With the introduction of badges last season being able to track more complex things like doging skillshots, missing skillshots, keeping opponents off the wave etc. They'd probbably already have a good delegate and listening system in place to calculate it with the new Truskill

@celfhelp

5 ай бұрын

wouldn't be surprised if tokens and eternals were a series of tests to see how well they could track literally everything that happens in games for this reason

@stunlockedirl

5 ай бұрын

that's what i was thinking. they already track a lot of different stats that can pretty accurately state your skill. i'm thinking this new change will likely be really good.

@juliogomez2534

5 ай бұрын

This is why I keep saying it's bs that they can't fix the LP system. They already have a system that tracks all the players performance and assigns rewards based on their contribution, sure the badge system might have been them testing the system, but them saying that tracking all the variables it's hard when they are already doing is a straight up lie.

@nikdonobody1327

5 ай бұрын

Yeah they can track a lot of things sadly most is micro not macro so as per what i heard master+ players say its meaningless after gold/plat where micro mostly platoes what im most scared about is of meta its going to fuck any Complex system either performing ridiculously well or be impossible to climb with lp is flawed but statisticly over long enough its going to give most accurate information about your ability to win games and work with team

@ChronicTHX

5 ай бұрын

If you go to your stats in your account in the client, you can see metrics like gold to objectives conversion ratio or damage to objective ratio. They already have usable stats that are meaningful per game that don't need honor system or badges. Also, badges are irrelevant by themselves

I personally think the root of nearly all toxicity stems from the frustration inherent in the LP grind. Nobody is happy to start their day off with a 3 game win streak and end with a 3 game losing streak, undoing everything they accomplished that day and turning the game into upwards of 4-6 hours of completely wasted time. Especially if the losses weren't the player's fault. I think we've grown enough as a community that we can admit that not every game is winnable, especially if a teammate actively does not want you to win.

@sirbobulous

5 ай бұрын

Especially because you can't know if those 4 others you're paired with are currently running on a high from good games, just started playing for the first time today or are basically tilted already after 4 losses and playing with 2 trolls in their previous game so are literally waiting for one single thing to go wrong to throw the game (I dunno...maybe someone with AoE damage walks past a jungle camp too close, procs damage on it, pulls them out of position and now the jungler loses 2 seconds running to get them - time to throw!)

@MrNicoteco

5 ай бұрын

In Valorant I found 3 afk in a row and lost all 3 games, did I lose less points because of that? No.

@HeresorLegacy

5 ай бұрын

Which in my opinion is another form of toxicity. Maybe I'm just oldschool, but "back in my day" games easily lasted upwards of 40 min. Either you actually win with your early/midgame comp in that time, or you get outscaled or outplayed. People feel entitled to easy wins and don't know how to play from behind. And because they throw the towel as soon as they get behind they never LEARN how to play from behind. It feels like a lot of League-players don't play for the game's sake, but for the LP. Of course you don't want to play a game out that you are struggling with if you hate actually playing it. And tbh, I don't need those people in the game.

@michaelprior1966

5 ай бұрын

While I see your point and understand it, it is still a videogame which is made for having fun and improving on yourself.

@metodar2906

5 ай бұрын

Back when i started playing i was playing with 5 people all the time and they were blind games and was fun all of the people i played lol i found them in lol even today 7 years after i still play with the from time to time But the game and the comunity change so much over the years its not even fun to que anymore you just lose from the lobby and about toxicity how can u not be toxic when u lose vs voli tabis;doran ring and the 40 armor as garen with ignite the game its try too mich to be the Game for all Mobas there are too many items that fix champ weakneses sy for my english if is bad

I have zero faith on riot's ability to balance anything, including rank systems.

@guigs4467

5 ай бұрын

Luckily for you, TrueSkill 2 was invented by Microsoft.

They' re fixing it already? It's only been 15 years!

@Digi4DD

4 ай бұрын

True, but multiple factor assesment of performance requires tech and computing power (A LOT) which they didnt have back in the day

I’ll believe it when I see it. They’ve said this so many times

@ziggs123

5 ай бұрын

The noobs biggest fear is a system where they can't blame someone else

@louisrobitaille5810

5 ай бұрын

Have you not read the Trueskill paper? It's not surprising they've failed to implement a good system after trying so many times when it's basically 10x as complex as building a rocket from scratch 💀. ("It's not rocket science", it's worse.)

@kjax139

5 ай бұрын

@@louisrobitaille5810 never said it wasn’t. They shouldn’t be saying they’re “fixing it” when each time it’s never fixed is the issue. Your point makes my agreement even stronger. They keep failing yet claiming they’ve figured it out each time

@agustingomez3284

5 ай бұрын

​@@kjax139 This time they say they are changing it not fixing it...

@Sniperfuchs

5 ай бұрын

@@kjax139 genuine question, what do you think "fixing" means? And have they every even used that term? From what I've seen, they have always used words like "improve" and the likes, because you make it sound like they claimed they solved 100% of all problems with matchmaking (a problem that probably will never have a perfect solution in your or my lifetime). This sounds very much like players reading way too much into what the developer is trying to convey and then being disappointed at their own misinterpretation.

Riot is allergic to announcing things in a proper manner. They rather scatter all information on random posts and replies throughout many different social media.

@zonkodraw3480

5 ай бұрын

Cause they are afraid of facing backlash which is a fair reason to never engage with the community. Imagine being a rioter which engages with the community every single day. The moment something goes wrong in something else not related to your area: you become the target. You get called out. Death threatened even.

@gamasannin5668

5 ай бұрын

They do announce things in the dev videos. And these infromation from reddit and such come from Riot employees not from Riot itself. I think they will announce this for sure but these things might be in early state or they want to keep it a secret for a while.

@opportunity3278

5 ай бұрын

​@@zonkodraw3480 phreak momennnt

@siduxjxhdgzhdjxhxuuxxyhgg1079

5 ай бұрын

​@@zonkodraw3480 you could, use official accounts instead of announcing things with manchildren like phreak or hiding behind developers to face the backlash. but who am i to judge big corpo

@louisrobitaille5810

5 ай бұрын

The problem is people associating Riot employees with Riot itself. A Rioter talking about something on Reddit is literally just a dev talking about what he's working on. It's not Riot announcing anything. If the project the Rioter worked on doesn't succeed, Riot will just shelve the idea and never even mention it.

The reason why a system like this would be great for the game (if it worked correctly) is that it would reduce the average necessary number of games you need to play to get to your actual skill rank by reducing the influence of random factors in lp gain (mainly matchmaking) and thus reducing frustration. If everyone had endless amount of free time there would be no reason to get frustrated over being unlucky for 30 games straight because eventually you will get it back, but many people don't have the time it takes to even out the influence of the random factors.

@TheMastertbc

5 ай бұрын

you would have to pick your lane an champion and then get matched with other players so the champs don't unbalance it

@christonchev9762

5 ай бұрын

thats a big IF tho cause how do i know that a player now is been reported even tho they are saying that the report has been sended and the X guy was banned?

@Fanofmusic106

4 ай бұрын

The only thing i see positive from this system is lessening lp loss if you played well and maybe slightly more if you played well. I feel like riot mmr system is fairly accurate tbh and lp gains are pretty fair this season. I can jump from +25 -29 to +30 -27 from a 3 win streak. But maybe thats just me?

@deny_s

4 ай бұрын

@Tomviel YES! THIS! Finally someone put the problem in the right words ! I don't understand that so many people, streamers, youtubers, pros, game devs talked about the MMR issue without ever pointing out the true issue: the necessity to grind like crazy to get to your actual MMR. You have to have at least 1000 games in a season to consider that you played enough time to compensate for the negative random factors (like having 4 noobs in your team, or trolls, or AFKs). Which points to another important factor: LUCK. You can theoretically win 80% of all your games by simply having better teammates that your opponents. In which case, you can rank up faster than light even if you're average.

years ago, i won 9/10 of my placements and got placed in bronze 5.... at the time there was nothing lower. years of messing around and having fun ruined my mmr without me knowing. after that i gave up on taking it serious.

@kiega123456789ooo

5 ай бұрын

yeah honestly, the whole smurf problem occurs more because some people wanna mess around and have fun (with friends, on their own) but want to still actually play 'at their level' on their main account. the bias towards 'people wanna dropkick newbies' is just from louder players like streamers and those inclined to complain/gloat online...

@mechanicalsilence1

5 ай бұрын

​@@kiega123456789ooo if you want to have fun go play normal, dont come into silver when you have thousands of games and a much deeper knowledge of the game than those at this elo.

@HighLanderPonyYT

5 ай бұрын

"Is just louder etc." No it isn't. Stop being a donut and play unranked if you wanna mess around. Smurfing is lame. @@kiega123456789ooo

@ReptarKingOfMeat

4 ай бұрын

I abandoned my first account years ago because I owned a lot of champions I was bad at and it screwed me over when ARAM was introduced and rerolls weren't as frequent. Now on the off day that I decide to log into it to play the mmr on that account is in hell. The people I get put on a team with are all trolls inting and saying weird crap in chat the whole time, every game, and it's norms. I guess the game thinks that account is being sold to smurfs or something since it gets used so little.

@Ziggyvu

3 ай бұрын

@@kiega123456789ooonot biased if true

As a support main I have numerous instances where I end up with a really poor KDA, but its my deaths that have allowed teammates to get pentas or aces or objectives. I dont know how they can gather enough info to see this as a net positive for my ranked progression, but I look forward to seeing if it will work.

