• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

Halo: Reach |OT4| This Thread is Not Your Grave, But You Are Welcome In It

Brolic Gaoler

formerly Alienshogun
Tawpgun, saw you on Gears, if you want a buddy to play with just hit me up when you're feeling the itch. I love gears and I'm not half bad.
 
I really hate the betrayal system in Reach. I have shotgun, I shoot enemy, his shields drop. I shoot again and at the same time a teammate sword lunges at him. My shot kills teammate, I get booted. I was having a good match. ='(
 

Louis Wu

Member
urk said:
Guests, like a great many things, are far beyond the reach of TrueSkill.
lol - I really wasn't griping, just amazed at how bad the other team was.

However, it's not just about guests: Our team had a Major, a Brig, two Generals, a Field Marshall, and a Noble - that's a decent amount of experience.

The the other team had a Warrant Officer, a Commander, two Colonels, and a Field Marshall.

Doesn't seem like an even split of the experience... though I realize that experience doesn't DIRECTLY correlate to skill. (Since nobody was in a party with anyone else, we should probably assume that all players are ROUGHLY at the same level, TrueSkillwise, right? So why wouldn't the teaminator (whatever you call the system that picks teams) split up the group into more evenly experienced sections?)

I guess i just don't know how any of this works.
 

urk

butthole fishhooking yes
Louis Wu said:
I guess i just don't know how any of this works.

Not too many people do. The expectation is that TrueSkill ensures that every match must end in a draw. The reality is that it's simply an algorithm designed to measure perceived skill.

Experience is largely a non factor. In fact, Trueskill needs very few games to become confident about a player's skill, but the more games a player plays the more confident TrueSkill will become about its assessment.

TrueSkill will never be able to account for guests, streaky players, players with babies on their laps, or shitfaced players (or streaky, shitfaced players with babies on their laps who are, in fact, in the game as a guest). These factors result in "surprise" results that make the maths look like guesswork.

And of course, nobody is going to scramble to NeoGAF to boast about the string of ten consecutive games they just played where each match was close and everybody was happy. It's the one game where you run into the splitscreen dynamic duo and they suck monkey nuts that demands attention.

:)
 

JB1981

Member
The elites with the green cannon gun things are so fucking annoying on legendary!! Also I'm watching Halo Legends on Netflix and I'm quite enjoying it but the episode with the clutsy Spartan and weird Japanese slapstick comedy seemed really out of place lol
 

Havok

Member
JB1981 said:
The elites with the green cannon gun things are so fucking annoying on legendary!! Also I'm watching Halo Legends on Netflix and I'm quite enjoying it but the episode with the clutsy Spartan and weird Japanese slapstick comedy seemed really out of place lol
Always be jumping. It helps when dodging.
 

Gui_PT

Member
JB1981 said:
The elites with the green cannon gun things are so fucking annoying on legendary!! Also I'm watching Halo Legends on Netflix and I'm quite enjoying it but the episode with the clutsy Spartan and weird Japanese slapstick comedy seemed really out of place lol


It's supposed to feel out of place, that's why it's called Odd One Out. I like it though. It's short but entertaining.

Which is your favorite so far?
 

Tawpgun

Member
Alienshogun said:
Tawpgun, saw you on Gears, if you want a buddy to play with just hit me up when you're feeling the itch. I love gears and I'm not half bad.

Sure thing.

If you're not half bad you might be able to balance out me being half bad. Until I get my lost skillz back.

I used to be a torque boss.
 

Brolic Gaoler

formerly Alienshogun
A27 Tawpgun said:
Sure thing.

If you're not half bad you might be able to balance out me being half bad. Until I get my lost skillz back.

I used to be a torque boss.


I got 3 straight torque kills earlier today, felt good man.
 

Plywood

NeoGAF's smiling token!
Just experienced the worst most laggiest game in BTB, the host switched every 5-15 seconds (not exaggerating) there was 6 other gaffers with me and it was stressful but that isn't the rage inducing moment. The rage inducing moment is when WE were winning only 3 kills away. Black screen and we all get dropped out of the game and it counted as a loss.
 

Tashi

343i Lead Esports Producer
Steelyuhas said:
See I don't understand this. That was a 2 point slayer game. How is that a bad trueskill match? Yes, some players played very poorly, but that is always going to happen, people have bad games. It happens. You can't expect every player in every game to have a kill to death within a few of even.

No this guy didn't just have a bad game. He was outclassed. It was so obvious. Nobody goes -13 in a slayer game unless you're being out classed. It's superrr rare.
 

GhaleonEB

Member
RIP animated Soffy.
Tashi0106 said:
No this guy didn't just have a bad game. He was outclassed. It was so obvious. Nobody goes -13 in a slayer game unless you're being out classed. It's superrr rare.
I think his point was, the team was balanced overall. My understanding is TrueSkill tries to find similar skill players but overall it also tries to balance out teams of comparable overall skill. And it did that.
 

