: > [{quoted}](name=Exarch Leona,realm=OCE,application-id=FjGAIbRv,discussion-id=PvHWTvdT,comment-id=0001,timestamp=2019-11-05T02:24:55.958+0000) > > They're worse in Gold, like holy crap man. > A dead fish on land is more useful than the people I get on my team. > "OMG, DEY GOT FB, AFKING AND THEN FF AT 15 REEEEEEEE" > > And the best part is, Riot says that crap is okay because "nut against dah rulz" Come on, this is getting old. You think of yourself better than everyone else. All the while, ignored all your flaws and conditioning. I would bet my bottom dollar you wont try this in your actual work place. Saying your Co workers is shit, you can do a better job, and the reason you cant get a promotion because the team you are placed in is holding you back. But then again, i could be giving you a view you have not yet reach. Prove me wrong eh?
> [{quoted}](name=TheDemonEmperor,realm=OCE,application-id=FjGAIbRv,discussion-id=PvHWTvdT,comment-id=00010000,timestamp=2019-11-05T02:35:18.235+0000) > > Come on, this is getting old. You think of yourself better than everyone else. All the while, ignored all your flaws and conditioning. > > I would bet my bottom dollar you wont try this in your actual work place. Saying your Co workers is shit, you can do a better job, and the reason you cant get a promotion because the team you are placed in is holding you back. > > But then again, i could be giving you a view you have not yet reach. Prove me wrong eh? Well Season 9 was basically about creating new tiers and making people think they improved a whole tier higher when they actually didn't. So it's not surprising that some players are vocal about it.
: new champion copy paste has been released
So if what you are saying is that: * Her gun looks like a large Caitlyn skinned gun * Her Q is like a Lucian Q, just with healing for allies * Her W is like a Zyra root, just without the initial plant rolling * Her E is like the inside of Akali's shroud * Her R is like a global Sona/Jhin/Jinx ult, only wider I would tend to agree. But a lot of the new champions are like this. The ones that are not that come to mind are Kled and Tahm Kench.
: > Picture 1 Because there is a lot more that goes into MMR calculations than binary win/loss. The average MMR of an unranked account is mid-low Silver, and it is set there for a reason. Namely, that is the average skill (as far as I know a little lower under the same principle as the soft reset they do every year) of the player-base, and thus the most likely skill bracket they will fall into. It is essentially another compromise. Where data is lacking you use *all* the data you have to make the best-educated guess you can and go from there. It isn't perfect, and this means that mid-low Silver will always have a few outliers, but new players have to start *somewhere*, and this method results in the most accurate results possible. The point of including some blue data was to point out there really wasn't a significant imbalance, because while red looked overall 'worse', when you really dig into it they were basically even. The point of mentioning that Trist is now Gold was to point out that a winrate over 10 games is beyond meaningless and using it as the sole indicator is misleading *at best*. The Leona has still played <30 so not much can be said. > Picture 2 I mentioned the Lee Sin simply because they didn't have a visible rank and it was to put the whole into perspective. The Vel had and still has played barely any games. Not anywhere near enough for their visible rank to catch up to their hidden MMR. I mentioned their performance because their visible rank is Iron 1, but they are in a mid-Silver MMR range game and performed well, which would imply that their MMR is mid-Silver, and also likely climbing from there. This comes down to MMR and visible rank not matching while people are climbing. MMR catches up to true skill much faster than visible rank does because it isn't impeded by divisions, placements, dodges, remakes, etc. Take my own climb as an example, in previous Gold-ranked seasons I would persistently be matched against Plat+, i.e I had a Plat~ range MMR, but I also never played enough games for my rank to catch up to my MMR. Now I have played those games and my visible rank has caught up to my MMR, and I've hit the plateau where I am no longer climbing ranks simply by playing. > Picture 3 Because Flex is a substantially different environment. Teamwork is a key skill when it comes to League, and the way that skill is executed differs wildly between Flex and Solo/Duo. In the same way that TT utilises all the same skills, but in different ways and to different depths as SR does, Flex vs Solo/Duo is much the same. People also treat Flex differently. For a player like myself, the ranks should be fairly similar because roles, champions and pre-mades are pretty consistent between the two, but for some players the difference is enormous. For example, a solo Mid main in S/D may play ADC with their Support mate in Flex, a role they don't normally play with a different lane dynamic because of the extra person, so relative skill and rank will differ greatly. Some people also don't take Flex as seriously and will play off-roles or champions, which puts their effective Flex rank lower because they have less skill with that given role/champ/playstyle. When dynamic queue was split back into Solo/Duo and Flex both queues used cues from the same MMR rankings. The reason they don't continue to do this for new Flex players is for the same reason they don't use normal MMR to indicate for new S/D players. The modes are different enough to cause more inaccuracy using that method than simply using the existing ranked player base to determine their MMR. > Picture 4 The accuracy of ranks is irrelevant to matchmaking health. Rank is purely an individual indicator of progression. So long as the hidden MMR is placing them in the appropriate games, which given they were in a premade and their solo rank reflects a similar skill level seems like it was, then matchmaking is essentially healthy because they are versing and playing with people of similar skill. Premades will always account for a degree of inaccuracy here. > Picture 5 21% over 38 games tells you barely anything. When were the losses? (Were they early in placements before their MMR could catch up?) How did they perform in those losses? (Losing a game doesn't mean you personally lost the game for the team) Have they been getting filled/second pref roles? (Maybe they play a role that wasn't popular last season but is this one and so are being forced into their worse roles a lot) Do they play at weird times and so they are versing wider ranges of MMR resulting in odd visible results but perfectly accurate hidden metrics? 38 games just isn't enough data to answer any of those questions which is precisely my point. Win-rate can only be used as any sort of accurate indicator past 100 games. Using win-rate or visible rank before that is only going to cause you grief by misleading yourself. > Picture 6 That 3-4/min number hasn't really changed all that much over the seasons in my experience, but also consider that the pace of the game has changed. First 15 minutes used to a lot more just farming up, where now there's a lot more fighting and trading. Naturally that impacts CS. The game has trended to a more aggressive and less passive pattern over time. The introduction of Iron only exaggerates the perceived lowering of skill because it has pushed the ranks up. People are still in the same *relative* skill bracket that they were, but the true visible rank of that skill bracket has changed for many players. Gold remained about the same, but that means that anyone in Silver and Bronze essentially has their true rank dropped a div or two to compensate for the extra ranks. So Silver 4 of this season was the equivalent of Silver 2-3 last year at a conservative estimate. A full reset will result in exactly the phenomenon that you were criticising