@sirbobulous

5 ай бұрын

Do the Assists you're giving by setting up the Pentas not get tracked? Or is there too much time/distance between your death and the kills? (e.g.: you delayed 2 enemy carries in the Jungle for 20 seconds preventing them from reaching the big teamfight in midlane)

@ace_sweethell292

5 ай бұрын

Vision, landed skill shots, time spent in enemy jungle without deaths, etc. A lot of possibilities there

@bohomazdesign725

5 ай бұрын

Its actually easy. Riot already tracks a shtload of data that they are using for those useless Token and Eternal systems like for example "Dodged Skillshots", "Ward takedowns", "Save an ally with a shield / heal who would've died with the next damage input", "Immobilize Enemies that afterwards die", "End laning phase early (before 7 minutes)" and those are just a few. They are tracking all the data they need to make it fair also for Supports. The only role that might be a Problem is TopLane due to the roles nature of applying map pressure that is hard to measure. For example if Im constantly drawing attention of 2-3 enemies to my part of the map, but my team is sht and doesnt do anything with the numbers advantage how do you even measure in that case my contribution to the game when there is no hard data to back up that you actually did something correctly, but your team fcked up (TopLaners have consistently the lowest Kill Participation in the game and very often your contribution to the game is avoiding to be killed while drawing the enemies attention and often you dont even get that many towers, because those are covered by multiple enemies that are busy dealing with you and not doing anything else). Support is easy to measure. Measuring TopLanes performance is the real problem.

@neethaii4689

5 ай бұрын

I feel you. I'm very curious as to how this system will work for supports and tanks.

@mufasa.alakhras

5 ай бұрын

plot twist for supports = the more you die, the higher LP you get

The grind in ranked is unreal. Was playing as Silver 2 in full Plat lobbies because apparently league thought I belonged there (reached Plat 1 back in S3) but even though the games were more challenging, my LP gains were the same as other Silver players in full Silver lobbies. How am I supposed to climb out of Silver if my lobbies are full of equally skilled players ranked higher than Silver. Managed to hit Gold but stopped there as it is just the biggest time waste...

@ddkillswitch

4 ай бұрын

Fr… I was plat and got placed against 5 diamonds. Like.. there’s not a whole rank of emerald before I reach diamond from plat??? Yall couldn’t give me emeralds or plats who are IN my rank? Diamonds??? Really riot? 5 diamonds on enemy team is the only people you could find to match me up with

@miarawidat1

4 ай бұрын

not possible lol

Can't wait for riot to apply the changes so they can match a high skill midlaner against my silver 1 hardstuck midlaner and call it fair cause my trueskill balances it out (the same thing they already do, on 3 out of 4 lanes)

@julianramos9368

5 ай бұрын

I mean if they do that they should match LP gains and Losses accordingly

@LudwigVaanArthans

5 ай бұрын

​@@julianramos9368spoiler from the future: they won't

@monkjoint7284

4 ай бұрын

you mean the high silver hardstuck filled support player against 80%wr low plat ? yea for past 100 games it was mostly role fills that determined lanes xd

I feel it's a double edged sword, because sometimes some statistics don't translate that well to actual performance in game

@daydreamteam1366

5 ай бұрын

case in point: Braum. Braum has NOTHING in graphs and stats. He doesn't mitigate that much damage, You can't see the number of damage he "saved you from", his CCs are all very impactful but short, His damage is impactful but low because it's bursts and unsustained, Him being "threatening" can't be counted. ALL OF HIS STRONG POINS are unaccounted for.

@sirbobulous

5 ай бұрын

@@daydreamteam1366 I wonder if it tracks stuff like damage caused by another champion but given to them by another? E.g.: Leona passive means an ally deals more damage on their attacks. Is that tracked as damage from Leona or from the champion who actually dealt it? Or how about healing prevented by things like Grievous Wounds?

@Master00Ra

5 ай бұрын

@@daydreamteam1366 Time to buff Braum (?).

@mendronesg

5 ай бұрын

​@@daydreamteam1366 you're just thinking of his battle stats. You have to account for all the Support metrics that will count to the overall "score". wards (only here there are more than 5 different ways to earn a good grade at the end of a match) - clearing, warding, useful wards, advanced wards etc.. assists objectives there's a lot...

@pwl2992

5 ай бұрын

@@daydreamteam1366 i think that for invidual champions they will tinker this stuff after release, because lets be honest how many people play braum

If one person leaves match, they should pay for everyone elses loss while others get mitigated -lp If one person leaves, and team looses then everyone would be counted for around +-20 lp loss, but everyone who played gets -15 only, 5lp loss less And the afk/leaver loses 40lp Also that would push afker/leaver more below others, so they loose their rank faster than the ones actually playing Everything else i think should be counted as it is now, but they need to deal with the new accounts and smurfs creating inflation in ranks so it's more fair

@maliapaffenroth8335

5 ай бұрын

I really haven't played league enough to fully understand, I have 6 total ranked games, but isn't there a decent market for iron 4 accounts? Would that help them make iron accounts? Just leave every match

@fasasplay

5 ай бұрын

Not a good system, why would the afk guy get to stomp lower level players just because he is afk, Low priority Q is actually a pretty good punishment for beeing afk, it should just active more quickly I think, and start at 10 min or something

@arcadeyous

5 ай бұрын

@@maliapaffenroth8335 no, it wouldn't because they would get banned?

@ukuviispert9599

5 ай бұрын

@@arcadeyous if they get banned whats the point of the -40lp thing in the first place? and what if they went afk because their computer crashed or something?

@sStyleBender

5 ай бұрын

Less LP lost while you got a leaver is actually a "fake" thing Riot invented to make players feel better. While on surface you see your LP not going down at that time, your mmr is actually not affected (its still counted as you lost) at all and the LP you didnt lose by that time will be lost little by little on your next games.

The real problem is the toxicity in ranked. Last split I only played the placement matches because 3 of them had the top laner rage quit after 1-2 deaths, 1 had an adc with 5 cs at 13 minutes, the final had a midlander intentionally walk down mid trolling. Lost all placement matches because of that shit and just gave up on ranked.

Dunkey brought this idea 10 years ago. His video was ahead of its time. It literally foreshadows every issue and solution people have today.

@Conflixx-92

5 ай бұрын

There was no need to foreshadow anything. We've had this problem since day 1 in League. Simple as that. It's been an issue since ranked released and Riot simply doesn't know how to fix it.

@DeathDespairDestruction

5 ай бұрын

​@@Conflixx-92cuz they are a small indie developer

@ManicRipper2

5 ай бұрын

These issues and solutions were already solved 10 years ago. Riot just chose not to intentionally.

3rd party apps made a calculation on how you do in a game. For example, blitz calculate your warding points, the time you upgrade your support item, vision score, dmg (relative if you are enchanter or dmg support). You can see even you are 0/8/22, you can still be the game's MVP. I hope that's where the system will base the next MMR

@blinddarm8478

5 ай бұрын

yeah but blitz isnt doing things in a good way. for example most games i get a "u had 2 stacks on your ward item for 4 minutes. you waste free wards!". Well dude, if im top, i see mid in lane and jgl was botlane 1 second ago. why would i waste my ward? THATS the acutal waste, and every time blitz tells me this im like "yo wtf"

@jamilangon5798

5 ай бұрын

@@blinddarm8478 i said for example bud, not making it the basis solely for TS2. but there are things that are bad on blitz like the tracker if you got caught on a teamfight if you are adc etc but that's how it is designed.

@deluxegaming8816

4 ай бұрын

supports dont need to be any more boosted than they are

I heard it's the same system Halo infinite uses and I must say, that system was pretty good. When I started playing it, it gave me much more mmr for wins that I was losing for losses until I reached the level at which I genuinely felt I belong (not because of it being the highest, but because I finally felt like Im in a fair match, before that it was playing with clowns vs clowns) I noticed it not looking that much on a kda, main focus likely being objective points (the most points gained were in the matches where I played objectives only, not chasing kills) League has much more variables and is much harder to solo (especially depending on roles), it's not an fps where you can just farm the whole enemy team over an over, so idk how it's gonna work here but I have hope in it

@gobbledeeznuts4045

5 ай бұрын

Yes the Halo infinite matchmaking was pretty insane, like the majority of my games in plat were always very close. Like 49/50 kind of close

@bj_cat103

5 ай бұрын

@@gobbledeeznuts4045 it was probably the best ranked experience I've every had (splitgate had exactly the same system). Sadly, the game itself wasn't all that great and got boring very fast

@vanjamenadzer

4 ай бұрын

MMR system in LoL is actually garbage and once you tank your mmr, climbing on that account simply becomes nearly impossible. I currently lose -34 and gain +16, same with my friend. We are both diamond players, our mmr is now tanked and we can't climb out of plat. 80% of the games it two of us who actually play and know what we are doing but we can't play for other players and the skill gap between us and average plat 2-3 player is massive. We both don't belong there but we can't climb out. Win 3 games, gain 45-50 LP, lose two games, lose 70. 5 hours of our lives gone for a negative LP gain. One more example of how bad mmr system is. I had a horrible mmr on my first account, was stuck in gold elo and it took me 1000 games that season to reach plat 4. Then my acc got perma, bought a fresh one, queued up, played placements and got placed into plat 2. This was also the first time I reached diamond. So basically, same player, same champ (I'm otp), different account, higher elo. This goes to show how bad the mmr system is.

there clearly is (or was) a problem, because i have a low plat account newly made this year, and my main account where i stuggle to get out of silver, this isn't right, something is definetly wrong here

I'm looking forward to see how this goes. They already brought me back with quickplay after quitting for 3 years. I would believe they at least have good intentions with this. I wouldn't mind seeing the 0/10 yasuo who didn't do anything gets -30, but the adc on the same team who did their best was 15/1 with 300 cs, several tower kills, etc., gets +5 or +10 lp for completely outperforming their opponent even though they were still on the losing team. It would make losses for those who actually put their heart into it worth it.

