Destiny 2 Tracker Network
Destiny 2 will receive an update tomorrow. Players will be required to log in to Destiny 2 again after installing the update. Please stay tuned to @BungieHelp for updates.
Love the site? Show us love
Learn More Donate
Yin
Yin

311 Credits 336 Posts
1 Credit(s) Received
Piazza03 said:When I check my stats for each crucible game type, it appears that none of the games I've played recently are registering for true skill.

I know I've played control, skirmish, and trials games within the last few days, but Skirmish isn't showing an update in games for 10 days, 15 for control.

Is this a bug, or am I doing something wrong? I always do the refresh and/or update stats, but that doesn't seem to help.


I am seeing recent games in Control and Skirmish from four days ago on your playlists page. Your TrueSkill may not be increasing because your in game performance hasn't surpassed its requirements. The blue bar under 'TrueSkill Progress' shows you how close you are to reaching the next rank. It does appear to working properly on your account, your Trials TrueSkill increased five days ago, and you Elimination increased a day ago. Just in case though, monitor that blue bar for a while and look for changes while you play.
Piazza03
Piazza03

0 Credits 5 Posts
Thanks for the response. Shouldn't those recent games change my True Skill history to correspond to those days of games? Understanding it may not change my actual True Skill score, it still should at least update my history to show those games are being used in the calculation.
Yin
Yin

311 Credits 336 Posts
Piazza03 said:Thanks for the response. Shouldn't those recent games change my True Skill history to correspond to those days of games? Understanding it may not change my actual True Skill score, it still should at least update my history to show those games are being used in the calculation.


Your TrueSkill history won't change until you either increase or decrease in level. This is just a design decision. If the games you play for over a single day balance out and result in no change in TrueSkill, it won't be reflected in the history. You can tell if your games are counting when they appear in the "Recent Games" section of the TrueSkill page, the contents of those games will have counted.

The idea of updating TrueSkill only on increases/decreases is merely a design decision. The history operates based on collection, so if you level twice in a single day they will appear twice (see image below). We could record your increases or decreases in TrueSkill based on your collections, even if you stay the same level; but we wouldn't be able to log your journey from beginning to end and you would have a new row every time you collected your stats (some people do this every three-four games, that's a lot of lines to add to the history).



Piazza03
Piazza03

0 Credits 5 Posts
Ah, I got it now. Thanks for the breakdown.
Zenki
Zenki

0 Credits 1 Posts
Why does it feel like such a grind to go from 49 to 50. I've played over 30+ games undefeated in trials and only getting 2% per win. but when i finally lost i dropped 10%. i feel everyone should be allowed a bad game. or not take such a burden when you lose. just don't under stand going a constant 2.0+ and the losing game one goes .80 and took a chunk out of percentage.
Yin
Yin

311 Credits 336 Posts
Zenki said:Why does it feel like such a grind to go from 49 to 50. I've played over 30+ games undefeated in trials and only getting 2% per win. but when i finally lost i dropped 10%. i feel everyone should be allowed a bad game. or not take such a burden when you lose. just don't under stand going a constant 2.0+ and the losing game one goes .80 and took a chunk out of percentage.


The distance between 49 and 50 is almost the same as the distance between 48 to 49 and even 47 to 48. You're also getting more than 2% per win on average.

I understand that you are anxious to get your 50, but stay strong. Setbacks are part of the process, and 10% is something that you can make up relatively quickly from the looks of your recent games.
Micky O Neil
Micky O Neil

1 Credits 13 Posts
It must go mostly by W/L ratio and if your in a group. I have a high K/D but a low W/L and I'm always solo in the Crucible.
Yin
Yin

311 Credits 336 Posts
Micky O Neil said:It must go mostly by W/L ratio and if your in a group. I have a high K/D but a low W/L and I'm always solo in the Crucible.


A big portion does come from winning, but your actual ratio doesn't play a roll in it. You could have a 20% win rate on average and it has no bearing on the speed at which you accrue points from games.
alonei r leet
alonei r leet

0 Credits 2 Posts
It really only goes up when you get wins, which is sad because if you play by yourself most of the time, you get the best around but you can't carry the kid that goes 1/14 (just happened to me). Last four games 2/4, Best around in both losses, but my skill level goes down... Doesn't make much sense considering my skill was > then those in the lobby.
Yin
Yin

311 Credits 336 Posts
alonei r leet said:It really only goes up when you get wins, which is sad because if you play by yourself most of the time, you get the best around but you can't carry the kid that goes 1/14 (just happened to me). Last four games 2/4, Best around in both losses, but my skill level goes down... Doesn't make much sense considering my skill was > then those in the lobby.