in Flex rank. You can't have both a full reset and take cues from past or other rankings. They directly contradict each other. > Picture 7 Trying to base anything off of 3 minutes of a game and flat data is not a good way to come to a solid conclusion. You seem to be of the mindset that win-rate (regardless of games played) is the only factor in determining the outcome of a game. I highly doubt that is the case but it is the impression you give. We don't now know the exact circumstances of the Ryze nor the Yasuo. The system *has* all that background and in-depth data. > Picture 8 Not necessarily, because again, which games is he winning/losing? Is he winning until the system puts him into Gold MMR games and then losing until he drops back into low Silver? Would that not indicate to the system that he belongs somewhere above where he is but below Gold? Well ok let's put him in mid-silver games and ramp it up slower this time until he starts losing again and then we'll even it out. His match history would indicate that he *was* getting ramped real fast and that he was hitting the slow down point. Given that he's now Silver 3 with a substantially dropped 65% win-rate (and not a huge amount more games) it would seem to indicate that his true rank lies in the high Silver-low-Gold range, which is about right for the games that he was in with you (a mid Silver game). You should be interested in what happened in that game because it's first-hand experience and can help guide the conclusions you draw from other insufficient data. The fact that the Jhin (who had an *apparently* low MMR) outperformed the Twitch (who had an *apparently* higher MMR) means that more than likely there is far more behind the scenes that is indicated in their win-rates and visible ranks. Someone who's true rank is lower than Iron 1 (if we base it solely off his negative WR) is only going to outperform someone who is supposedly high Silver-low Gold (again basing it solely off of visible WR) under exceptional circumstances, and yet both their performances were consistent across the match and ross subsequent matches. Given that, clearly there is more to the story than what is shown in visible rank and WR, which is basically my entire running point. > Picture 9 Perhaps they won the first 10 and got catapulted into low Gold MMR and the system is now dragging them back down to where they belong (and consequently dragging their early WR down too). There is simply not enough visible data to make those kind of conclusions. > Picture 10 Because MMR and visible rank does not match until a player plateaus (i.e reaches their true skill and MMR, ceases climbing and is just breaking even). *Subsequent* results give you a strong idea of where their MMR was trending before their visible rank had caught up. Given the Vlad's subsequent climb the system clearly knew he was meant to be higher than his visible rank indicated and got him there, giving you a good idea of why a visibly Bronze 4 player was in a mid SIlver game. *Because his MMR at the time was mid Silver and trending upwards*. > The Last Picture. This comes down to population. Queue times in this MMR range are already excruciatingly long, and there just aren't enough people to get matchmaking of the quality that most people would like. The system accounts for this by adjusting MMR and LP gains and losses based on whether you are "expected" by the metrics of the system to win or lose the match (Expected to lose and lose? Lose less LP/MMR. Expected to win but lose? Lose more MMR/LP. Expected to lose but win? Big gains.). The alternative is that these players either have limited queue times, which has its own caveats of some players simply not being available at any of those times, or never getting a game at all. The current system is the compromise of wider matchmaking ranges to allow for games to actually exist, with adjusted gains to compensate for the imbalances that will inevitably occur. There is no perfect solution, the scale simply makes that impossible.
Hi Seras, I only started taking screenshots of these example, because I was sick of playing games where the one player in the game would cause everyone else to have less enjoyment. There were clearly players out of the skill depth for those games. And as a support main, I can fully understand where people like Exarch Leona are coming from. Because the support role is least popular, a lot of times the current system is placing the only supports it sees in the pool in with the monkeys. And that is frustrating to know that whatever your level is, a lot of time you will be lumped with monkeys. My idea to revert champion select back to pick order where the player chooses their champion AND their role, I think is a good one. It will make much faster matchmaking (just like back in the day), hence giving the algorithim more time to choose the correct mmrs from the pool and not having to find ROLES from the pool. And the bonus is the playerbase skill level will rise, having to sometimes play roles that you currently don't mean that you learn each role. If you consider the autofill function at the moment, players hate that with a passion. But its really no different to what I'm proposing. Sometimes you will be last pick, and sometimes you won't. I think Riot's idea to have the current role selection was short term thinking that they could curb toxicity. As for the matchmaking discussion, I don't have too much further to add, but I will address some points below. My position is still the same, Riot could do so much more to help the players and themselves. > [{quoted}](name=Seras Dragon,realm=OCE,application-id=FjGAIbRv,discussion-id=7qIE3EQp,comment-id=00050000,timestamp=2019-11-13T12:21:01.824+0000) > > Because there is a lot more that goes into MMR calculations than binary win/loss. The average MMR of an unranked account is mid-low Silver, and it is set there for a reason. Namely, that is the average skill (as far as I know a little lower under the same principle as the soft reset they do every year) of the player-base, and thus the most likely skill bracket they will fall into. It is essentially another compromise. I don't see many people complain abot DOTA2's matchmaking system, or chess's. I would say that the current system needs to be improved, and more streamlined (less complicated). Because the more games you play like these, it is very obvious who the weak link is in the game, and why they simply don't have the same skill level. > Take my own climb as an example, in previous Gold-ranked seasons I would persistently be matched against Plat+, i.e I had a >Plat~ range MMR, but I also never played enough games for my rank to catch up to my MMR. Now I have played those games >and my visible rank has caught up to my MMR, and I've hit the plateau where I am no longer climbing ranks simply by >playing. Understood, so I think the MMR system should be far quicker to change. Like NEACE says, it's simply outdated. >Because Flex is a substantially different environment. Teamwork is a key skill when it comes to League, and the way that skill i>s executed differs wildly between Flex and Solo/Duo. ? >People also treat Flex differently. That's true, but I'm only interested in the matchmaking algorithim, which should be the same for normal ranks, apart from duo-trio-penta player adjustments. In reality though, the system for flex should just have the same amount of premades on both teams, to save on calculations in that department. If I were in charge of everything though, I would simply delete flex queue and bring back team builder and the OOL. > The accuracy of ranks is irrelevant to matchmaking health. I agree, but the accuracy of MMR is. > Premades will always account for a degree of inaccuracy here. That's right, but there are no premades in most of these examples, as stated. The Twitch/Jhin game is the most glaring. If you are saying that the ranked system gives more weight to a players first few games (i.e. placements) than the others, then