@vladys5238

5 ай бұрын

but what if then the 15/1 ad wanted to stay 15/1 and didn't take a risky play that could've won the game cause knowing they might fuck up their score and if they lose they get less LP... I don't feel like encourgaing extremely safe playstyle and KDA players will make the game more fun

@fireemblemaddict128

5 ай бұрын

I think in that scenario, something like -10 LP. Winning or losing should still be the biggest slice of the pie. Performance is relevant but not the primary factor.

@vegardkjrberg1929

5 ай бұрын

@@vladys5238 can also look at stuff like DPM. Anyone can be 15/1 by just sniping low hp people and essentially do 0 dmg. Had lots of games where me as the adc have had 10+ kills and low deaths and I think I did lot of dmg but turns out i barely did anything. Forces you to fight more but dying shit ton of times with bad kda punishes so you have to play proper. Like knowing when to take fights to keep up the dpm and kda while at same time not completely run it just to keep up the dpm.

@koy7896

5 ай бұрын

@@vladys5238 That's so oddly specific but also I think the objectively good thing to do in a scenario where you are way ahead of the enemy team but your team is 0/10 is to just try and stay alive. If you die you lose the game, like there's no chance. Even if I was the 0/10 Yasuo I think he would deserve to win get a reward.

@byeguyssry

5 ай бұрын

@@koy7896 Imagine the situation where all 5 enemies are hitting your Nexus. Only you are alive. If you have a 0.1% chance to stop them and win a 1v5, you wouldn't try it if it ruins your KDA, if KDA is a big part of your LP gain/loss

Would be cool if good pings also contributed to final score. As a jungler I feel like I'm the only one looking at map

@gamingandchill4186

5 ай бұрын

I agree, pings can be vital if used correctly. However, that’s also hard to gage considering how much people spam them sometimes… Additionally, I think Riot should look at the actual built-in stats/graphs displaying what you did in the match when it comes to giving LP/MMR/whatever they turn it into now. That can give accurate results more often since, well, damage dealt numbers don’t lie. Kills can be stolen, but the damage dealt can’t be removed in the total output of the statistics

@yoshiwhere148

5 ай бұрын

i think views like this are required for Trueskill 2 to even come close to being better than a standard elo/mmr system

@vegardkjrberg1929

5 ай бұрын

@@yoshiwhere148 Ye I also thought of using stuff like DPM. Anyone can be 15/1 by just sniping low hp people and essentially do 0 dmg. Had lots of games where me as the adc have had 10+ kills and low deaths and I think I did lot of dmg but turns out i barely did anything. Forces you to fight more but dying shit ton of times with bad kda punishes so you have to play proper. Like knowing when to take fights to keep up the dpm and kda while at same time not completely run it just to keep up the dpm cause there are tons of games where the 0/10 guy did more dmg than the 4/1/2 guy simply cause he just sprinted in and did all his buttons before he died.

@TheBossManBoss319

5 ай бұрын

You are.

@keanneraymesobre1171

5 ай бұрын

So after they fed the enemy team they just gonna spam ping to ease up the lp reduction?

Getting put in plat after promos are finished requires you to have master-challenger normal game mmr, which with a bought aram/botgame account is around 8 wins in a row. This is actuslly hard since in the last games you often face actual challengers and rarely face anyone below diamond. All this is just to get placed in gold from your first game. After this you still have to win your games to get plat while facing people in that rank and possibly higher after a few games.

@stephanojenkins7636

5 ай бұрын

No fresh account in promos is facing challengers lil bro. Look at an account like drututt in korea where he went 28-4 and promod to emerald 2 and the highest mmr player in the lobby is a 50 lp low master player. It's literally impossible to see high elo players in early games, you're stuck playing with diamond and low master players. Once you get to d2+ with still an 80%+ winrate (which takes at least 50+ games...) you can start seeing high master and gm players Virtually all accounts with emerald/diamond borders you see in gm+ games are decayed alt accounts, not fresh smurfs xd

@antanassmetona4054

5 ай бұрын

@@stephanojenkins7636 sorry to tell you this but you're unable to read. He literally said he has to play normal games against challs just to get placed gold and then play games stacked up against him over and over in every game. Never did he say "play against gm in first games of ranked", you're just lying and making shit up. What the guy above said though is relatively true.

@hdkfshkskfe5186

5 ай бұрын

@@stephanojenkins7636 are you able to read? I’m talking about the normal games before entering ranked to max out your normal mmr. Don’t embarass urself on a topic ur clueless about

@SevenTheMisgiven

4 ай бұрын

@@stephanojenkins7636 It's like you are 8-0 or 9-0 in promos and you are facing diamond players in the last 5-6 games but even if you went 9-1 they will place you a lot lower while still remaining in the same MMR. So you are gold 2-3 but you have to play what previously were diamond players. In other words to get to platinum you have to play better than players who frequently make it to diamond while you are valued at gold 2. It's an ancient issue that just doesn't really reflect how someone is performing 80-90% of the times. If you are gold 2 stomping low diamond players you should probably be plat 2-3 at least. Not some mid gold player who is just on the shit end of the MMR system. Nowadays we have Emerald in between as well but it's kind of the same story. In 3v3 I went just 17-0 before I faced a rank 33 solo queue player while just being diamond. I don't think you have any idea how big the problems are nor what is being described. But Riots new system won't fix it. You don't get more LP because you just made the game winning play. And I don't believe they can do that.

@etmon.

4 ай бұрын

legit just had a level 34 player in high plat game. dude died 5 times in 5 minutes, didnt react to pings, didnt leash.

Im actually so interested in this since what is killing league's comp scene more than you would expect(granted still not much) is that master+ cant duo. So curious to see how premade ques will change, im assuming them combining flex and solo whatever i was called maybe dynamic queue, was kind of part of the first trueskill so i wouldnt be surprised if they tried it again. This does just sound like them saying trust me guys we are fixing it while its only been getting worse for the last 8 years but ig ill have hope in that they were focused on rebuilding over improving.

@MSusername69

5 ай бұрын

How is master+ being unable to duo killing the competitive scene?

@HSE331

5 ай бұрын

@@MSusername69Because he can't duo abuse obviously.

TrueSkill2 is already a working concept in the unofficial ranking system for MMORPGs such as Warcraft Logs/FFXIV Logs. If they make apm/parsing stats strict you should average out in far less matches and find people who can oppose you in lane at your skill level or better sooner.

i'm looking forward to a future where you do your placement games and by the end of the 10 you basically are exactly the rank you should be (whether a brand new player, a first time ranker on a 10 year old account, or a smurf/high elo player) no more being 2 ranks below your target, overinflated and get a solid lose streak, none of that

I suspected as much that they are doing something with the net LP with three ranked splits. Reading the lines, three ranked splits means they are expecting to make more adjustments than previous seasons.

The funniest part is Riot pretending they aren't using EOMM when it's OMEGA clear when you're getting a teammate or enemy that clearly shouldn't be in your lobby.

I quit ranking when I calculated, that id need atleast 65% wr to stay in my Elo, not even climbing. I still had ~60% but ofc it went down the more I grinded.

'Fix' is a strong word. You may not want to tie you reputation to it.

Something to mention about KDA - it has made me a better player and have gained a lot of LP because of it. For a majority of my league career I was low to mid Gold and would often force bad situations and play reckless. I play jungle and main Shy, recently started focusing on getting as much gold as possible and staying alive, playing more selfish and playing for myself. While there was times I got flamed because of not making stupid decisions with my team. The difference has been very noticeable, I've been in what feels like a lot of 1v9 scenarios, and am now emerald 1 almost Diamond without much resistance.

@MrDaffy15

5 ай бұрын

As JGL main one thing really stuck with me - Better get guaranteed gold ( camps ) than path to lane and try to force risky gank because you lose a lot of tempo. You should try to maintain high temp and outmanouver your enemy role. Play consistent, make gameplay every game depending on matchup how you will win the game.

@Z7Games

5 ай бұрын

KDA is an extremely poor way to assess skill as many times a good play involves dying. The easiest example is a toplaner doing some splitpushing. Who cares if the guy is 2/7/1? If the enemy team has to send 2-3 guys to stop the toplaner and they are losing every single objective for it, that toplaner is hard carrying.