Out of curiosity, I found and punched the numbers behind your TrueSkill for the game you mentioned, and your own personal gameplay padded a majority of the points you lost.

I'm beginning to think that displaying the percentile progress for TrueSkill is a mistake. Other players, such as yourself, express disappointment in the system using winning and losing in the way that it does as a solo searching player. If you've played ranked modes in Halo 3 or League Play in Black Ops 2, these systems are also highly based on winning or losing the game, with added input coming from the individual stats of the player as well. These systems are less transparent than ours, but individuals continued to seek higher placements, despite the system being driven and focused around team-based play. Our system is similar, only that we cannot influence who you actually match against. The problem here, is that players are putting a magnifying glass to the progress percentage and are becoming frustrated at the first sight of lost progress (a setback that most, if not all, ranking structures include in their experiences).

You've lost progress, sure. But overall your progress has been steadily increasing, despite these strings of losses. Having a few unfortunate losses back to back is not going to revert you to depressingly low TrueSkills; given enough time, the system is forgiving to players who clearly excel beyond the expectations set for the TrueSkill that they are currently rated at.
alonei r leet
alonei r leet

0 Credits 2 Posts
Darkened Night said:
alonei r leet said:It really only goes up when you get wins, which is sad because if you play by yourself most of the time, you get the best around but you can't carry the kid that goes 1/14 (just happened to me). Last four games 2/4, Best around in both losses, but my skill level goes down... Doesn't make much sense considering my skill was > then those in the lobby.


...

Out of curiosity, I found and punched the numbers behind your TrueSkill for the game you mentioned, and your own personal gameplay padded a majority of the points you lost....

^^ maybe I'm taking it the wrong way, but that did seem like you were saying my gameplay was bad, and none of those games were bad games by any means. Maybe it was just bad in the terms of your formula, I'd like to think the latter. It is interesting, my KD was higher than my average in every game and scored the best around, what else could have I done?

On that note, I would take the opposite approach to your thinking, not that you're wrong because I DO see your point of view on the nick-picking of true skill because of the progress bar. That being said, questions won't be posed to you if the formula was known as to why people went up or down, that would solve the issue of people saying "ah man I don't understand" - it would be clear.

Just playing devils advocate here, you can understand the frustration of destroying everyone in the game, and losing true skill though right? "My play was > than my own statistics, and beat everyone else in the lobby by a good margin - but my skill went down because of bad player X's score I can't control"

Maybe it's just a obstacle I don't understand given I don't see the stats you receive from the game and the best way to create a TS score. That being said, thank you for doing it because I'm here, trying to raise mine.



Yin
Yin

311 Credits 336 Posts
alonei r leet said:^^ maybe I'm taking it the wrong way, but that did seem like you were saying my gameplay was bad, and none of those games were bad games by any means. Maybe it was just bad in the terms of your formula, I'd like to think the latter. It is interesting, my KD was higher than my average in every game and scored the best around, what else could have I done?


What I was implying was that your own gameplay was strong enough to negate a large percentage of the points that were lost from losing the game. By padding, I was referring to a "cushioning" of a majority of the blow that the loss would have held against your TrueSkill. My implied meaning was the exact opposite from what you interpreted, my apologies for not making this more clear.

I can most definitely sympathize with those who consistently carry members of a team, only to lose the game. My recent run in Iron Banner saw 12 losses in a row that were largely out of my control (little bit of a pun, there), an event frustrating beyond measure even without considering TrueSkill at all.

The reason we have wins set the way we do is a direct result of the shifting gameplay that providing stats causes. Competitive Guardians (unfortunately) compete with K/D in mind, it's the misleading go-to statistic that is used to gauge someone's skill. This led to an influx in camping, where players would search solo and camp in the back of the level to avoid dying altogether (in Halo 4, we saw players with 120.0 K/Ds and higher). While their performance was strong from a statistical perspective, these players would be spending the entire match out of consideration most of the time, leading games like Control to play out much like a 5v6.

