off as TrueSkill ranks you and matches you appropriately.

Agreed, I don't think Arena should be used. I think it presents some good ideas but the execution isn't there. That's important stat data that shouldn't just be thrown away though.

Every playlist should be like the Arena but secretly...maybe it already is

User 73706 · Feb 23, 2011

lol late I know, but still:
Favorite Gamemode: Campaign
Halo 1: Assault on the Control Room/Pillar of Autumn
Halo 2: Regret/Quarantine Zone/Gravemind
Halo 3: Entire game after Floodgate.
Halo 3 ODST: Mombasa Streets/NMPD HQ/Coastal Highway
Halo Reach: Long Night of Soalce/Pillar of Autumn
Overall: ?

xxjuicesxx · Feb 23, 2011

Dax01 said:
1. I don't have a rating in the Arena (at least I shouldn't). Haven't played a single game of it. Haha. And I'm determined to keep it that way!

Wouldn't wanna ruin that 1.7 k/d!

Devin Olsen · Feb 23, 2011

Tashi0106 said:
Agreed, I don't think Arena should be used. I think it presents some good ideas but the execution isn't there. That's important stat data that shouldn't just be thrown away though.

Every playlist should be like the Arena but secretly...maybe it already is

I was going to suggest if there was an arena style rating for every game you played (kept secret) then they should use that data.

But then I remembered Urk talking about the difficulty in calculating 'ranks' in Objective game types. So that makes me think there isn't an arena style rank/tracker for the other play-lists.

Also Grifball and living dead style play-lists would really mess with the formula, I am sure.

GhaleonEB · Feb 23, 2011

Devin Olsen said:
I was going to suggest if there was an arena style rating for every game you played (kept secret) then they should use that data.

But then I remembered Urk talking about the difficulty in calculating 'ranks' in Objective game types. So that makes me think there isn't an arena style rank/tracker for the other play-lists.

Also Grifball and living dead style play-lists would really mess with the formula, I am sure.

I think the Arena formula is a good example of why this kind of stuff just doesn't work for providing satisfactory metrics of player skill.

Keep it simple. Don't try to invent something like this, it will. Not. Work.

Steelyuhas · Feb 23, 2011

Devin Olsen said:
I was going to suggest if there was an arena style rating for every game you played (kept secret) then they should use that data.

But then I remembered Urk talking about the difficulty in calculating 'ranks' in Objective game types. So that makes me think there isn't an arena style rank/tracker for the other play-lists.

Also Grifball and living dead style play-lists would really mess with the formula, I am sure.

Speaking of those specialty gametypes, it would be nice if a playlist like Grifball didn't count towards your matchmaking K/D on Bnet. I mean you don't even shoot guns in the gametypes haha.

Tashi · Feb 23, 2011

Devin Olsen said:
I was going to suggest if there was an arena style rating for every game you played (kept secret) then they should use that data.

But then I remembered Urk talking about the difficulty in calculating 'ranks' in Objective game types. So that makes me think there isn't an arena style rank/tracker for the other play-lists.

Also Grifball and living dead style play-lists would really mess with the formula, I am sure.

Yea they couldn't figure it out. Bungie even said, if you've got a way to calculate it, let us know. But then they said there was important data they weren't even tracking in the game so it became pretty much impossible to do it because it was too late to put the system in place to track that data.

Grifball and Zombies are joke playlists, they shouldn't count at all, not for anything, ever.

What if the system was RPG style and gave EXP based on win/loss and how "strong" the opposing team was. I understand that sometimes you get carried and sometimes your teammates quit out or are idling but I don't know if you could rank an individual based on stats in a team game. How do you do that? edit: With 100% accuracy.

Also, we never got a detailed break down of the spawn system in Reach. I think Shishka mentioned it in here before but, well, you know. Anyway, I think that's important game information that could be used.

Steelyuhas · Feb 24, 2011

Tashi0106 said:
Yea they couldn't figure it out. Bungie even said, if you've got a way to calculate it, let us know. But then they said there was important data they weren't even tracking in the game so it became pretty much impossible to do it because it was too late to put the system in place to track that data.

Grifball and Zombies are joke playlists, they shouldn't count at all, not for anything, ever.

What if the system was RPG style and gave EXP based on win/loss and how "strong" the opposing team was. I understand that sometimes you get carried and sometimes your teammates quit out or are idling but I don't know if you could rank an individual based on stats in a team game. How do you do that? edit: With 100% accuracy.

Also, we never got a detailed break down of the spawn system in Reach. I think Shishka mentioned it in here before but, well, you know. Anyway, I think that's important game information that could be used.