@Skryzer

5 ай бұрын

Here's the thing. You are totally correct about maximizing gold income actually improving your chances of winning. But the thing is there will be cases of you actually needing to die to maximize said gold income. Huge wave to collect knowing there will be enemies nearby who will kill you (0/5) but you can keep them from doing something on top of that? That's actually something TheBaus abused on Sion. He practically dribbled the system and won by being a threat to the enemy team without the enemy team gaining anything for stopping him. It's a totally ****ed up way to play the game and it works. You will also see other pro's being in the negative regarding KDA and still consistently winning. Playing with a mindset of earning much gold & xp very often correlates into a positive KDA since most of your gold and xp comes from farming instead of killing others, the biggest misconception practically any player below diamond has. But if you want to go a step further you will sometimes find yourself in places where the best decision to maximize incomes and minimizing enemy gains requires you to actively die. What I am trying to say is: Never indulge yourself in the illusion of KDA mattering when the KDA alone doesn't give you information about player skill. Combined with other numbers how ever, like win rate, elo, gold income, participation in objectives, destroyed towers, cs etc. it can give a lot of hints of how the player is performing and what he's doing right or wrong.

@jarvis2892

5 ай бұрын

People are selfish beings 9/10 times. You will always be flamed when someone wants you to cater to them even when the situation is completely against it. That's exactly why you play for yourself and help only when there is nothing else for you to gain for the team on your own. I main mostly late game mages (syndra, lissandra, lux, taliyah) and i've comeback and often carried so many games that we were behind 15 minutes earlier, and all I remember from most of them is how everyone calls you trash or useless until your champ came online and started sweeping.

@djjazzyjeff1232

5 ай бұрын

@@Z7Games It could be more extreme than that, I've lost many a game to a top Sion that was LITERALLY 1/15.

You know what doesn't feel good? Getting someone genuinely running it down mid from minute 0 for no reason, 6 games in a row, and, while still having a positive win rate, going from gold 4 with neutral gains and losses to silver 2 with +10/-30LP. 6 games. That's all it took for my account to hit a point where it felt pointless for me to ever play ranked again. And every single one of them had someone run into the enemy team's mid turret right from minute 0 for literally no reason at all. Nobody spoke in champ select. Nobody banned champs that someone else was hovering. Those people just decided that they wanted to lose games that day apparently.

@DontKnow-hr5my

5 ай бұрын

I was omw to Diamond in Emerald I with over 60% winrate 5 days ago. Now went down to Emerald 4, to Emerald, Emerald 4, Emerald 3 and now am in Emerald II again. I went from those 60% winrate to 50 and now am at 53%. I had more than one smurf account in my team just going afk or inting because "it is not their acc and idc", countless people first timing their champions and *surprise* feeding on it. It felt absolutely cursed. Just lost another two Games in a row due to people not able to play their champions and feeding.

@d3d952

5 ай бұрын

Same problem here. I was in promos plat 1 I'm a toplaner won my lane hard got inhibs, but there is some asshole playing support illaoi with assassin build or the enemy yone is at 10-0 in 15 min. if I get one game out of 5 with decent people I'm happy, but now every game someone will int from the start or will just throw the game later for fun.

@gillybuzz

3 ай бұрын

Funny that cause about a month ago Riotphreak was getting -10lp and +30lp over multiple games lmao and to top it off he literally never loses more LP than he gains.. ever

instead of trying to define the features that are important for trueskill 2 from just human analysis, riot can just go the machine learning route (which is somewhat alluded to by the 'looking at millions of games' part), and then analyze/manipulate output to pick out the most relevant features. Then iterate on feature weight configurations until arriving at something that performs well enough.

i wonder how split pushing would be counted in the LP gain/loss function. will it be able to calculate the enemies you are going to pull when u try to break a top tower when your team is killing/contesting dragon/baron?

@MementoMoriGrizzly

5 ай бұрын

They can already calculate it, apps like porofessor, league of graphs etc can detect it and show how much damage to turrets you did and how long you were split pushing etc.

New players losing the rank they shouldn't have gotten in the first place means that they're gonna troll quite a few games wasting their time (and fun) as well as all 9 other players. Many people care just about winning but I would rather lose and have a fun game than stomp the enemy in 15 minutes

@gabrieldu3474

5 ай бұрын

Normal game is for fun so huh If ranked exist, it's for the competitive player who just want to win (I don't really know fr since im not touching this hell)

@stawkey9186

5 ай бұрын

@@gabrieldu3474 if you play ranked to just win then you're not playing a game, you're working. Ofc winning is important and no one wants to lose but if it's all that you care about then I'm sorry

@stawkey9186

5 ай бұрын

@@gabrieldu3474 also, sadly there is no point in playing drafts cause you're literally losing the rewards you could get by playing ranked. So unless you want to play singed support or something you're probably gonna be mediocre as you should play ranked cause literally the game encourages you to do so

@SlayNorway

5 ай бұрын

you work towards the goal of improving. if you dont enjoy improving then why even play in the first place? you gain nothing. complete waste of time. @@stawkey9186

@tadashii02

5 ай бұрын

Ranked game, its just grind version of normal game. If u want to have fun u just go play something else. There like no point in playing if u wanna have fun, but if u wanna compete, break ur mental state or ruin ur day for high level of dopamine and lp, cool border. However its not that bad, ranked game its test that u need to pass, in order to ger precios(idk how to spell) lp, But if u wanna pass u really need to improve ur mental, macro, micro. Its really good place to improve and start carry ur team. But its also teach u too be patient, lose will happend, its not really affect much in long run, rank will come its just not time.

2:25 I would love to see it too. Literally every friend, or friend of a friend of mine, have at least one smurf. Riot banning people, while they can just login in another account it's just pointless. Specially when infamous toxic streamers do this with everyone seeing and supporting, I kinda remember Riot talking about how they didn't allowed smurfs many years ago, but I may be wrong since it's clear a lot of people use it. I hate the concept of smurfs, but hey I can't play with my friend because he moved two tiers while playing solo, so I fixed it with a smurf until he got close to me again.

@TheOzumat

5 ай бұрын

I do agree that smurfs are a problem, but I don't agree that they make bans meaningless in League. You have to grind to buy champs and losing years worth of progress IS in fact a punishment.

It's pretty wild how the ranked has worked up until now. My account is 64% winrate, but when I win I gain 19LP and when I lose I lose 22. You would think that with my winrate the system would acknowledge that I should be higher than my rank and boost my LP gains. Then I go watch a streamer on a 50% wr account in my own game gain 35LP for wins, really irritating.

5:30 no, it indeed isn't the brightest people but a large portion are average players e.g gold (you know the ones that gring ranked the most again and again) The game clearly benefits from the vicous cycle of failing promo after promo the average player is in causing them to buy every new shiny skin that is released so they get another hit of dopamine to que for rank again.

As someone who spam created 4 new accounts this is true. my main was bronze and I was out raged. I climbed all the way to gold on my main and started losing 30 and only gaining 12 I made a support ticket and they said this is normal and should end fast. I basically kept playing and de-ranked all the way to bronze. Basically I just made new accounts after and placing them. My first got out silver. My second one got put gold 3 my third one I didn’t like the name and I lost my first game so I scraped it and the 4th one got put in silver but my opponents were all emerald/ diamond I get like 40 plus lp a win

@SevenTheMisgiven

4 ай бұрын

I have played games where I lost 0 lp and would win 52 lp. This was 3v3 but still, the system has always been hugely flawed. A lot of this has to do with the fact that you can be extremely good on some champions and then on another champion you are literally bronze quality. There is so much to league that is hard to define. And hard to rate. Other sports are much further a long the way in that area but 5v5 is also not good enough. :p You need bigger teams where everyone can have their spot.

@vanjamenadzer

4 ай бұрын

This right here is exactly why MMR system is pure garbage. I know what you are talking about as I went through similar things with multiple accounts. It's literally better to spend few bucks, play some ranks, see where you landed and if you don't like the placement, buy another one and try again. Ranking up on a dead mmr account is nothing but masochistic

I dont care that new accounts start in gold/plat. I care that my 12 year old account that I havent played ranked on in ~4 years would have to start in Bronze despite never being below gold since season 2. I dont want new accounts to have to play 100 games to get to where they need to be, I want it so **I DONT** have to play 100 games to get where I should be. I want to grind LESS to get good games, not make other people grind MORE.

I feel like the main factor would be Objective control, followed by Vision (AKA, wards), then CS/PM or just G/PM. KDA, if factored at all directly, would be low on the factor list.

@Slowp0w

5 ай бұрын

I feel like APM is also a valid metric, as long as it is champion bound.

Not gonna lie althought i am excited about this, making trueskill 2 sounds extremely difficult. Like how do you estimate how important shielding amount is compared to gold income/per minute for example in specific numbers? It will be very complicated and i am sure if it stays(hope it will) it will be something that will take a lot of time to balance and even be something that they will keep developing/balancing bbasically every season when looking at it longterm.

@Vafarien

5 ай бұрын

more shields that prevent deaths skill badge + teamfight participation + roaming sheild/vs/in laning sheilding. i think riot already has a solid data set to work with in telling who the winners are, now they just gotta implement it into the role system where it's all tabulate and weighted right, using true skill 2.0. which i think they already have done. they only need to turn it on now.