With this in mind, we needed to enforce strong win requirements; but at the same time, we needed to do so in a way that didn't ultimately punish the players who consistently carried the team and still ended up with a loss. On average, a player who consistently places at the top of the scoreboard will see wins more often, as they are the corner-stone that leads to those victories. In theory, a consistently strong player will undoubtedly be capable of reaching the max rank, given enough time. The amount of time that is required for this theory largely depends on the number of random variables (namely random players). The more of these random variables you eliminate, the easier it is to predict your outcome. I would argue that you would have a much easier time earning TrueSkill in Skirmish than you would Control, because you are dealing with half the members on your team. Your contribution, therefore, is much more significant and more deterministic of whether you are going to win or lose a game. With that in mind, we prioritize our wins to force players to compete for the win. Players like yourself who exhibit strong gameplay are doing so as a result of playing for the win, and not relying on camping and losing matches as a result of more passive gameplay.

From there, you can shift your direction to the concept of solo versus team searching. Guardians that play at the competitive level almost always acquire a team to do so. Hell, even Trials of Osiris requires a team in order to prevent teams from stomping over players who are searching by themselves. Because of this, we needed to cater the system to be more demanding to keep it challenging for Guardians who search as a team.

TL;DR We base our TrueSkill primarily off of wins to prevent camping players from reaching the highest TrueSkill. The difficulty curve for TrueSkill is used to cater to the majority of players at the top of the leaderboard, who play as a team. This, unfortunately, makes it more difficult for solo searching players, but this idea is no different from ranked play in games such as Halo 3 or Black Ops 2.
Is Endless
Is Endless

0 Credits 1 Posts
The system is somewhat flawed, sure. However, how could it be perfect. I think it works well for the most part. Anybody who takes this too seriously is only creating problems for themselves. It's just a guide of sorts. For the most part, it's accurate. It's not perfect.
jrgeek
jrgeek

0 Credits 11 Posts
I'm way late to the party, but this tool is pretty darn cool. While I'll never - probably never - get past a 28 - which is my goal - my eyes aren't what they used to be and I don't play in front of a PC monitor. All that being said, I love the statistical analytic's y'all do here. As for you folks that are battling the arc of being on losing teams, I'll apologize for the moderate players like me where I'll have a few crappy games and then I'll have a streak where they are really good ones - at my level at least -.
blue23454
blue23454

0 Credits 5 Posts
Darkened Night said:
Danymac89 said:Stupid system seeing as I keep getting put into losing matches of 18000-7000 point control matches


-snip-
If you are intent on insulting the features that we provide, we ask that you provide constructive feedback for the sake of improving the features for everyone to use or simply dismiss them entirely.
-snip-



Yes because there's so much constructive feedback that can be given on a "super secret formula"
Maybe some transparency would help here, you guys run a stats website, which makes it a science. There can be no peer review when you treat your methods like it's a Krabby Patty.
Sorry but exact formulations, computations, and data collection methods are the only way people like this guy could even begin to offer constructive feedback.
Until then we will all just have to trust your apparently vast higher knowledge and hope that somehow the method actually represents skill or move on to another site.

Darkened Night said:Those losing matches don't seem to be affecting you. You've been consistently going up in Control and you have a solid 33 in Trials of Osiris. The system is working as intended, and hundreds of our members have 50s. Remember that even the best player won't earn a 50 overnight, it takes time.


Translation:
Our method for determining skill works because, according to our methods, some people are good at this game. You're probably just mad that your score isn't going up fast enough and are in no way criticizing the theory so get good.

Look, I know that, for me personally, the system seems to work, it usually reflects my skill at the time that I checked the site. I also know that wins weigh into the algorithm heavily, and I can tell you, as someone who frequently plays on his own, that my wins are not representative of my skill. My wins represent little more than what kind of teammates I was playing with, but I can't offer constructive feedback unless I know exactly how much wins weigh, and how much any other factor weighs. I know I'm commenting on two old posts but the sheer audacity you displayed in that one response was staggering.

Speaking for myself, and probably a little for Danymac
I'm definitely not saying DTR should be a flat average, like K/D or S/G, but many people would probably agree that wins don't represent skill, and the fact that they're a factor at all, let alone that it's apparently heavily weighted, makes us question if it should even be referenced as a skill rating.