That would be nice. I still get a little confused by Reach's spawn system at times. Seems much less predictable than H3's.

I don't really see how a fair ranking system could be made for objective games, there is just to many variables, many of which are not really traceable by stats. Speaking of which though, now that Arena will be Win/Loss starting next month, will it be possible for objective games to be implemented? Daily ratings (which are just for show now) would have to be disabled because obviously you wouldn't get a rating for objective gametypes.

bobs...onGaf · Feb 24, 2011

Cocopjojo said:
lol

Remember, this isn't Truskill or anything that's being used to match you in-game. I honestly don't have a problem with the way Bungie did it for Reach, I think it works as intended and I've never noticed any issues with getting matched up regularly against people who are much better or much worse than I am. It works great IMO. What we're talking about here is just a quick way to get a good idea of how good someone is at a glance.

I should also point out that I don't have control over how this formula works, but I can talk to the people who manage it.

I think it would be nice for everyone to have the KD that they feel is appropriate for them but like Dax said, he plays the playlists that he's good at. Those players are going to be upset if we weight their "official" KD ratio by the playlists that we decide are the "important" ones.

I agree that seeing KD ratio by playlist (like how B.net does) is good, and I don't see any reason why we couldn't put that in eventually. I honestly think that might be the best solution since that way you can always drill down and see how good someone is in "your" playlist.

This wasn't a response to whatever complaints people have about the Arena system, that never even crossed my mind; the "simplicity" point is directed more toward how Halo players have felt about ranking in general over the years.

But again, we're not talking about matchmaking ranking, we're just talking about a number on a website that gives you a quick idea of a player's relative skill level.

Fair enough.

I only brought it up because I would hate it if that was sort of a general opinion over at 343 which could lead to the next Halo game having a crappy overly simplistic ranking system to compensate for the fact the complicated one in Reach didn't work! As far as the BPM thing -as you said its just a nice feature, its not critical to gameplay and so should probably be kept simple. Its just that since you guys have 'powah' over the halo series now, I just wanted to put my opinion across that 'complicated but consistent' is a better approach than 'simple but not exactly 100%'.

GhaleonEB said:
Cocop, my big suggestion for you to pass on (or not) is this: don't try to create a new measure of player skill. The "battle proficiency rating" doesn't really add anything. We all speak the stats language we're used to after playing Halo over the years. Some new, merged and unclear method of parking a big number at the top of a service record is just going to add confusion. If you want to list a stat, list K/D. And for that matter, list the K/D for Firefight as well.

If you're going to migrate the b.net stats support to Waypoint, the team should obviously put their own stamp on it. But Reach wasn't made with some big number to show off ("I'm a 50") like Halo 3 was. Reach split that into stuff like Rank and Commendations and other supportive stats. No need to try and add one now.

As much as I would love a global BPM which also somehow incorporates Campaign and Firefight (this really would be overcomplicated lol) - Ghaleon speaks the truth - I also personally dont think Reach is well suited for this big overall number. I totally understand and love the reason for it and if it was done right it would be brilliant but as it stands its not exactly a good metric to use to compare players.

Karl2177 · Feb 24, 2011

Cocopjojo said:
Right, but there are two problems with doing it by playlist:

1) It starts getting really complex, and from what I've read (and how I feel personally), Halo players want simplicity in a ranking system. We want to be able to understand how it works.

2) What playlists would we weight it by? SWAT? Team Slayer? What if there's someone who's really good but they don't like those playlists? Now we're hurting them by not picking their playlist.

I think weighting it by your global KD ratio is a pretty fair way to do it when looking at all of your options.

Thoughts?

It all depends on what all you guys are tracking or can track. Mainly, if you can track the rate at which a person's KD spread and ratio go up/down, then you can get a better representation of their skill. A person's spread should increase at the rate of the KD ratio. I'm completely braindead right now, but a way of incorporating KD spread into the formula would improve it... I think.

Tashi0106 said:
Also, we never got a detailed break down of the spawn system in Reach. I think Shishka mentioned it in here before but, well, you know. Anyway, I think that's important game information that could be used.