@ksip3131

5 ай бұрын

It’s simple, they look at trends and the correlation of a particular action/event and how often a team won as a result of that measured action. A higher value or amount of that action performed in a game would yield higher lp gains for that category

@sololeveling7390

5 ай бұрын

@@ksip3131 Still ldont think it is that simple since actions don't exist in a vacum and can be dependant on one another. For example the winning team in a match has more map control so the support has the ability to go ward more easily so that advantage will always be there. So what do you gain extra lp for your actions and skill only if you are winning and can do those things? Say you play an enchanter and need to ward but the enemy is full of a assasigns and your team wont help you ward. Wouldnt you basically not be able to get that extra lp just because of this, especially since warding score counts more for supports? Ofc thes is all solvable with an advanced enough system but i don't know how it would work with an incomplete/imperfect one. Wouldn't it have biases imbedded into it? I would think that such things would take time to balance out and fix. Added note, you would think that even if the system is kind of unbalanced, it is as unbalanced for all players so one game you get the advantage, one the enemy, but what if you are a player that doesnt play a lot. Yould could get straight disadvantaged by such a system for the few games you have time to play without having that big pool of games to balance it out. That is to say the system needs to be balanced even from game to game quite well. This is all just thoughts that might end beeing irrelevant. Very excited though.

@ksip3131

5 ай бұрын

@@sololeveling7390 yes, you’re right that it won’t be simple in the fact that there will be a lot of variables l but I meant the concept of how it would work is simple. To your concern about unbalance, I think that because there are so many variables, they will be able to balance discrepancies that don’t make sense, but in the end it makes sense for players to see an unbalance in their LP gains from scenarios like the one you described due to the opposition denying their ability to perform said actions. Typically this translates directly into wins, therefore should grant LP for those who participated in them. Bottom line, better players will get more LP individually of their teammates as well as opponents. This is the system players have been requesting for so long. A system based on individual merit but also team based actions/results. Finding the balance is key to prevent players from completely ignoring their teammates or solely focusing on doing one thing.

My guess is that it's just a better smurf detection system and won't affect matchmaking unless the player is tripping excessively high trigger warnings. Like going 18-2 with 10 cs/min and getting a performance grade in the top 0.01% of players in that rank. Smurfs and boosted accounts are glaringly obvious, and just accounting for extremely abnormal trends could be a good way to deal with them.

@mtk1808

5 ай бұрын

That's terrible because what if someone starts improving and getting better? Wouldn't he trigger the alarms.if he starts playing like that? It's terrible advice... Just link a phone number and done like they do in overwatch

@anon1963

5 ай бұрын

​@@mtk1808nobody starts playing like that after 1 day, its a gradual increase

@taddeustentakel8598

4 ай бұрын

@@mtk1808 your example is utter nonsense, NEARLY noone, who starts incremently getting better at the game will have the stats as the OP showed, like 18-2 in 10min or so, legit noone, IF he is in fact, getting better steadily.. So its a good approach he mentioned.

This week I went up against a smurf who slapped my team. One guy went afk 5 minutes in and in the next game our midlaner stole the support pick banned my champ went naut mid and had ignite smite. That was just in one day.

I wish they would reach out to Capcom and figure out whatever makes their Mmr so good. It’s amazing. It puts me at my skill level after a game or two and then I actually improve because if I want to get higher skill I need to learn how to use more of the games base mechanics.

@SirDubbs

5 ай бұрын

Different games. The MMR is so good in SF because it’s a 1v1. It’s purely your skill. Team-based games inherently have more variables with teammates

Riot can pretty easily implement trueskill 2 because it's still a similar system to trueskill but with more factors included. In the client we also have the stats page, which most likely will be integrated to the system but in ''real-time''. As someone who works with statistics and those kinds of models, it can either be a great success or the next spaghetti code for riot to work through. I just hope I don't have to play in master's with +15 -32 gains/losses anymore.

@sololeveling7390

5 ай бұрын

I don't think the problem is neccessarily data collection, but data processing. How do you put a number on how much importance one stat is compared to the other? Sure for some it is easy, but for other not so much. There are wayy too many stats that can be accounted for in game and should be if you want a great system. It seems like something that will constantly get changed at the start and another one of those things that they often change for balance. I highly doubt it will be perfect from the start. Im excited tbh cause not matter how good or bad it is at least we know they are trying to improve the game

@tskgamerr8302

5 ай бұрын

@@sololeveling7390 Yeah I agree with you, just the wording might've been off. Riot probably have done analysis on what variables contribute to most wins with other variables added in to not have bias in it. But they can find out pretty quickly if someone finds a way to abuse it. It probably starts off as something like wins contribute to 60% in the system, kda 10%, macro variables 15% and other micro variables 15% and then slowly make those variables percentages higher depending on how they feel is better.

5:22 Yes, making ranked worse makes rito money. Simple, if you think about it from non-fan standpoint. Worse climb, worse matchmaking -> higher chance player would need more play time to get were they want -> this leads to overall increase in play time across the game -> increase play time = higher chance a player might buy something (there is a reason why game companies love statistics like "play time"). Its the same logic casinos use to fool addicts and NEW customers. Give free drinks, free gifts (for higher tier of customer), free entry bank/spin/chips etc, so that "player" spends more and more time "playing". Once they got you, there is higher chance they earn more profit from you. Same thing worse for gacha games. Difference comes from economical standpoint of said customer and how many "gifts"/perks each one get and how much attention from higher ups they receive. At some point, you have your customers (at least lower tier) hooked, so you have to do less for max profit. After a time, you to do bare minimum and allowed to do with them anything you want, because addiction has kicked in :) So again, Necrit, all do your respect, but profits for Rito, gacha devs and casinos are the main focus and they have A LOT of psychological techniques to achieve best outcomes.

@DontKnow-hr5my

5 ай бұрын

Yup isn't the Capitalist thought great? Instead of creating the best product imagineable it is "how can we milk the most money". Like Washing Machine, Keyboards etc breaking down as soon as warranty is over. Capitalism is holding us back now.

@anon1963

5 ай бұрын

True, but league addicts will still cope and say that a profit driven company cares about them and is fair to them lmao

I’m curious to see what they do for people like me… I’ve placed gold 4 once in the 10 years I have played ranked and usually stay at Silver 1. FYI normally I don’t do ranked except the mandatory 10 games to be placed for any free rewards. The problem I encounter is that I play games with my friends who love rank and are of the rank of Emerald, Diamond, etc (no challengers) so I typically play against players in their elo while we are together. The games are tough but I usually do very well unless I’m mid. The problem I have is when I play solo, and I’m still up against those high elo players… and when I play solo ranked, the people who are playing against me as a silver, don’t play like silver players. Their macro is way above silver. I feel like the ranked elo I’m in is for the birds who are just flying by on their way to the rank they should be in. But my teammates are definitely silver players or below. The lack of wave control, ward placements, and map awareness and roaming tells you everything you need to know about my teammates. Because of this experience that I get 100% of the time in ranked. I don’t do it because of the toxicity and seemingly unfair advantage. Like the comment said about badges, im challenger for the dps badge, diamonds for doing 40% of my teams champ damage etc. so im doing well and involved in the journey to the win but I can’t win games when my teammates are in thier elo but my opponents are otw out to much higher elo’s. Sorry this was a long one and yes! I’ve considered that I’m just a silver player but I do have a platinum gem on my boarder and I’m almost diamond (couple hundred points off) so I feel as if I’m not in my elo.

@robertenbre8188

5 ай бұрын

Challenges are nothing when it comes to how well someone actually plays. I commented under the video, saying nearly all stats are irrelevant. What I can see from your comment is that you're clearly disrespecting how complex league is as a game. It's a mindset problem. None of what you've said helps you to learn the game / get better. I absolutely don't mean to be mean just for the fun of it, but it does come off as coping. I've done coaching and I had my own journey aswell before that, I'm just trying to help you realize that you're holding yourself back by focusing on the wrong things.

@TheOzumat

5 ай бұрын

Placements are always tentative and conventional wisdom states that you need about a hundred games before your rank even means anything. There's one thing you can do for yourself and that's just to play more games. When you're settled at a steady 50% win rate, that's when you'll know you've reached your "true" rank.

If i recall correctly the original trueskill system was used in Halo 2 which helped standardize completitve multiplayer on the console for years to come showing that it was possible in the first place.

Also as an emerald player. Any smurf I played on this season was back in the same ELO as my main SO fast.

The more nuance they put into a system, the more people will game the system. Whats stopping people from deciding their midlaner doesn't deserve LP and taking their farm/kills. Game the system by sacking one of your teammates to boost your performance. Or just doing funnel strategies. What about champions like Kayle, who simply try to exist until 26m and may or may not do something before the game ends.

Riot already has a solution. They use it to show how well you did with a champ vs how others did.

@robertenbre8188

5 ай бұрын

No. You can't come to a conclusion of how well someone REALLY has played in a game by looking at raw stats. Maybe you could get a rough idea, but sometimes even that's not true. You would ignore a lot of things if you were to look at raw stats only. That's why people have coaches aswell. Keeping up good general stats isn't enough. Example: I had a duo jg mid, they were the ones with like 80% of the kills and fed af after laning phase (as they were playing Ekko jg Zed mid and taking everything they could) and we were so ahead in gold as a team, it's crazy how a game like that can last 40 minutes. But it did. Why? Cause even though Ekko had like 30 kills at the end of it, he was permanently running after useless kills instead of taking objectives, or his also fed duo Zed died on the sidelane for literally 0 reason instead of backing off, or they forced a fight just after we took baron and traded 1 for 1 on a target that's 5k gold behind them. Looking at their stats they were angry at their team for not playing well, but it was them holding us back from winning in 25, even if everyone else wasn't so ahead. Ekko was taking cs from laners, saying he's the one carrying anyways. The more ahead you are, the more of a liability you become. Yeah your team might have underperformed to differing degree, but who would you rather blame for a loss, someone who goes 800 gold behind their opposing laner on the losing team, or someone who wins by 3k and still doesn't impact enough to win. Raw stats mean nothing. You can say, well uh, but they must have played well and that's why they had good stats. yeah right. Go and try to maintain kda and safely keep high cs and other good stats in games where your only job is to pressure, even if u end up dying for that in the end, and don't get confused of the outcomes. Oh and trust me I have experience, I have been hard stuck with good kda, good cs, good turret damage etcetc, just cause I was only focused on ME winning. It would only hold people back from playing as they should, if we were only looking at ingame stats to determine lp gains.