it should be fixed. Consider this: A player starts a new level 30 account and tanks it into Iron 4, then tries to smurf on it (they are Master level). Perhaps this is what happened with the Twitch player, and why their MMR is still so low after their 78% winrate across a number of games. This is simply an outdated system, and needs to be tuned into what chess or DOTA2 does, which is to give equal weight (for MMR purposes) to each game and adjust after each game. And perhaps the Jhin is in the complete opposite direction for these calculations. This could be the only logical reason why those players are in the same game. But, its stupid and needs to be adjusted faster. Because the Jhin was clearly out of their depth, and they only performed better than the Twitch because Twitch was griefing. By fixing this, there will be less incentive to troll like the Twitch and smurf from a new tanked account. Btw this Twitch's main account is called "MMM Predator" which they told me after the game. So not quite Master level as they were claiming, but still higher than the game in question. I'm guessing that because they were actually not as good as their D2 rank, they started smurfing to have fun. Because in game, they clearly lacked macro knowledge (we can talk about this in a whole new paragraph), and because they could not handle that a silver was calling them out when they disagreed with a call made, they started griefing and inting. By fixing the MMR system, you would also encourage less of this behaviour and make player reach their true ranks quicker. > There is no perfect solution, the scale simply makes that impossible. That's right, but from where I'm standing Riot can do a lot more, which will help not only them, but the playerbase. Leaving an outdated system in place is in my view unacceptable. Thanks for the discussion.
Kureneye (OCE)
: Ranked Matchmaking
I'm heading off for now, but food for thought. Whether you can pass this through to Riot Games headquarters or whether they would consider it is another story.
: 1: You will increase wait times even with an option, because even if only 10% of people choosing that option, the pool for players that don't is still smaller, and so the wait time longer. Increased wait times means less players that can or want to wait that long for a game which means less population which means longer queue times which means etc etc etc. Once the population gets too small, matchmaking health goes out the window almost entirely (see TT and it's Challenger v Gold games on the regular, hence it's retirement). It's the kind of thing that sounds great in theory but isn't scaleable in practice. Remember that a change of any sort to matchmaking can't be relevant to a few games, it has to be relevant to *every* game at *every* level *all* the time. By virtue of the same, compromise is a thing that has to exist. No matchmaking system of this size can be perfect. 2: I wouldn't be against something like this, the idea has merit, I just don't know how it would pan out in practice. Already visible rank has to catch up (or down) to hidden MMR when there is any period of change (climbing, reset, etc). This would create more times when MMR is not reflected by visible rank. 3: MMR and visible rank are not always equivalent, there's more to it. The Yi most likely was playing in games *above* your MMR, and has now dropped into your MMR range. If he performed poorly in your bracket he will drop further. 4: Does true rank Golds with a visible rank and MMR of Diamond sound like good competitive integrity? The only way for a reset liek this to work as intended is to systematically match every player against every other individual player rather than being randomly matched into games (which is impossible). The system wouldn't have any idea what rank a person should be, so when they win or lose a game it doesn't know what kind of gains to give them, and the whole thing goes out of whack. It's just big ol' dominoes. Why is season 8 MMR any more relevant than s9? By your own words you believe an MMR decay system would be a good idea, how is using 2-year-old ranked data better than using data from the previour year with a soft reset?
> [{quoted}](name=Seras Dragon,realm=OCE,application-id=FjGAIbRv,discussion-id=7qIE3EQp,comment-id=00040000,timestamp=2019-11-13T10:18:58.422+0000) > > 1: You will increase wait times even with an option, because even if only 10% of people choosing that option, the pool for players that don't is still smaller, and so the wait time longer. Increased wait times means less players that can or want to wait that long for a game which means less population which means longer queue times which means etc etc etc. Once the population gets too small, matchmaking health goes out the window almost entirely (see TT and it's Challenger v Gold games on the regular, hence it's retirement). > > It's the kind of thing that sounds great in theory but isn't scaleable in practice. Remember that a change of any sort to matchmaking can't be relevant to a few games, it has to be relevant to *every* game at *every* level *all* the time. > > By virtue of the same, compromise is a thing that has to exist. No matchmaking system of this size can be perfect. Well, this was my idea for compromise. Another Idea that I have (which some people may not like) is to do away with choosing your role in champion select, and choose it in pick-order, like back in the season 5 days. This is will drastically reduce queue times, because searching for roles will be removed. It will also increase the overall skill of the playerbase, because players will learn every role and understand what the team needs best. It means that some players will need to learn support, but they always had to back then also. And then again, because players lock in their champions AND choose their role at the same time, they may not need to learn support if there is a support main locking before the last pick. > 2: I wouldn't be against something like this, the idea has merit, I just don't know how it would pan out in practice. Already visible rank has to catch up (or down) to hidden MMR when there is any period of change (climbing, reset, etc). This would create more times when MMR is not reflected by visible rank. > 3: MMR and visible rank are not always equivalent, there's more to it. The Yi most likely was playing in games *above* your MMR, and has now dropped into your MMR range. If he performed poorly in your bracket he will drop further. But how could he have been playing in games above my mmr, which is say silver 3-4 ? With such extremely low winrates the system really needs to adjust better. > 4: Does true rank Golds with a visible rank and MMR of Diamond sound like good competitive integrity? The only way for a reset liek this to work as intended is to systematically match every player against every other individual player rather than being randomly matched into games (which is impossible). No no, you misunderstand. The ranked reset resets the MMR also! Everyone would be at some predetermined mmr (it could be silver, could be gold, whatever is a good middleground) and we go from there. > The system wouldn't have any idea what rank a person should be, so when they win or lose a game it doesn't know what kind of gains to give them, and the whole thing goes out of whack. It's just big ol' dominoes. It would operate exactly how it is doing now. > Why is season 8 MMR any more relevant than s9? By your own words you believe an MMR decay system would be a good idea, how is using 2-year-old ranked data better than using data from the previour year with a soft reset? Season 8 mmr is more relevant because it was accurate for that time before the cockup in early season 9 matchmaking and the two new tiers were introduced. End of season 8 ranked data (1 year) would be less painful for those players that have worked for their mmr (and we are assuming that the balls up of dynamic queue matchmaking has been sorted out by this time).