This would be extremely nice.

feel · Feb 24, 2011

Account Attempt #4 said:
Hey what's up HaloGAF. After years of trying, I finally know what it feels like to be on "the other side" of the registration process, hence my username. Hopefully it's only temporary haha. I am quite casual when to comes to gaming. Not by my choice of course but rather my real life circumstances, writ family man. While I'm unable to log hours/days/months playing the game I love most (Halo), I spend idle time lurking GAF, watching rendered films/streams (<3 Tashi!), and try to never miss an MLG tourney stream. I just love the energy of the broadcasts and the level of gameplay.
One last thing I'd like to say is RIP BB. Even though I was stuck in read-only mode and didn't know him I really felt for his family's loss. I wanted to dedicate my first GAF post ever to his memory. Anyway, thanks and it's nice to be here!

I like you. Welcome.

blamite said:
Looks like the Halo Waypoint website got a pretty big upgrade. It no longer uses silverlight, and it has it's own page for tracking your reach stats.
http://halo.xbox.com/en-us/stats/GANNONSMASH
Looks like the "Battle Proficiency Rating" is a 1-100 number meant to act like your overall skill level?

This is very lovely. No Halo 3 though??

kylej said:
wape

I laughed

Domino Theory · Feb 24, 2011

Hypertrooper said:
No code. I guess Domino lives in SF and so he got an invite for the Microsoft press event.

Ding ding ding.

But I live in Cupertino which is just an hour away. So excited!

xxjuicesxx · Feb 24, 2011

Getting matched up with a wolfpack. I love it.

Striker · Feb 24, 2011

Tashi0106 said:
What if the players were put into divisions, much like Arena, that signifies what level skill they possess? Maybe this could be based on TrueSkill. The BPR could then be compared to players in the same division. I'm an 81 in Onyx, while Dax is an 84 in Gold or something. Basic stats don't work without context. It's far from accurate because it doesn't take into account the opponents. If I go to Dax's house (he's got a higher BPR rating) and play on his gamertag and hop in matchmaking, I'll destroy the opponents he's matching against (if True Skill is doing its job) And if Dax plays on my gamertag, he'll get wrecked on a regular basis.

Why so serious?

Tashi · Feb 24, 2011

Striker said:
Why so serious?

???

Cocop asked for serious suggestions...

Steelyuhas · Feb 24, 2011

xxjuicesxx said:
Getting matched up with a wolfpack. I love it.

Wolf Pack!

A27_StarWolf · Feb 24, 2011

Steelyuhas said:
Wolf Pack!

Dibs on leading the wolf pack

xxjuicesxx · Feb 24, 2011

http://www.youtube.com/watch?v=fdMxRG9Jol0&feature=related

Sai-kun · Feb 24, 2011

67 battle proficiency

does this mean i'm a bad kid?

Hydranockz · Feb 24, 2011

xxjuicesxx said:
http://www.youtube.com/watch?v=fdMxRG9Jol0&feature=related

Perfect!

Plywood · Feb 24, 2011

Tashi0106 said:
Every playlist should be like the Arena but secretly...maybe it already is

I'd have serious trouble believing that.

Steelyuhas · Feb 24, 2011

xxjuicesxx said:
http://www.youtube.com/watch?v=fdMxRG9Jol0&feature=related

"Tonight I make a toast" hahaha

Tashi · Feb 24, 2011

Plywood said:
I'd have serious trouble believing that.

Pshhhh I don't, at all. It was probably how they were going to handle all the playlists before Arena was invented.

Tunavi · Feb 24, 2011

I have 213.5 hours of Reach Playtime. I must say I got my $40 worth

Stripper13 · Feb 24, 2011

Wow. My profiency rating is 71. I'm pretty happy with that - though I'd like to know what it involves when calculating.

kylej · Feb 24, 2011

You see wolf pack, I see scary ass furries.

Domino Theory · Feb 24, 2011

My BP is 100. What does this mean?

User 73706 · Feb 24, 2011

Domino Theory said:
My BP is 100. What does this mean?

Step up your game, man. It use a top ten-list system.

Stripper13 · Feb 24, 2011

Domino Theory said:
My BP is 100. What does this mean?

Pics or it didn't happen.

Support NeoGAF

Halo: Reach |OT3| This Thread is Not a Natural Formation

Member

Member

Member

Member

Member

Banned

Member

Member

Deleted member 21120

Unconfirmed Member

Member

Member

Member

343i Lead Esports Producer

Member

Member

Member

Member

343i Lead Esports Producer

Banned

Member

Member

343i Lead Esports Producer

Banned

Member

Member

Member

Member

343i Lead Esports Producer

Member

Member

Member

Member

Crystal Dynamics

Member

Member

343i Lead Esports Producer

Member

Banned

Member

Banned

Member

NeoGAF's smiling token!

Member

343i Lead Esports Producer

Banned

Member

Banned

Crystal Dynamics

Banned

Member

Similar threads