@PassivesAbseits

5 ай бұрын

Well... the problem with that is, that you can then artificilly increase your stats, without contributing to winning. Specific example: If I go into normal draft, to farm my S-chest with a champ, I go for the most selfish build and playstyle. As Karma Support, that would mean: "Q-Max, Full AP, Full KS, let's go!". But in Ranked, I would start putting points into E earlier and get items, that help my adc... but that will decrease my own stat. Which is totally fine, unless it does decreases my Elo. And you can do that with every support. And don't get me started witht he "Should I die for my ADCs overextension or mistake"? Right now, I sacrifice myself every damn time to keep him alive (ok, twice a game, eventually, it becomes dumb), because that is the correct move for my team. But if that loses me Elo, I will instantly stop doing that and play for my KDA... And there are so many examples. You are a 3-0 Kennen, but your other lanes lost. They really need you to help them in teamfights, but you just push out sidelanes, win a couple of meaningless 1vs1s, finish the game with amazing stats and a loss... instead of at least trying to turn things around with a flanking TP Ult. But that 6-0 300 cs stat line looks great. Or you are the 10-2 Yi, that thinks "What a jucy Red Buff the enemy has, I will steal that... oh shit, FNAtic death bush, I am dead"... The enemy gets baron and 2 inhibs, before you respawn. You have the best stats in the game, but your dumb ego invade still lost your team the game. And that last example makes it so damn impossible to judge players on stats. The game is so complex, that you can be the best player for 25 minutes and still be the main reason your team lost.

@robertenbre8188

5 ай бұрын

@@AnsuTalrei if your personal performance was consistently D4/D3, you would get D4/D3 eventually. raw stats mean nothing. it's just you and everyone disrespecting the complexity of the game to try and cope by blaming others. Drop the ego and be curious, stop limiting yourself.

@robertenbre8188

5 ай бұрын

@@PassivesAbseits exactly.

@beilight6523

5 ай бұрын

You also have to take into the account of winstreak and lose streak. Personally, I've been of 7 game winstreak and gone on 7 game lose streak. Anyone, who enters the mood allows for plays above their skill level. Hell, I have flashed a maphite ult in silver when I was on a 15 game winstreak. Thats what my winstreak did for me. I was suprised myself that I even did it.@@PassivesAbseits

For split 1 of 2023 my mmr gains was +16 -24 and the only way I was able to fix it was literally play 1000 games at a 54% winrate for it to finally get fixed. (Context ended 2022 plat2 started the rank season plat4 with 8 win 2 losses). Even dropped all the way down to gold 4 which normalised my mmr gains but the second plat4 is reached it’s back to +16 -24 (lowest I’ve ever seen was +12 and +13 winning 4 games in a row so +48 only to lose 2 games and losing -50) promoted and demoted 7 times so have the sample size. (My lobbies was always plat3+ so it’s not a “plat rank but in silver lobby meme”) to explain the low gains. I think there is legitimacy in new account claim because existing players should not be expected to play 1000 games to stabilise their mmr so ofc it incentivises making a new account. I know the tagline is “the grind” but having to play with 4 randos is “grind” enough. Clearly the lp issue is bottle necking. I find it strange iron is only bottom 7.5% and they even added Emerald and people still have these issues. Wouldn’t it just make more sense to populate iron and shift everyone’s rank 3 divisions down so skill disparity is evened out for a linear lp grind. 🤔 Understand the necessity of low population in diamond+ but who the fk is arguing iron 4 should remain as bottom 0.54% of players.

funny how riot think that forcing the ranked distribution to be equal across bronze, silver, gold.... is not ARTIFICALLY adjusting their ranks

If they are able to implement a system that can actually detect how well you play, it would be huge for well-meaning players stuck in a game with an int-er and would also more severely punish said inting.

@Wildjesta

5 ай бұрын

I mean, with the badges they can keep track of a TON of your personal stats and achievements. I have one that says I took more of the enemy's jungle than he did (pretty stompy game). It also tracks your ability to dodge skill shots, your successful kills under the enemy turret, and there's even one where you get a badge for not letting the enemy team take a single objective (tower, baron, Harold, dragon).

I was telling people for YEARS now that the player grades after a match are just an experiment to train an algorithm to track important in-game metrics relating to wins, that will eventually be shadow incorporated into the actual game to judge your MMR...except I was sure they would NEVER admit to this but this seems to be exactly what True Skill 2 is going to be The system probably needed hundreds of millions of games to analyze to understand all nuances with accurate enough predictability that it would be fair to implement because, as you said in the video, raw number stats aren't always what actually wins the game...I fully believe they trained the system to look at tens of thousands of hypotheticals based on current game state like "does the 0/3 toplaner now accurately split push and apply pressure implying that their scoreline is not accurately representative of their skill but rather a set of circumstances beyond their control" (for example, enemy being jungler top duo and camping but losing hard bot because of it" And whenever I would post this to Reddit it would either get removed or I would get laughed at or downvoted

@Thekisame27

5 ай бұрын

Lol, I had the same problems on Reddit. Sadly its an echo chamber and its hard to change people's opinion with facts, only with emotions

@CxerRy96

5 ай бұрын

@@Thekisame27 well the problem is reddit represents the broader league community whose average skill rank is like gold IV (now it's plat but only because of the artificial cosmetic inflation from adding new tiers)...they don't have enough nuanced understanding of the game to be able to tell these things, but high elo players and streamers have speculated on this for years

I hope they fix wild Rifts matchmaking too. I don't know if you've been following it, but a hilarious trend has been extremely popular recently, where are people just play inting Sion how to get to Challenger with an 80% win rate. This works because the game mainly looks at kills and deaths when choosing how good you are to pair you off with other good or bad players. Naturally, inting Sion will end games with horrible scores despite winning through taking plates, so the system thinks the player is garbage despite having great game knowledge, and they great paired with borderline pros.

I spotted you yesterday on Guzu's Raid in Season of discovery Good luck with the loot mate !

It would be interesting if they had an AI that evaluated whether the person played like people of their rank or played like someone with a higher rank and used that to give/receive LP. But this is probably too complex to apply on a large scale and evaluate all games.

@guigs4467

5 ай бұрын

Nah, I can't see a reason why. How would you model this? How could a NN be more accurate than TS2? What metrics would be fed to the input layer? How many parameters would the model have? and so on. It would be a massive R&D expense for little benefit.

@MohammedAli-pj1it

5 ай бұрын

thats basically what hes saying at the end. All that stats munmo jumbo about bayesian statistics and intuiive ransom process checked against data and refined? thats just what everyone else calls machine learning

@anon1963

5 ай бұрын

​@@guigs4467you arent a ml engineer are you?

@AzureFides

5 ай бұрын

@@guigs4467 AI is really good at notice patterns that human can't. Like google's AI can tell the gender by just look at a scanned image of an eyeball. Also in term of inputs, video game is perfect for that as it can access to raw data right away without need of human to process and digitize all the information for it. And the thing is AI isn't that expensive anymore because it's self learning. Actually it probably much more cost efficient than hiring a team to build something like TS2. Also in a long run it can automate evaluate itself and adjust to new meta.

@guigs4467

5 ай бұрын

@@MohammedAli-pj1it That's absolutely not what machine learning is. Machine learning is a catch-all term for Neural Network weight refinement.

I have a huge problem with the grade system after a match, people that plays toward KDA (and thus more likely to get an S) has to purposefully not play together and wait for something to happen, may it be a partner intiating (and likely dying in the process) or the enemy making a mistake. This stalls the game and takes a bunch of the fun away, but most importantly, goes agains the spirit of "lets work together" when a dangerous engage occurs and you don't see the worth of trying to help the teammate :(

@robertenbre8188

5 ай бұрын

This. This is why looking at stats is utter bs, and if they make a ranked system around raw stats, i hope it kills league. I love the game for what it is and how complex it is, so if Riot did that, they'd deserve to fall.

@daedalus6993

5 ай бұрын

Clearly you don’t get many “S” ratings because if you did, you’d know that KDA has almost nothing to do with it.

@PixiiBomb

4 ай бұрын

I actually think the opposite. I still get tons of As and S's on Support characters who do not play for KDA. Regardless of the role, 90% of the time, I agree with the grade I was given at the end of the match. A's and S's, I'm always like: yea, I had a great game. I feel like I played well. When I get a B, I'm like: yea, that wasn't my best game. I don't play for KDA on Support or Tank, and still feel like the grade was correct, and still get A's and S's. And when I play ADC or AP, I also agree with the grade at the end of the match.