Kureneye (OCE)
: Ranked Matchmaking
**Picture 1 of your "OCE Matchmaking 2":** you wrote "RED: Over this many games win-rate is not an accurate indicator of true average skill. There just isn't enough data there for a single number to accurately represent anything. These kind of numbers only become something resembling accurate after 100+ games." And that's exactly my point. How on earth is that matchmaking remotely accurate? Btw saying that "Trist is now gold" is not a logical argument about the matchmaking at the time. At that moment in time who knows what their skill level was. The LEONA on the other hand has played a reasonable sample size to tank from their account starting mmr (which people say is in silver) and should be placed in lower category games. I have no problem with the blue team, just the completely randomness of the red team. **Picture 2 of your "OCE Matchmaking 2":** My point was that the Lee Sin's mmr at the time would have no bearing on why there is a an Iron 1 in the game. Like you say, they have the standard pre-placement silver mmr. But the Iron 1 should simply just not be in the game, considering that they had played a number of games to establish their level. You cannot compare the Iron 1 with the Vayne over the course of the rest of the season to logically argue that they should have been paired in this game. **Picture 3 of your "OCE Matchmaking 2":** I understand that Solo and Flex mmr are not the game. But in pairing players for their first Flex ranked games, why doesn't Riot take into account the soloqueue mmr, instead of starting them off fresh with a pre-placement silver mmr. It would certainly help to avoid these sorts of shellackings. Another thing that Riot could do is balance the TEAMS better using some soloqueue mmr weightings until players have established their Flex queue ranks over a specified number of games. **Picture 4 of your "OCE Matchmaking 2":** The Miss Fortune in this game has stopped playing Flex queue after completing their placements with a result of 4W 6L Iron 2 placement. By using the idea specified in the previous paragraph, Riot would get more accurate Flex Queue ranks and avoid some players simply giving up because their placements were lopsided. **Picture 5 of your "OCE Matchmaking 2":** Then information given was that this player (name withheld because I was afraid of the name/shame board rules police) was placed into my silver promo game. Considering that they were not a duo player, the questions are how and why? Consider the mmr, not the rank: mmr needs to be made simpler, like I said in my point 1. A 21% winrate player after 38 games is not even a high bronze level player, so you cannot even consider that argument that high bronze and low silver are not so far apart because in reality they are not high bronze level, unless of course mmr tanks very slowly. I would think that losing 22 out of your 30 games would tank mmr pretty quickly, but maybe this is not so. But it needs to be. **Picture 6 of your "OCE Matchmaking 2":** The cs numbers point was made to show the drop in skill level of players over the seasons and also because Riot invented the new Iron tier, this exaggerates it even more so. But 5 minutes into a game a Darius has 8 cs vs an Urgot, something does not seem right. Perhaps they are having a bad game, its true. I remember that when I first got placed into Bronze 3 at the start of season 5, back in the day when everyone had to play every role, players were getting between 5-7 cs per minute in the opening 15 minutes of the games consistently. The point is, resetting the ladder will make the cream rise to the top once again, and those players that are out of position will be put back much faster. **Picture 7 of your "OCE Matchmaking 2":** Perhaps the word 'caned' is not the correct term. Maybe "jipped", or "ripped off" would be better terms, considering that they have to face someone a whole tier above them, without any compensating factors on the teams. Silver 4 vs an established Gold 4 player at the time (over 600 games played with 49% winrate) is not fair matchmaking. **Picture 8 of your "OCE Matchmaking 2":** You wrote "Twitch did the worst in the game, is probably a second account yes, but if it is they are either intentionally losing (report) or are certainly not Master/Diamond. Also, 27 games is again, nothing. The Jhin had barely played any games (and still hasn't), and also outperformed this supposed Master tier Twitch so he must be doing something right. Bronze 1 rank also doesn't mean he has Bronze 1 MMR. With a WR like that he almost certainly doesn't." My point here, is that 27 games with a 78% winrate should be enough to kick start the twitch's MMR into higher placements. Why are they being placed into a game with an mmr level around his own rank?? Jhin did not outperform the twitch by himself because the twitch was inting and running it down after some disagreements on strategy in game (check the build for example to confirm). I'm not interested in what happened in the game, only the matchmaking. According to the screenshot, the Jhin had a 42% winrate, was in Iron 1 at the time after 57 games. This would suggest to me that they are nowhere near high bronze, low silver mmr. Consider this situation. 10 players. Team A has MMR: Top 2500, Jungle 2200, Mid 2500, Bot 2500, Support 2800. Vs Team B with MMR: Top 2500, Jungle 2800, Mid 2500, Bot 2500, Support 2200. Using this example because it is from high elo, the game MASSIVELY favours Team B. Why? Because Jungle at this time has way more influence on the game in the early stages than support. This is an example of lopsided matchmaking. With the idea to give players the option to sit in queue for a little bit longer, this is can be greatly eradicated. I hope you see the point here. HOWEVER in the Twitch/Jhin game, these players ARE ON OPPOSITE TEAMS. Which makes the matchmaking STUPID to say the least. **Picture 9 of your "OCE Matchmaking 2":** The 24% winrate Yi situation I've already discussed in point 1 in previous post. But just to reiterate. A player that starts on a new account, that is not duo-queueing and has the standard silver mmr, but then only wins 24% of their games over a reasonable enough sample size (enough to change mmr) should not be placed in silver games. **Picture 10 of your "OCE Matchmaking 2":** I don't see how you can use **subsequent results** to justify the placing of players in games at a certain snapshot in time. it doesn't wash with me. If the Vladimir has just finished placements and is coming out of a loss streak, why should I be placed in the same game when I'm a whole tier higher ranked? Again, by giving the matchmaking algorithim more time to find players (i.e. supports, which might have been harder to find at the time), this thing can be avoided. **The Last Picture of your "OCE Matchmaking 2":** (the high level game with Eyla). You are right here, the Qiyana wasn't gold last season. I was swayed by what someone said on the stream at the time, and didn't check it. But yes, they were autofilled and had a horrible game. At the higher ends of the spectrum there are many discrepancies that need to be fixed by resetting the ladder. For example at least 3 of the players in this game were no higher than Diamond 2 last season, and by maintaining a 50% winrate are somehow in games with the top 5 players on the server. This is a frustrating skill gap in these games, and its a major reason why pros are simply not playing many soloqueue games throughout the season, when they need their practice the most. (Some of them are playing now, to reach challenger at the end of the season). This happened because of the balls-up in matchmaking at the start of the season, combined with the introduction of the new tiers. Ok, I hope you understand my points more now, and why I've selected the ideas to help remedy the situation at hand. Cheers.