Havok

Member
Steelyuhas said:
See I don't understand this. That was a 2 point slayer game. How is that a bad trueskill match? Yes, some players played very poorly, but that is always going to happen, people have bad games. It happens. You can't expect every player in every game to have a kill to death within a few of even.
The guy that was doing poorly has a .86 k/d. This isn't a perfect measure of skill, I'll admit, but there is a massive indisputable skill gap between him and the rest of the team. Sure, the game was close, but that's not the right way to balance teams. You can't put a few good players with a bad one and expect that since they might be able to carry him to a victory that you've done a good job balancing. Individual average skills should be close between teams, not the averaged individual skills of a group of people.
 

Tashi

343i Lead Esports Producer
GhaleonEB said:
RIP animated Soffy.

I think his point was, the team was balanced overall. My understanding is TrueSkill tries to find similar skill players but overall it also tries to balance out teams of comparable overall skill. And it did that.

Ohhh I see. Hmmm, I guess if you look at it that way then it makes sense. However, I don't the guys who did poorly on the other team were as bad as our teammate. The stats say that but even if they didn't I could just tell. I was looking at this from my and the 2 others I was partied with. We had to carry hardcore just to win. That's why it seemed off.
 

GhaleonEB

Member
Dani said:
Did you flush him down the loo?
He was flushed for me.
Tashi0106 said:
Ohhh I see. Hmmm, I guess if you look at it that way then it makes sense. However, I don't the guys who did poorly on the other team were as bad as our teammate. The stats say that but even if they didn't I could just tell. I was looking at this from my and the 2 others I was partied with. We had to carry hardcore just to win. That's why it seemed off.
Yeah he was definitely the odd man out in the group performance wise. If the game wasn't that close it would be a case study in one guy getting tossed in with players way over his head. I really don't know what can cause that sort of thing to happen if the player has some decent experience behind him in the playlist, other than to guess the system couldn't find a perfect match on a player per player basis instead tried to balance the teams. I have no idea, though.
 

Ken

Member
Plywood said:
Just experienced the worst most laggiest game in BTB, the host switched every 5-15 seconds (not exaggerating) there was 6 other gaffers with me and it was stressful but that isn't the rage inducing moment. The rage inducing moment is when WE were winning only 3 kills away. Black screen and we all get dropped out of the game and it counted as a loss.

http://www.bungie.net/stats/reach/gamestats.aspx?gameid=564422883&player=Chocopolka

Why would the game give them first place when not all of us DNF'd and had a higher score in the end.

And good games to everyone I just played with.
 
Tashi0106 said:
Ohhh I see. Hmmm, I guess if you look at it that way then it makes sense. However, I don't the guys who did poorly on the other team were as bad as our teammate. The stats say that but even if they didn't I could just tell. I was looking at this from my and the 2 others I was partied with. We had to carry hardcore just to win. That's why it seemed off.

Okay I looked up that player's service record and I take back what I said. He has played 2 games in that playlist, so I have no idea why he was matched up in your game.

My point in general though was that individual players have bad games, it happens all the time, and the majority of the time (I see your complaint though), I don't understand players complaining about a 2 point slayer game.

Also, do keep in mind that if you don't select the skill search modifier, playlists such MLG and others with low player counts are not going to be able to give you great trueskill matches every time.
 

Tashi

343i Lead Esports Producer
GhaleonEB said:
He was flushed for me.

Yeah he was definitely the odd man out in the group performance wise. If the game wasn't that close it would be a case study in one guy getting tossed in with players way over his head. I really don't know what can cause that sort of thing to happen if the player has some decent experience behind him in the playlist, other than to guess the system couldn't find a perfect match on a player per player basis instead tried to balance the teams. I have no idea, though.

I see. Now I'm thinking though, what does True Skill take from this match? My 2 teammates and I did very well despite out 4th being horrible? Does it poo poo the other team for losing basically a 3v4? Does it do nothing because this was a very tight and competitive match? I wonder.

Steel said:
Okay I looked up that player's service record and I take back what I said. He has played 2 games in that playlist, so I have no idea why he was matched up in your game.

My point in general though was that individual players have bad games, it happens all the time, and the majority of the time (I see your complaint though), I don't understand players complaining about a 2 point slayer game.

Also, do keep in mind that if you don't select the skill search modifier, playlists such MLG and others with low player counts are not going to be able to give you great trueskill matches every time.

Ahhh, I knew it LOL

I understand that players have bad games. I go negative all the time lol. Sometimes I put up huge negative numbers. I could tell though that he was completely mismatched. I guess I'm complaining because if we had a decent 4th, that game is likely a blow out. I mean, whatever. Winning that game against those odds definitely pumped me up and increased my confidence.

Usually True Skill does a good enough job to where I don't have to put that modifier on. It also slows down search times. If I do put on the modifier, it's during the day and I prefer language. Otherwise, I just let it rock.

Urk mentioned it earlier. We point out when True Skill fucks up occasionally, like twice or 3 times a day but for the most part it's pretty solid.
 

urk

butthole fishhooking yes
Players with few games recorded in a given list are always going to get mismatched until TrueSkill can gather enough data to make the call. The only way to place them is to let them play. Remember the playlist resets in Halo 2? Feeding frenzy for good players.
 

neoism

Member
urk said:
Not too many people do. The expectation is that TrueSkill ensures that every match must end in a draw. The reality is that it's simply an algorithm designed to measure perceived skill.