@ilbgentyl

4 ай бұрын

Huge issue is shut down goals. You can’t be as agro or you get the team back in the game

@Fanofmusic106

4 ай бұрын

Yea gotta say kda does not effect s ratings at all and idk how you dont know that lol. Ive seen so many supports have a kda of like 0 5 15 for ex. And get an s. Riot actually has a pretty fair system in place for grading. Im a jg man and have gone 10 1 15 and still didnt get s due to not having as much cs as i should have.

This split I got hard stuck with 62% wr for close to a hundred game, were I kept getting win streaks followed by loss streaks and going nowhere. lp and winrate stayed more or less the same. So I just started trolling and first picking in positions I never played before since the grind doesn't matter.

@penguinmonk7661

5 ай бұрын

Sounds like you got hit by some EOMM bs

I speculate that Riots' current system will take into account your past games, likely the most recent ones(for the benefit of the doubt), then go through its stat checker at the end of a game to determine how much you contributed to the game then it will match you based on how well you have done previously. If you played really well, you will be matched against tougher opponents with average teammates. If you play badly then you will be much more likely to get easy games. This is why you see 1 or 2 good players on a team and typically 3-4 mid - or lower players in dimond.

It's crazy that no one mentioned that Riot already has a system that tracks your progress on a given champion vs players of other ranks on the same champion. This is also how they determine your letter ranking(A, B, C, etc.) of a given game performance. If they implemented Trueskill2 or their own MMR calculation, they could simply use that letter ranking data to feed the Trueskill2 algorithm.

@mastaaceexclusive

5 ай бұрын

Ive never heard of this. What?

@mwx5314

5 ай бұрын

the letter ranking system is also TRASH lols.. you dont know how to EXACTLY get A, S, or whatever letter..

@robertenbre8188

5 ай бұрын

@@mwx5314 you do know how to. you can go to profile -> stats, and see what they rank you on. they probably just average all of those for you for a single game, to give you a rating. but it's all irrelevant anyways so...

@mwx5314

5 ай бұрын

@@robertenbre8188 exactly what you said.. PROBABLY.. you are not sure.. hence.. my point STANDS..

@robertenbre8188

5 ай бұрын

@@mwx5314 okay? there was no point anyways, cause it's all irrelevant. I just said probably because yeah, no one knows what else of the stats they track use for those grades, but one can assume they use the ones they already grade u on, under stats.... but still, raw stats are all irrelevant. You're trying hard to win an argument when there's no argument.

should be a rank by role I think It makes the most sense! then kda system can be apply to adc and junglers where supports could be like warding assisit etc it makes much more sense this way instead being called solo duo should be rank mid rank top rank jungle rank support rank adc then your rank goes by other adcs other tops mids junglers supports that makes the most sense!

@julianramos9368

5 ай бұрын

I'd think you'd also have to implement the champion in role like they already rank you in the ranked tab since champs like Shen shouldn't really be compared to like fiora for example

@MementoMoriGrizzly

5 ай бұрын

Imagine being diamond in adc and mid and getting autofilled as support and having to play in silver or gold. Complete waste of time.

They also need to fix how I'm stuck in plat 3-4 because i keep getting matches with high emerald and diamond. I can't queue with my emerald 1 friend but I can get matches against diamond 1 people? how does that work?

What about the grades like S+, dont those depend on performance in per the lane/role in comparison to everyone else?

@SkittleBombs

5 ай бұрын

yes, its basically going to be if you get an S in your game but lost, you will probably get like 2-5 LP loss mitigation or something.

The problem with ranked LoL has always been, and will always be, Riot's focus on Solo Queue as the premier play mode. It's a 5v5 team game, and they've completely ignored the ranked 5v5 game mode. Until the focus somehow shifts, which is highly doubtful after 12 years, solo queue will always be a toxic hell hole.

@laurelkeeper

5 ай бұрын

There just aren't many people willing/able to get 5-man groups together. They'd be stiffing roughly 90% of their ranked playerbase, minimum.

@Thergood

5 ай бұрын

@@laurelkeeper The ship has most definitely sailed there. No reason they couldn’t give a shot at making 5v5 useable in parallel. Do both. Maybe at some point it would pick up steam. Things like making a “best of” 5v5 queue where teams play best of 2 or 3 games. Or have some kind of single elimination or double elimination tournaments like the old ESL Go4LoL’s back in the day, but automated and kicking off a certain times. I don’t know, something.

@laurelkeeper

5 ай бұрын

@@Thergood isn’t that just Clash?

I played last season ranked and was placed iron after my placements. I was being paired up against platinum players and both win and loss about 20 lp each game. I quit ranked after that because I don't understand how to get out of iron if I have to win games against plat players

The key to overcoming the lp skew to where if you lose you lose more lp than if you win, is to employ strategies that result in CONSISTENT winning. Yes some games are going to be out of your control but if you are breaching platinum/emerald and moving into diamond, you need to know how to consistently perform. In high elo the skill floor is higher. So even the worst player on the team does fundamental things relatively well consistently. If you can't do that, you don't belong in that tier. Overcoming that hurdle is HARD. It really is a good way to force people to ACTUALLY play better to climb.

@vanjamenadzer

4 ай бұрын

You sound like Rioter.

I quit ranked way back in season 6, after failing to move up into gold one too many times. I've never looked back since and the air has been that much fresher for it.

I'm very surprised to hear no mention of Wild Rift's ranked system in this video😮 It already has the new ranked system, and is a great place to study what LOL pc is about to get. I personally love the ranked system on Wild Rift, it's a huge improvement from pc for sure!

@johncarlfuentebella9375

5 ай бұрын

Same here! Even as a support main I always get MVP for playing good.

@johnronaldo00

5 ай бұрын

Lol what you talking about im diamond and it feels like I'm playing still with silver players.

@Angkarpadevat

5 ай бұрын

​@@johnronaldo00they just want to get the rank to flex it and are not interested in actually improving. if they were they'd just deal with the current system which worked for 15 years and was fine

@JAYNEX

5 ай бұрын

does it really has the TrueSkill2 system? i think not, cause that will be completly new to LoL no? You mean the "upgraded" one that is coming with the begin of new season?

@MillenniumEarl014

5 ай бұрын

Lol Wild Rift matchmaking is garbage having bad stats is rewarding because the game will put better teammates for you.

As someone exclusive play tanks and enchanters. I tank a gazillion damage and peel for my carries like a parent protecting their young still gets negative LP because my carries suck so much.

6:58 the insertion point is dependent on MMR and if you are in smurf queue... smurf queue is a thing because there are a bunch of level 10-30s playing at a emerald+ level... instead of playing with a bunch of level 700s with emerald and diamond borders..... if you are in smurf queue you will see level 15s with 188cs at 18 minutes... or players with 30+ kills...and lots and lots of master yi's... idk why but its a popular champ to smurf on.

Wild Rift seems to have a heavily KDA influenced matchmaking system. Obviously it's not confirmed, no one outside Riot actually has the algorithm. But the subreddit is flooded with people who do believe it and others showing themselves climbing to Masters with 100 games or similar by basically running the Baufss Inting Sion strategy. The theory being that Wild Rift overvalues KDA in matchmaking so by getting your KDA down to 0.3 or something, purely focusing on towers, etc. the games thikns you're a poor player and pairs you into a team of higher skill team mates so you are more likely ot get carried on the rest of the map while you split push. True or not? I dunno. But there's several examples of people using that strategy and getting the positive result they expect. I don't have any evidence for the otherside though, confirmation bias could be that only the people who are successful at Inting for towers are posting their rapid climbs while the people who are failing to Int to the top (or who just get bored and swap back to playing normally) are not posting.

@TheCruelWarden

5 ай бұрын

can confirm.. i tried it and the inting Sion worked several times better than playing normally regardless of the role you play (played top, jg, and supp)

@MillenniumEarl014

5 ай бұрын

Because it is true. It has been proven time and time again. Riot should stop adding stupid gimmick to their matchmaking. Make the matchmaking purely about rank and make it as random as they can.

I'm personally hype af for this change so many games I go 20-30+ kills and have 85% kp and we still lose because top decides to afk or some random league shenanigans

i had a winrate from 80% in silver (i had a long pause and was usually around high plat low dia) and couldnt climb. i gained 10-15 lp and lost 30-40 lp. i needed a winstreak of 30 games to climb out of gold. after i reached plat i gained around 10 lp and lost around 45 lp, i lterally couldnt climb any longer. i needed a 85% winrate to even gain lp and one or 2 games with people that tilt and run it down or leavern/afkler and 5 wins are gone. stopped playing ranked cause of that, made a smurf and got easy into dia with it. i just dont like smurfing, cause all my shit is on my main account.

Friend and I were stuck in high-low gold, both of us bought a 1$ lvl 30 account and now we play consistently in mid plat without any issue.

Double edge sword. It would be cool to be placed on your actual ELO as quick as possible according to your skill, but this could lead to an increase of boosted accounts, because boosting could be done faster. Now imagine arrive to diamond+ games with silver players who bought their accounts, which actually happens now but is not that common.