Kureneye (OCE)
: Ranked Matchmaking
Hi Seras, I'm not going to spend too long on this, because it takes time out of my day. But I will address points 1-4 and then move to the pictures later. 1: The system does not do this satisfactory enough. Earlier in the season I was dodging many of these sort of players that have been placed into the wrong level of skill, but when you have to dodge more than once in a row, it is very time consuming. So by having a button that players themselves can choose or not to, is quite beneficial. Consider that not everyone has time to play league all day and only play 1-3 games at a time. I would rather wait an extra 10 minutes in queue for every game to be matched fairly, than to be waiting in a dodge queue for 1 game for 30 minutes. 2: It's been tested. E.g. NEACE the streamer came back to play league after 1 year away. And even though he knew nothing about the new champs etc, he was placed into his old level of games (d1+). I gave an arbitrary number. Use whatever number is needed to stop player that have been out of the game for a long time, coming back and ruining the experience for others. They need to get to grips with the meta, etc etc. BTW Neace dodged the game after testing, because he knew he was not at the level at the time. This idea also will kill off a lot of account sharing, smurfing, selling, and boosting. 3: You say it will tank the MMR real fast. But 30 games is not fast enough according that example given (the YI in the pictures). When you have 24% winratio, you are clearly out of your league, or trolling. Generally when its below 40% winrate after quite a number of games, you know that player is out of their league, or a griefer, wintrader etc. 4: I understand that it would be "painful" for some players. But competitive integrity and skill matter more. I can understand why Riot will not do this, because they are afraid of losing their playerbase. But think about those players at the start of this season when the ladder wasn't reset. They were in the same position. A compromise would be to reset players to their season 8 mmr.
XzMln (OCE)
: Then you're an asshole and you deserve to have all accounts banned on site, you seem like a racist kid with no understanding of the ramifications.
> [{quoted}](name=Désired Name,realm=OCE,application-id=T8eq2lFQ,discussion-id=PP26tb3f,comment-id=00010000,timestamp=2019-11-12T22:17:16.720+0000) > > Then you're an asshole and you deserve to have all accounts banned on site, you seem like a racist kid with no understanding of the ramifications. But Taiwan is #1. J-Team whipped FPX good {{sticker:slayer-jinx-wink}}
Kureneye (OCE)
: Ranked Matchmaking
Riot thinks their matchmaking is fine. I've had enough of poor matchmaking, it happens all too often. However there are the skeptics and forum nannies that try to despute these things so I've been gathering evidence over the past few months. Now it's time to show it. Before you consider the pictures, note that for all Ranked "Soloqueue" games displayed, they were checked for players in duo, and there were none. So for those who actually care about the issue, [here is the evidence.](https://imgur.com/a/r8shadT) Now, for some solutions, they would be quite easy. 1. Wait longer to pair players of equal strength. Even have an option for players to say "wait longer for me" or they can keep it the same as it is now, if they are happy with it. 2. Make MMR decay after 6 months for all players. 3. Don't make the pairing algorithm complicated. If you see a 24% winrate player over more than 30 games, they are not at their normal mmr. Adjust accordingly. 4. Reset the ladder fully at the start of season 10. This will clear out the monkeys and make a level playing field again. Its frustrating to see pro players having to cope with idiots time and time again. The cream will rise to the top eventually, just do it.
: > [{quoted}](name=Gehirn,realm=OCE,application-id=3fteLeTW,discussion-id=v02mAZr5,comment-id=0024,timestamp=2019-10-17T06:43:30.361+0000) > > Winning teams have been determined! > > First place goes to two teams who managed to take down the nexus within the same second. For prizing, we've combined first and second place and split it equally among the 10 players of each team, granting them 4K RP each. > > https://matchhistory.oce.leagueoflegends.com/en/#match-details/OC1/259185115/201645071?tab=overview > https://matchhistory.oce.leagueoflegends.com/en/#match-details/OC1/259156226/200066214?tab=overview > > Third place goes to this team who were only 1 second behind! Players on this team have earned 1K RP each. > > https://matchhistory.oce.leagueoflegends.com/en/#match-details/OC1/259175894/201963485?tab=overview > > Thank you to everyone who participated, we hope you had as much fun yesterday as we did. The celebrations continue with 10 days of gifts for all players who had accounts before the Riot Pls stream started yesterday, [check out the details here](https://www.riotgames.com/en/news/riot-pls-anniversary-edition-recap). I'm disappointed with how poorly organised this competition was. Lack of communication on Riot's part lead to some entering teams not having the advantage of having player feeders on the opposing side.