If you're a College Football fan, here's an example from last Saturday, MSU @ UofM. Blake O'Neill is the reason MSU won, the game should have been over on this last play, but O'Neill seriously screwed up; his team lost because of him. Does that mean De'Veon Smith, also on the team, is a bad player? No, and if it wouldn't make sense to have one person's skill affect another person's stats in Football, a game where every player on the team knows each other personally, actively communicates, and practices together on a daily basis, then how does it make sense for your entire team's skill to affect one person's stats in Destiny, a game where you play with random people you don't know, never played with before, and don't speak to at all?

TL;DR
Don't pretend like a valid complaint wasn't made when you can't even be bothered to share your methods so that constructive feedback is at least possible.
Yin
Yin

311 Credits 336 Posts
blue23454 said:Yes because there's so much constructive feedback that can be given on a "super secret formula"
Maybe some transparency would help here, you guys run a stats website, which makes it a science. There can be no peer review when you treat your methods like it's a Krabby Patty.
Sorry but exact formulations, computations, and data collection methods are the only way people like this guy could even begin to offer constructive feedback.
Until then we will all just have to trust your apparently vast higher knowledge and hope that somehow the method actually represents skill or move on to another site.


While being fully transparent may be the only way to assure our members that it works appropriately, we have good reasons attempting to balance the elements that we reveal and disclose to our users. Were we to do so (particularly in combination with adjusting the formula to not account for wins, as you later state), there is room for exploitation. We have experienced this in the past over at our sister site, HaloTracker, and try to keep our formulas under wraps to prevent it from happening. It's not a claim to have "higher knowledge" and an attempt to solicit trust, but we do consider its inner workings as a private matter.


blue23454 said:Translation:
Our method for determining skill works because, according to our methods, some people are good at this game. You're probably just mad that your score isn't going up fast enough and are in no way criticizing the theory so get good.


You have misinterpreted my words; I reviewed the game history of Danymac89 and noted that he or she was consistently increasing in TrueSkill regardless of the unfortunate Join in Progress matches that occurred. It wasn't a joust insisting that their performance was inadequate, it was a prompt to remind them that their abilities could very well carry them to a 50 if given enough time.


blue23454 said:I also know that wins weigh into the algorithm heavily, and I can tell you, as someone who frequently plays on his own, that my wins are not representative of my skill. My wins represent little more than what kind of teammates I was playing with, but I can't offer constructive feedback unless I know exactly how much wins weigh, and how much any other factor weighs.


This dwells on the previous notion and decision we had to make when developing TrueSkill whether to cater to members who play in teams or those who do not. By creating a system solely (or primarily) around individual participation, you open up a world of issues related to stat padding/camping, as well as idea that players who compete on a team will often reach the highest rank because the skill ceiling caters to that of non-team searching players. It's definitely a drawback to this sort of system, where individual players are hit more by penalties; but upon weighing the options, is ultimately what we decided on. One could very easily argue that this is not the case, and that creating an individual formula that doesn't include wins could be equally valuable, I'm certainly not denying that. However, with the system we have developed, we use wins because each member has a "pull factor" that will inevitably lead to a higher percentage of wins (theoretically, assuming everyone has a similar distribution of skills for teammates).


blue23454 said:Don't pretend like a valid complaint wasn't made when you can't even be bothered to share your methods so that constructive feedback is at least possible.


It wasn't that his or her response wasn't a valid complaint; it is that after reviewing their game history, their TrueSkill remained unaffected from those join in progress matches. That said, despite that other users are similarly met with the unfortunate events like Join In Progress matches, those capable often always end up pulling through it. And, this isn't a result of the wins being too big a factor, their individual performance is preventing this from becoming an issue. I regret that I cannot justify this point with a transparent breakdown with our formula to reinforce it.


blue23454
blue23454

0 Credits 5 Posts
I believe I understand your points, however:

Darkened Night said:

It wasn't that his or her response wasn't a valid complaint; it is that after reviewing their game history, their TrueSkill remained unaffected from those join in progress matches. That said, despite that other users are similarly met with the unfortunate events like Join In Progress matches, those capable often always end up pulling through it. And, this isn't a result of the wins being too big a factor, their individual performance is preventing this from becoming an issue. I regret that I cannot justify this point with a transparent breakdown with our formula to reinforce it.