Experience is largely a non factor. In fact, Trueskill needs very few games to become confident about a player's skill, but the more games a player plays the more confident TrueSkill will become about its assessment.

TrueSkill will never be able to account for guests, streaky players, players with babies on their laps, or shitfaced players (or streaky, shitfaced players with babies on their laps who are, in fact, in the game as a guest). These factors result in "surprise" results that make the maths look like guesswork.


And of course, nobody is going to scramble to NeoGAF to boast about the string of ten consecutive games they just played where each match was close and everybody was happy. It's the one game where you run into the splitscreen dynamic duo and they suck monkey nuts that demands attention.

:)




seriously laughed at this part for like 10 minutes... :p
 

Tashi

343i Lead Esports Producer
urk said:
Players with few games recorded in a given list are always going to get mismatched until TrueSkill can gather enough data to make the call. The only way to place them is to let them play. Remember the playlist resets in Halo 2? Feeding frenzy for good players.

So it was just unfortunate for my team.

Also, I totally read, "Feeding Frenzy" in the Halo announcers voice. LOL
 
urk said:
Players with few games recorded in a given list are always going to get mismatched until TrueSkill can gather enough data to make the call. The only way to place them is to let them play. Remember the playlist resets in Halo 2? Feeding frenzy for good players.

Wait I thought new players in a playlist started with a "1" or "low" trueskill, so they match up against other new players or scrubs to start out. Is this not true?



neoism said:
seriously laughed at this part for like 10 minutes... :p

Yeah same here that was gold.


Tashi0106 said:
So it was just unfortunate for my team.

Also, I totally read, "Feeding Frenzy" in the Halo announcers voice. LOL

Glad to know I'm not the only one that does that haha.

Also, power slayer is kind of cool except for one glaring problem. How the hell do I know what gives me power ups and then when I get them, how am I supposed to know what they are?

Also, still never played speedflag. The voters suck.
 

Tawpgun

Member
My god that laggy BTB game.

It was annoying. So annoying it was almost hilarious.

Also, great games of Invasion versus those cool other guys. It's always great to get matched up against a chill group of guys and have fun.
 

urk

butthole fishhooking yes
Players start low and TrueSkill begins to assess them accordingly after the first recorded game. You might imagine it's zeroing in by going low, then high, until it can converge on the correct skill level. Ping pongin'.

Tashi, pretty much. Sometimes you get a player mid-assessment and they're suspected to be higher than they really are. Sometimes it's the other way around. Try not to think of that poor dude as dead weight. For a fleeting moment, you were his mentor, and now he will forever aspire to make you proud.
 
urk said:
Players start low and TrueSkill begins to assess them accordingly after the first recorded game. You might imagine it's zeroing in by going low, then high, until it can converge on the correct skill level. Ping pongin'.

Tashi, pretty much. Sometimes you get a player mid-assessment and they're suspected to be higher than they really are. Sometimes it's the other way around. Try not to think of that poor dude as dead weight. For a fleeting moment, you were his mentor, and now he will forever aspire to make you proud.

Okay makes sense, thanks for clearing it up.



Hockey would be very cool if you couldn't assassinate or even damage people.
 

Tashi

343i Lead Esports Producer
urk said:
Players start low and TrueSkill begins to assess them accordingly after the first recorded game. You might imagine it's zeroing in by going low, then high, until it can converge on the correct skill level. Ping pongin'.

Tashi, pretty much. Sometimes you get a player mid-assessment and they're suspected to be higher than they really are. Sometimes it's the other way around. Try not to think of that poor dude as dead weight. For a fleeting moment, you were his mentor, and now he will forever aspire to make you proud.

:') Make me proud son
 

Karl2177

Member
Two games today, I almost had Perfections.

Game 1: Karl2177 is on a Running Riot.
*enemies ignore my teammates and focus on me*

Game 2: Karl2177 is on a Killing Frenzy.
*teammates enter betrayal mode*

People's mentalities while playing this game somewhat baffle me.
 
Karl2177 said:
Two games today, I almost had Perfections.

Game 1: Karl2177 is on a Running Riot.
*enemies ignore my teammates and focus on me*

Game 2: Karl2177 is on a Killing Frenzy.
*teammates enter betrayal mode*

People's mentalities while playing this game somewhat baffle me.
Glad the enemy team/teammates didn't do that when I was on an Untouchable. lol.
 

Brolic Gaoler

formerly Alienshogun
urk said:
Players start low and TrueSkill begins to assess them accordingly after the first recorded game. You might imagine it's zeroing in by going low, then high, until it can converge on the correct skill level. Ping pongin'.

Tashi, pretty much. Sometimes you get a player mid-assessment and they're suspected to be higher than they really are. Sometimes it's the other way around. Try not to think of that poor dude as dead weight. For a fleeting moment, you were his mentor, and now he will forever aspire to make you proud.


It's more likely he will go back to betraying Gui for sniper and bitching at his mother to bring him chocolate milk.
 
Top Bottom