> [{quoted}](name=Indy Savvy,realm=OCE,application-id=3fteLeTW,discussion-id=v02mAZr5,comment-id=00240002,timestamp=2019-10-17T13:52:30.952+0000) > > I'm disappointed with how poorly organised this competition was. Lack of communication on Riot's part lead to some entering teams not having the advantage of having player feeders on the opposing side. 3:58 was the winning score?? Would have to agree with Indy Savvy on this one. Looking at those match histories, theres heaps of improvements that could be made. Here's a couple of simple ones: 1) 5 feeders on enemy team, not 4 2) LULU 3) Some enemy team members (take smite!?) could be farming (after initial int) to buy speed items to get to lane faster, also could be farming to provide shutdown gold.
Rioter Comments
: Amazing this
[REMOVED - REPORTING OTHER PLAYERS] = SWEEP UNDER THE RUG
: I think you've missed my point a bit. I didn't say that not naming and shaming them would magically make them reform, I said being named and shamed *discourages* reform. Behaviour is not black and white, there is no magic reform button. You can create a combination of factors that gives players seeking reform the most chance of success and the most reason to try, but if they aren't willing to change they won't, simple as that. The behaviours you describe are on an *entirely* different scale. Consider instead a thief, or someone who has been to jail for a non-violent drug crime. They could be perfectly willing and able to reform, maybe they already have, but now they can't get a job because applications require you to state you have been convicted of a felony, without giving you the opportunity to explain your situation/how long ago it was/the severity. They can't get a job to pay the bills/feed their family so they fall back into the (illegal) means they have of surviving. Some will still persevere in reformation because it's better in the long run, but it's a struggle and everyone will be against you along the way. It's *better* in the long run to reform, but *easier* in the short run not to. To encourage reform, you have to make it as easy as possible in the short term.
This is such an interesting post that it needs addressing. > [{quoted}](name=Seras Dragon,realm=OCE,application-id=T8eq2lFQ,discussion-id=wY8Ejqot,comment-id=000000000000000200000000000100000000,timestamp=2019-09-05T07:56:22.324+0000) > > I said being named and shamed *discourages* reform. I would argue that name and shame helps to protect innocent parties from evils, and draw attention to a needed punishment, why protect the troll itself? E.g. sex offender lists. > Behaviour is not black and white, there is no magic reform button. You can create a combination of factors that gives players seeking reform the most chance of success and the most reason to try, but if they aren't willing to change they won't, simple as that. Exactly, so permit naming and shaming to help protect others. > Consider instead a thief, or someone who has been to jail for a non-violent drug crime. They could be perfectly willing and able to reform, maybe they already have, but now they can't get a job because applications require you to state you have been convicted of a felony, without giving you the opportunity to explain your situation/how long ago it was/the severity. They can't get a job to pay the bills/feed their family so they fall back into the (illegal) means they have of surviving. While I would dispute that they couldn't get a job/some form of income, I'll play along with this and say that they could get some fake id (smurf account) and work from that basis. If this sort of person is still unwilling to change their ways they will face the law once again on their new persona. If we still compare this "felony" to league/esports then this is the reason why some players were sanctioned before worlds for breaking rules. And those challenger players were named and shamed on Riot's own news page. > To encourage reform, you have to make it as easy as possible in the short term. To encourage reform, a person firstly must be remorseful for what they did. Proper punishments are essential. Let's talk about why Riot hold this double standard with esports professionals and the average player. From their point of view, they do not want to spend the finances needed to make such proper punishments. They have obviously calculated the numbers. So prohibiting naming and shaming is simply to draw less attention to their ineffective system and keep costs down.
: If he get crap teams all day he get demoted, if he get good teams all day he get promoted, if this, if that. No! stop. You dont rely on luck if you want to climb rank. The chances of him getting demoted from getting a series of bad teams, is the same chance he'll get promoted from having good teams. The guy is simply not good enough to even float on average. For god sake, all he vsed was silver 3-4, rarely gold 5. If he cant even manage to carry at that rank, why does he think he can maintained gold?? http://oce.op.gg/summoner/userName=SeibaPendragon Look at the stat, bronze 1 - silver 2 -1 top. BS that he can get to gold if he cant consistently win against bronze and low silver. Check the games that he have lost, most of it is due to playing with and against gold players. The answers is right there, just analyse it.
72 sins (OCE)
: Most mechanically intense champion?
Be prepared to lose a lot of games if you decide to master a mechanically intense champion before learning how the game works. If you want to really test yourself on both a micro and macro level, then Nidalee is your champion. Please note than very few players on this server have mastered her (can only think of 3 in fact - Shernfire, OnIy and Impaired).
BeefyBill (OCE)
: LF Support player Clash Tier 3/4 team
I was hoping to join a team in tier 4, but not gonna happen with this guy. They will judge you based on your current soloqueue rank, not on previous ranks, or accomplishments, or actual understanding of the game. Just a warning. He's looking for an elo based member. Typical OCE attitude, need to cut the western shit bro. [Oh well, his loss.](https://imgur.com/a/GwdTijs)
Bookbash (OCE)
: 8 games in a row now there has been an iron player in a silver mmr game. Please fix.
And talking about the sort of players with less than 20% winrates. Cmon this is getting stupid.
Holy Fish (OCE)
: Plat/Diamond Support Main - LF Plat/Diamond ADC duo (Silver/Bronze/Iron rank).
[Nice preseason.](https://old.reddit.com/r/leagueoflegends/comments/a4knol/kr_challenger_player_rush_just_got_promoted_to/?sort=old) [ Another example from Reddit (NA server, btw).](https://prnt.sc/lss6u1)
Bookbash (OCE)
: Can I please stop having monkeys in my team
8 games in a row now there has been an iron player in a silver mmr game. Please fix.
Rioter Comments
F4R W3ST (OCE)
: SoloQ matchmaking
[Another example, a few hours later.](https://imgur.com/a/ar9LDTL) You get used to it.
: Draft Normals matchmaking is so garbage
It's normals, and it's oce. You should be happy to play against higher ranked players though! It's boring for them, but great for you!