In this case, you were saying the formula must be working for him, because despite losses he still increased in score. I even believe I stated I typically agree with my score, so for me, in practice, my score works. Just because a theory works in practice doesn't mean the theory is correct though.
I could develop an entire theory about how umbrellas keep me dry in the rain because they are magnetically charged and repel water, sure in practice it works because the umbrella keeps me dry, but if I walk outside with an electromagnet taped to my head I'm going to get wet... Point being, you can have 100 instances of your formula working, that doesn't mean it always works.

To your point of padding, that would be difficult to do on a game where no private servers or lobbies exist. I suppose you could pad your stats to a point on Rumble with two friends, but it's really not that possible beyond getting good, deleting your character, then starting over (which is really more of a gamble than padding, as this could easily go the other way for you). Also, campers are going to camp, period. My opinion, if revealing the formula for skill rating would encourage camping, then the formula is fundamentally flawed.

The only part of your argument I can agree with is
said:players who compete on a team will often reach the highest rank because the skill ceiling caters to that of non-team searching players.

But I know the game also matches you against similar fireteams. That is, if I'm in a fireteam of 6 and go into control, the game will try and find another 6man fireteam for me and my friends to play against. Everyone is playing at the same level of communication, and yes if my 6man fireteam goes up against a team of randoms we will probably dominate and our skill will go up faster, that's going to happen no matter what, but I don't think catering to individual scores will necessarily skew the skill ratings so that all or most players who always play in fireteams with friends will cap at 50 in a short amount of time. In a game where matchmaking finds a similarly matched team to put you against, I would think a skill rating focused on individual performance would balance out.

In other words what I'm hearing is "Instead of making it easy for fireteam members to increase their individual skill, we made it hard for solo players to increase skill" which is, not only the same thing but, makes the assumption that full fireteams are going to dominate every match when the reality is full fireteams are going to be playing against full fireteams more often than not.
Point being, since anything regarding the DTR skill points out that wins are important, in a fireteam your DTR represents your overall skill in that team, and that's kind of cool, but as a solo player your DTR skill represents your luck in randomly being selected into the better team.

So here's an idea, give each player two skill ratings
Solo and Fireteam skill
Solo skill only collects data when playing alone, geared more towards individual performance factors (S/G, K/D, etc.)
Fireteam skill only collects data when playing with at least one other person in your fireteam, geared more towards team actions (KA/D, Wins, etc.)
If there's a way to separate that data, I'd say this would be a good alternative given your concerns. There has to be a way to make that distinction in the data, the game does it (roster separates fireteam from lobby, teammates are blue while fireteam members are green).
Yin
Yin

311 Credits 336 Posts
blue23454 said:Just because a theory works in practice doesn't mean the theory is correct though.


Absolutely. This is one of the main reasons we offer the details we do, and the manner in which I reply. When potentially flawed circumstances do appear and I receive a message from a user, the data gets reviewed and are looked at more in depth (from other players) to determine if there is an issue or if a rebalance is necessary. This has occurred a number of times since we first debuted TrueSkill, and we are proud of creating a feature that can be refined in such a way.


blue23454 said:To your point of padding, that would be difficult to do on a game where no private servers or lobbies exist. I suppose you could pad your stats to a point on Rumble with two friends, but it's really not that possible beyond getting good, deleting your character, then starting over (which is really more of a gamble than padding, as this could easily go the other way for you). Also, campers are going to camp, period. My opinion, if revealing the formula for skill rating would encourage camping, then the formula is fundamentally flawed.


Consider the idea of having a formula where winning games has no bearing on your skill, and factor in the elements that you feel provide an adequate representation; lets consider K/D for an example, an element that might not be accurate of skill alone, it does provide a means of determining how well someone does in contrast to the opposition. In matches of Control, players could simply avoid going for capture points, purposely picking points by which they can continue to kill players to increase their K/D. Completing ignoring the objective causes the team to lose the match, but the player picked up some awesome stats along the way (mostly because the enemy team was focusing on playing the objective, not solely going for kills). Without factoring wins into the equation, this player can rapidly climb up to a high rank. This is the foundation for why we consider wins as an important element.

Similarly, Rumble can be "padded" by simply not actively seeking out players. A single player could stay in a single location, only attacking players who are in his/her immediate area. By not going for the win, this player manages to pick up high marks, and maybe even a few multikills when their super charges.