@mendronesg

5 ай бұрын

they wouldn't last much, though. It's only logical that they let the boosted players sink back to low ELOs if they legit play bad. I think this is the phylosophy behind the new ranked system

@x0Fang0x

5 ай бұрын

Silver players who bought diamond acounts will just be in silver in just 5 matches😂😂😂(exaggeration)

I wish riot would focus on how to get new players and old players back insted of focusing on ranked still

@anzullic

5 ай бұрын

What did Riot do with ranked?

@Kalesvol

5 ай бұрын

What's the point of getting new players, who aren't even guaranteed to keep playing, if you can't keep your current players happy? On top of that, a lot of old players quit due to not being satisfied with the ranked experience.

@RemixedYoshi

5 ай бұрын

@Kalesvol I quit a few years ago when they stopped doing new game modes and stuff aram and normal summoner rift was not cutting it chief a majority of the league player base was casual people not ranked back in the day I started league in 2013 and dropped it in 2019

@Kalesvol

5 ай бұрын

@@RemixedYoshi Those modes were dropped BECAUSE NO ONE PLAYED THEM. Its the reason why URF keeps coming back. Its the reason why Arena returned. It's why ARAM is a permanent mode. People actually play these game modes. Why would they intentionally drop modes that people actually played? You act as if Riot wants to lose money.

Only reason I got a smurf account is cause leagues loot progression after level 30 is so dry it was faster to make a new account and get new skins than to level up from 200 to 201.

I have a feeling it’s about how many clicks you do. CS, damage, damage to incoming damage ratio, wards, damage on objectives. That’s how I would make it

6:49 u just need to have a big win rate in normal game before start rank with a new account and u get plat/emerald

Hopefully this matchmaking system works out and is also applied to Wild Rift. Right now, we have confirmed that the system in WR is heavily KDA based. In essence, the better KDA you have, the better you play, the more you are punished by getting potatoes in your team. The worse you play, the lower your KDA, the lower your overall stats, you will get better teammates. There are folks with 99% win rates because of the current system. These guys have 0.2 KDA.

@omargtz2602

4 ай бұрын

The Wild Rift comment I was looking for.

@mstrikerwildrift

4 ай бұрын

They use WR as a test case. The reason why the matchmaking in WR is shitty, is because they are using the "changes" they are talking about doing for PC

@omargtz2602

4 ай бұрын

@@mstrikerwildrift aaaah hell nah, even now the system they have in LoL is better than WR. If the direction is to change it to something similar the fan will definitely be hit.

According to the details of Trueskill 2 you mentioned Supports sound screwed. The only two things beneficial for them are Teamfight contribution and Wards. Other then that supports are easily outshined by their team members or are heavily relient on them resulting in being at a bigger disadvantage when allies troll or otherwise sabotage.

@TheMastertbc

5 ай бұрын

they would obviously be compared to other supports you dont compare a support to an adc

@taddeustentakel8598

4 ай бұрын

@@TheMastertbci hope so :S

Isnt the simplest explanation that they will also take in count the grade you get after games? (e.g S+ on a support is given according to KDA + Vision score + CC score). I thought they were just gonna use that as a mostly accurate rating of players

The reddit commenter meant that by intentionally keeping the ranked system bad, incurring negative MMR states on purpose, you keep a player hooked for longer as a consumer of the game, because they get trapped endlessly looping wins and losses for eternity. And thus, if they play the game forever, they'll also be buying skins forever.

I mean I will admit I don't know how the in-game rating system fully works but its there and I always felt getting S+ in a game and winning should give me more lp than my teammate that got C-, same with losing S+ u should lose a lot less lp. It's already in the game so why not use it. Valo does it with combat score.

@AkshayPrasadrocks

5 ай бұрын

The reason they don't use it is because its easy to get an s rank without playing well at all. The letter grade system only exists to help lower elo players feel good about their performance. Not saying this to flame or anything, its a nice system for casual play. It is not relevant when you are actually talking about performance though. For example, say that you are toplane and you get camped, but still do your job perfectly and absorb pressure. You are probably getting a B grade, but you were a huge part of your teams win. Furthermore, this would incentivize people playing supports in lanes, because thats the easiest way to earn s ranks. (Farming on champs that don't normally farm)

@themonkeys96

5 ай бұрын

@@AkshayPrasadrocks This is the biggest factor. You'd end up getting less LP for getting weak sided. Now if they took into account jungle proximity that may change some things. They track how often your jungler was in the area vs. the enemy. They see the enemy was in the area more often, then go a little bit easier on your expecting your numbers to be lower.

@Kendymary

5 ай бұрын

The issue with that is, you don't need to play that well or even help your team to get an S+. You could grief your team with karthus jg, only farm, never gank, ult to ks and still get an S+. It's a flawed system that should never be used for anything but chests at most because it will encourage the KDA players, not the proactive players that are actualy trying to win the game so they take on more risks.

@danranger2821

5 ай бұрын

@@Kendymary then surely the correct path would be to improve a system ppl are already familiar with and make it more transparent on what constitutes an S+, for example in the stats tab where u can see a pentagon of metrics for how well u played a champs is a good way to start, that way an afk earths jg that is passive and only presses ult to steal won't get an S+

@viperdragon35lol6

5 ай бұрын

@@Kendymary i mean if the karthus ks for kda and still loses game he still loses lp the kda might reduce the lp loss by like 2-3 lp or something

Isn't predicting how good or bad we will be creating winner's and loser's queue ?

The most obvious fix to the issues the reddit post presented: Add an option to "Fresh start" on a main account. the main issue I see, myself included, is that an account that has 5+ seasons of ranked play takes 5 times longer to rank up than a fresh account (same person playing). I can hop on a new account, play 20 games and be in emerald games and stay there with over 50-60% winrate. but main account with 68% winrate takes over 50 games to get the same.

The statement about smurf detection is just so objectively false, it hurts. High MMR players have basically solved the concept of MMR control and will make themselves maintain their rank indefinitely. What needs to happen is if an account is detected as a smurf, ranked queues are COMPLETELY disabled. this would also solve the massive issue of ELO boosting as well.

You know what would be amazing for ranked games, if reported players (toxic ones) win less LP to teach them their behavior is not ok, while the rest of the team gets normal LP.

@SkittleBombs

5 ай бұрын

the only one that will get mitigation is leavers /AFK. but if they can sense a character underperforming to the level of basically being AFK/Inting that will be amazing

@LuxiBelle

5 ай бұрын

Then everyone reports the worst player (it's always you)

@gillybuzz

3 ай бұрын

If you didn't honour me, I will report you

I feel like most people have an issue like this. I myself am in Silver but in Draft queue I go against Emeralds and Plats constantly (occasionally Diamond). In my ranked games it’s not as high but I get in Platinum games still 😭

@MementoMoriGrizzly

5 ай бұрын

Draft =/= Ranked. Me and my friends are diamond-masters and just play drunk or just to chat with a friend who is low elo in draft. Nobody takes it seriously there besides some sweaty players here and there. Trust me if those emeralds and plats actually gave a shit you wouldnt be able to lane.

1) as Alois says, low deaths win games. No matter what role you play, if you don't die, you win much more. So playing for KDA essentially does make you a better performing player, since deaths give away so much gold on shutdowns. For this reason alone, i think introducing a KDA variable to MMR would only push people to play the game as it should be (considering the existence of 1k gold shutdowns). 2) nobody said it has to be based strictly on KDA - first of all, there are all kinds of stats. Damage done, Healing done, Damage taken, CC done, Vision, etc. With these, there could be an algorithm which evaluates each stats importance for EACH ROLE and then compares your performance to the average of your rank, with 90% emphasis on the stats which win games on your lane. Even through all this, you still can't do much as an enchanter with a shitty adc - except maybe not playing an enchanter KEKW

This won't do anything. If everyone sucks, it's the teacher (here, Riot) who's to blame. League needs a complete UI overhaul and A LOT of QoL to make the game more accessible while keeping the skill gap intact

People will always find cope as to why they are hardstuck low elo, new ranking systems wont change that.

I still do believe that they should give LP to a player who did a lot in a game even if they lost, I'm a toplaner currently plat1 I have been losing my promos when I was winning my lane hard and got towers, inhibs and everything but when a mid laner or adc gets fed I'm done for even if I try everything its ass and I lose LP for basically winning lane and helping other lanes.

Will the changes be moved to WR too? I play it cuz i have only 1 PC but I wanna play w my brother so we chose WR; we should be high diamond/master (stuck at plat) but the matchmaking is kinda bad so i was wandering if there are some things planned

In flex I play with Emeralds and Plats as a "bronze" player as thats the rank my friends are at. Every game I get less LP then they do despite the fact that in some games I literally win lane hard and have a much higher gold income against an actual emerald 2 solo queue player. How does that make sense?

I’ve experienced this negative LP before for a few months which I couldn’t get out of, unless I duo’d with someone of higher rank. I had a 70%WR in bronze 1 and couldn’t make it to silver cause I was gaining +10-15 and loosing -30. This happened cause I’d play a few games quite late at night, run into a few Smurf, get smashed then stop playing for a week until I had time off work again 😅. Rinse and repeat for 2-3 weeks and voila negative LP for months 😂.

For a while in emerald i was +14 , - 31. I’m the same rank now with +/- 25, but I just couldn’t be bothered to play the game for a hwile because of it. I took like a month break and the mmr was fixed when i came back and i dont understand why

Trueskill 2 just sounds like the newest league music band

this could potentially destroy climbing the ladder and elo boosting you just stay at your rank and thats it