: Seriously, Riots banning system has to change or i'm quitting this game
[Just stay cool, and send in the report.](https://boards.oce.leagueoflegends.com/en/c/player-behaviour/ksPHUhAm-intentional-afkers-not-punished?comment=0000) P.S. Player has been banned, hasn't logged in for 15 days.
slxw (OCE)
: Mastery lvl 7 / bullshit
There's lots of western invented things in the game that mean nothing in the grand scheme of things. Emotes, skins, mastery levels, borders, icons, have nothing to do whether someone is a good player or not. Don't sweat it. I only level up champion masteries when I have the shards to do it. Otherwise its a waste of Blue Essense, which could be used to buy new champions. You have the intelligence to find the forums, so how are you bronze 5? [Not trying to rank up?](https://www.youtube.com/watch?v=slgxQ8MW2WQ)
animos1ty (OCE)
: new ranked changes in 8.24 have screwed with my lp win/loss points!?
F4R W3ST (OCE)
: SoloQ matchmaking
Just oce things. [Here's a game played 2 hours ago.](https://imgur.com/a/Gh0Cq1C) Luckily for me (silver 3) the diamond 4 Riven was still a monkey at macro.
NaTionzOG (OCE)
: Sorry, 250 wins is miniscule :)
Well, make it whatever number of wins (500?). The number is not so important, but the experience of closing games, and playing to win is the point im making here. Monkeys entering ranked should just be placed in Iron 4 and if they are good enough they will climb. Punishing established players in ranked by allowing new accounts with no experience into higher mmr brackets in a small population server is just not acceptable. Especially when you see that a lot of these accounts are really just banned bronzers that bought one of those bot accounts with ids like "#@$!firstname".
Gehirn (OCE)
: OCE Rioter Q&A - (Group 1 of 3)
To Belquin: Can we please grow the scene by bringing back OOL and the ability to make your own esports team, and reward those teams that qualify into the OCS like previously? The scene will not grow with just having the same old orgs and organisers. Not only did this help showcase good players, it also helped showcase good managers, coaches etc. And more people were involved.
: KKJedi got his account back LOL. Good luck Platinum players.
Looks like they are already inting in every game.
NaTionzOG (OCE)
: I would have to disagree with you on this one, 250 games is miniscule and is less than what is required now. having level 6 on 20 champions requires many hours of investing, maybe 3-5 for champions you play, but an understanding on other champions would be exquisite in my opinion.
That's 250 wins, not games :)
DBZChiChi (OCE)
: Intentional afkers not punished?
Those sort of players have the worst attitude to the game. Had this inting Janna in my game recently. They tried to hold champion select hostage by saying "Someone needs to dodge". After the reply came back "yeah, you" then stated that they would int and just take cleanse and ghost summoner spells. They not only ran it down the lanes, but kept the surrender vote invalid until 20 minutes of our time was wasted. This is one of the main reasons why people quit the game, you cannot escape from the time wasters. [NEACE](https://www.twitch.tv/neace) said it best on his Fortnite stream when he compared this team game to basketball. If you join a few players for ball at a local court and they act like arseholes, you can just leave the game. You're not trapped in that situation. So my friend, it is going to cost you a few minutes to make sure they are punished correctly. If you feel that this is worth it for the community's sake, then do this: 1: Screenshot in game action 2: Copy the pre and after game chats 3: Upload to a site where you can combine them all 4: Include the link in your report Should probably take you about 10 minutes after the game, but they won't be coming back. Just like the Janna. [Link removed, censor the names if you wish to re-add it but I feel the point has been made.]
NaTionzOG (OCE)
: People in ranked are shit
A better solution would be to have new accounts score 250 wins in normals before being allowed to play ranked. And for the player entering ranked to have level 6 mastery on the minimum (currently 20) champions to enter the ranked arena. Would raise the quality substantially. OCE macro would still be garbage, but you might see less inting Luxs, Lee Sins, Vaynes and Yasuos out there. Not gonna happen.
: Not true at all, they have more hp, get more resistances the more plates have fallen and even more ad damage
He's talking about after 14 minutes, or the towers other than the outers. And he's correct, they lost their armour in 8.23 and you can knock em down with your pinky. Fake buffs to make the community think they are stronger, when Riot is looking for more quick clown fiestas that make games gravitate to their conclusions a lot faster. More games, more random results, more skin buying, everyone's a winner! A simple solution would be to give them their armour back, but that would defeat Riot's purpose of the changes. [So it's time to play Sion, Jax and Tryndamere!](https://boards.oce.leagueoflegends.com/en/c/miscellaneous/ozM8Aq0n-i-hate-823?comment=0000)
: i hate 8.23
Don't worry it's been reverted to 8.22 LMAO, causing afks in games. I guess CHN overlords didn't like that OCE got a preview 1 day early. Edit: and now it re-installs 8.23 wtf Edit: no one in low elo is trying the broken stuff yet, like 1) EZREAL with kleptomancy 2) Viktor and Sion top buffed 3) Towers are now butter, despite the "buffs" so use Sion and Jax to mow them fast 4) TWISTED FATE xD P.S. sign-in doesn't work, rip.
OCEshots (OCE)
: I think i feel so strong about not giving up so much is because i have accounts in almost every elo up to plat. bronze and silver are so easily carried purely from macro. wave control and objectives win those games. i Opt for champs that can solo baron at 25mins (with good farm and gold lead) and just put your whole team and your back and give them some free LP. But yes totally agree that people that give up that early are annoying but also no longer see it after my perma ban on my 3rd gold account because i just mute all so i don't speak. A lot of people should not be playing ranked and not because they suck or anything along those lines but purely because their mental isnt in the right spot.
Yep, people think too much about their rank/kda/ego/skins/memes, than the actual game and what they need to do for the team to win it.
Bookbash (OCE)
: Waiting for this! Equal ranks for roles please.
[Then we can avoid lopsided matchmaking like this.](https://imgur.com/a/86Vn0k7)
: From what I read, they want to test the new role ranked system for a while first, so might be a longer preseason. Most likely late Jan.
Waiting for this! Equal ranks for roles please.
Typhon (OCE)
: criticism or helping?
Helping is when you give constructive advice for teammates that will help the team progress. If kids have no balls and take it as toxicity, then unlucky. Examples: "Try a hexdrinker", or "be sure to react to the inting Sion before he gets close to the tower", or "You'll need a QSS", or "We need baron to close because we have no siege". None of the above is egotistical, or personal attacks. But kids still take it the wrong way.