With the elements we have available to us as a community site, it's difficult to handle elements that display individual performance without giving way to a method of boosting ones score through otherwise unsportsmanlike play. This is why wins is an important factor, to encourage players to actively compete for the win. To not do so, would leave the formula with a huge opportunity to be exploited.


blue23454 said:But I know the game also matches you against similar fireteams. That is, if I'm in a fireteam of 6 and go into control, the game will try and find another 6man fireteam for me and my friends to play against. Everyone is playing at the same level of communication, and yes if my 6man fireteam goes up against a team of randoms we will probably dominate and our skill will go up faster, that's going to happen no matter what, but I don't think catering to individual scores will necessarily skew the skill ratings so that all or most players who always play in fireteams with friends will cap at 50 in a short amount of time. In a game where matchmaking finds a similarly matched team to put you against, I would think a skill rating focused on individual performance would balance out.


I have to disagree with you on this one. Destiny offers very little in terms of skill matching; most of my games, even in last week's Iron Banner, often pitted me with a group of solo searching players against teams of six. This wasn't a by-case basis; I was consistently matched against teams of four or higher while searching solo. In my Skirmish matches, a mode where I often run with a group of three, I'm pitted against groups of two with a random. These odds are often hardly fair, and through my experiences (nearing up on 2,000 games) I am fairly confident that the game does a poor job of matching teams against teams and skill against skill. This doesn't serve to aid or hinder my opinion related to TrueSkill, but it is a major discrepancy from what you have said.


blue23454 said:In other words what I'm hearing is "Instead of making it easy for fireteam members to increase their individual skill, we made it hard for solo players to increase skill" which is, not only the same thing but, makes the assumption that full fireteams are going to dominate every match when the reality is full fireteams are going to be playing against full fireteams more often than not.Point being, since anything regarding the DTR skill points out that wins are important, in a fireteam your DTR represents your overall skill in that team, and that's kind of cool, but as a solo player your DTR skill represents your luck in randomly being selected into the better team.


It's not a point of making it easier for fireteam members to increase their skill, it's more of trying to estimate the moment of a player who competes in a team. This might be appear as an assumption, but majority of players at the top of the leaderboards are there as a result of competing as a member of a fireteam. We understand it may be more difficult to reach higher ranks as an individual searching player, but the difference in data between the two is too massive; as a result we try to balance team wins with individual play. A player being carried through Skirmish will not be able to reach 50, so wins aren't the sole determining factor of this formula.


blue23454 said:So here's an idea, give each player two skill ratings
Solo and Fireteam skill
Solo skill only collects data when playing alone, geared more towards individual performance factors (S/G, K/D, etc.)
Fireteam skill only collects data when playing with at least one other person in your fireteam, geared more towards team actions (KA/D, Wins, etc.)
If there's a way to separate that data, I'd say this would be a good alternative given your concerns. There has to be a way to make that distinction in the data, the game does it (roster separates fireteam from lobby, teammates are blue while fireteam members are green).


Sadly, there is no way to determine this; believe me, if this were an option, you can be sure that there would be two values for something like that. We work through Bungie's provided API, so not only are we limited to who is playing in matches, but what information we can retrieve from those matches. Personally, I would love to have an ability to determine whether a match was joined while it was in progress; the ability to do so would completely fix the single biggest issue facing solo searching players (the frequency of join in progress matches). This is the issue with creating a system such as TrueSkill as a community site. If we were Bungie, we would have a lot more tools at our disposal and a lot more resources on-hand to compute a more advanced formula. For us, being able to create a system that (we feel) balances individual gameplay elements with wins is something we are particularly proud of.


NuclearSynapse
NuclearSynapse

0 Credits 1 Posts
hatebreedsrage said:
Danymac89 said:Stupid system seeing as I keep getting put into losing matches of 18000-7000 point control matches


Yeah, its essentially tailoring your score to whatever random group of people you've been paired with. I've seen some truly amazing people with crazy high scores on the losing team because they got paired with players who had 4 kills and 17 deaths. I dont see how they can call this a TrueSkill score when it has nothing to do with your skill alone.


The ratings do factor in individual performance, so those who do extremely well while their team loses won't be destroyed in their TrueSkill score like it may appear.
Vinni-K
Vinni-K

0 Credits 13 Posts
Hey I was wondering how long it will take to get to 50 Trueskill from 49 Trueskill? I got from 48-49 in 2 days but it seems like getting to 50 is going to take forever...