QAPLA (OCE)
: > [{quoted}](name=Bookbash,realm=OCE,application-id=Ntey9fRZ,discussion-id=LRMaetzt,comment-id=0001,timestamp=2018-11-12T07:44:03.792+0000) > > I was away from league for 3 weeks and missed the whole of worlds and together with whatever token system they have. But who cares, I just play the game for the skill, not the token, kd/a weeb stuff or icons. Spend money on this game? Never. I like free skins, I hate the whole anime weeaboo thing it's into now. Now don't get me wrong, I've been watching anime since the 80's (when it was good. I haven't watched much anime made this millennium because it's mostly derivative trash), but this is not an anime game, so fuck off with the weeb shit Rito. Every remake of a champ becomes more anime, it's fucking sad.
Agreed with the champions thing. The older champions have character, with a jovial side to them. These new champions and reworks and "visual updates" are just sad. They need to focus on the skill and gameplay, rather than appealing to trends. Last good champion released was Kled. Don't rework any champions.
: The only time where I’ve reported someone then they got banned is when someone told me to kys. Otherwise, with the ones that int I get no player behavior banned blah blah blah message.
Same. Trolls that i've reported multiple times are still running around. Like the bard (no names because we can't handle the backlash) who ints 75% of their games by running around doing nothing.
A Bronzie (OCE)
: Player giving up
Kids will generally give up yeah cos they have had everything spoon fed to them and "everyone's a winner" mentality taught to them. So when they are losing they don't know what to do.
YuN0Die (OCE)
: How in the actual hell is this possible? Troll after afker, after inter, after troll, after afker
This seems normal for me. Usually one day I'll win 10 in a row and then the next lose 10 in a row. In the latest survey that was in the client, I gave Riot some good ideas to fix matchmaking, lets hope they read it :)
QAPLA (OCE)
: Why I hate Missions
I was away from league for 3 weeks and missed the whole of worlds and together with whatever token system they have. But who cares, I just play the game for the skill, not the token, kd/a weeb stuff or icons. Spend money on this game? Never.
Mister E (OCE)
: LFM players to do some last day flex ranking up
Add me if you need a solid support for your flex queue and can handle my shotcalling :)
Bookbash (OCE)
: I was in a game where the enemy Sion did this strategy. It's frustrating because after advising my team what to do against it, they didn't listen, and the 4-16 Sion won the game basically, so I can understand your frustration. The strategy basically takes advantage of monkeys with no macro understanding, and therefore quite good in low elo. In my game Sion did this vs an Kayn, and we had a Jax jungle. The problem was the Jax jungle didn't pay attention to the Sion before he got close to the tower, and went off splitpushing himself because he was clueless. I was trying to advise him on how to counter, but he didn't listen. Another case of that one player in a team costing the team the game. Eventually the split-pushing Jax in the bot lane lost baron of course, as you get with these types of players (Udyr is another common one). As usual, splitpushing is either misunderstood at low elos, or players just have no balls and do that as a way to silent int. In my game there was a time where myself (support), and mid lane Diana were in base and Sion is coming in the top lane, and the Diana just walked mid and ignored him. I'm pinging them over and they just ignore. Low elo monkeys. I see that in your game you managed to win at least with your bronze 3 top laner. Not too bad to get a result there. The other good news is that now Riot realises that they made a mistake all of those patches ago, making towers too easy to kill (looking at you bot tower #1) and are changing this for preseason. We will have longer games again, with more strategy, instead of a quick hose with animals running around looking for meat.
UPDATE: I had a think about this and decided to counter it with Lulu support. Just polymorph the zombie sion before he can kill the tower. [Problem solved.](https://matchhistory.oce.leagueoflegends.com/en/#match-details/OC1/223995109/201025528?tab=overview)
Elysian88 (OCE)
: how about you fix this inting sion crap already...
I was in a game where the enemy Sion did this strategy. It's frustrating because after advising my team what to do against it, they didn't listen, and the 4-16 Sion won the game basically, so I can understand your frustration. The strategy basically takes advantage of monkeys with no macro understanding, and therefore quite good in low elo. In my game Sion did this vs an Kayn, and we had a Jax jungle. The problem was the Jax jungle didn't pay attention to the Sion before he got close to the tower, and went off splitpushing himself because he was clueless. I was trying to advise him on how to counter, but he didn't listen. Another case of that one player in a team costing the team the game. Eventually the split-pushing Jax in the bot lane lost baron of course, as you get with these types of players (Udyr is another common one). As usual, splitpushing is either misunderstood at low elos, or players just have no balls and do that as a way to silent int. In my game there was a time where myself (support), and mid lane Diana were in base and Sion is coming in the top lane, and the Diana just walked mid and ignored him. I'm pinging them over and they just ignore. Low elo monkeys. I see that in your game you managed to win at least with your bronze 3 top laner. Not too bad to get a result there. The other good news is that now Riot realises that they made a mistake all of those patches ago, making towers too easy to kill (looking at you bot tower #1) and are changing this for preseason. We will have longer games again, with more strategy, instead of a quick hose with animals running around looking for meat.
: How are people not running locket on sion yet
Just keep doing what works for you. When you get to a higher skill level (it's oce with monkeys in terms of strategy, so probably diamond 2+), you'll see why locket on Sion top is not efficient. Don't complain about idiots not using good strategies, keep it a secret, or use it to boost for money :)
: Soz by looking at my history ill say i fuckking carry everygame and the point that pisses me off is people either leave half way in the game or start not playing as a team because as i see it with the past fuckking 700 games each game has been fuckking horrific and this system of matching with people is poorly done
Maining Vayne in low elo is not easy. People understand nothing. Try a better laning adc like Varus. Get that first tower, get ahead, and monkeys will gravitate to the win. I agree though, the system is trash and the players are just animals.
: Korea carries LPL to Worlds win!!!!
FeelsBadForMataMan. Couldn't carry from support.
Show more

Bookbash

Level 176 (OCE)
Lifetime Upvotes
Create a Discussion