241 thoughts on “Insider Talks – Skill MM Once Again”

1. soviets are scared of something.

   • They are scared of losing customers if skill MM is the next fluke. You would be careful too, if your living depended on it.

     • Why the heck don’t they make this a choice? Let players chose a skill or RNG MM server depending on their mood, preference or whatever.

       • because people are dumb and don’t know what they want. specifically, people who think skill MM would make randoms better haven’t actually thought out what that would mean.

         • People in California were given direct democracy and voted to shoot down property taxes. Then the state went broke, the schools were going down the drain and finally those dumbfucks understood the concept of balancing revenues.

           • Wow, wrong user is wrong. California passed Prop 13 in 1978. While the state certainly has problems and elements of Prop 13 contributed – we certainly did not “shoot down” property taxes. I still pay a little over 1% of the assessed value of my home per anum. A link to help you learn.

             http://en.wikipedia.org/wiki/California_Proposition_13_(1978)

     • They should avoid the concept alltogether. Skill MM will make this game boring.

       All this skill MM is just a Whale Whine.

       • No, it will make it thrilling because you will get more matches that are actually challenging, and not a roflstomp fest.

         • I don’t understand why this is an issue, you want “skill equal” matches and +-1 mm just play clan wars and shut up about it, Random matches are just that random.

         • how hard would it be to have normal MM but each game has to be between 45%-55% per xvm. this makes all the games more fair so u dont have 60%-90% massacres. i love when xvm had all my matches in a fair range. If a match is 70% then MM has to take out a good player and replace them with a bad player in a similar tank, or swap a good and bad player like in high school dodge ball, to balance the teams. I dont see this as being that hard.

           • I don’t know if you read the whole thing, they don’t like XVM and consider its prediction garbage. What you’re suggesting causes MM chokes and longer wait times for players involved.
             Splitting regular and ladder MM isn’t a solution either, because who’s willing to take a constant 5 minutes queue (consequence of stricter rules for MM + lower population avaiable) for a match that lasts 10 minutes? It’s not like DOTA2 where queue can be long but matches last a lot, here we’re talking about spending half your playing session sitting in queues.

2. Sounds like waking up a Hornet’s nest. If it ain’t broke don’t fix it.

   • Yes, I completely agree. At least they are aware of the fact that demands of a small group are rarely good for everyone else.

     • is it my fault that i like my loltractor? that i play it 30 times a day?
       i mean yes from the point of view of the comunity i am a “stats padder”.
       but to be honest i would love to hit a check mark in the setting that:
       1. makes it so i can get into higher tier match’s fully understanding that it will be supper chanaging will little chance of penning the enamy

       2.if someone TK’s me then they get banned or at least turn blue (because the formula for that is tied to heath so 130 HP to a high tier is nothing)

       3. no more then a tier 3 match normally.

       • all those “what is skill” consideration are a bit disappointing.
         Just simply take win ratio in the account.
         Or more precisely – win ratio from random battles without companies and clan wars.

         You statpaded you WR by platooning? You will be a bit punished in solo game (system overrrates your skill) and your bonus from having coordinated platoonmates will be lower so in the long run what you get in platooning you will lose in solo battles.

         They seriously consider league system? why?

         The same as with +/-1 MM there is no point introducing 1 to 1 skill balance – but some time ago tier 5 could meet tier 10 and it was considered bad, so “sometimes up sometimes down” has to be toned down from time to time

         This is however another case – better analogy would be the matchmaking again.
         In MM there are rules how to balance both team: no more then 10% difference on balance weight for both teams. There are other but that 1 is the only important one.

         So add the wr together and check if there is more then 10% disproportion. If yes, look for the tanks with the same balance weight (so you can swap them freely) find best player from weaker team, swap with worse player from better team and check again.

         Repeat it up to soime number of times (lets say 3) if it still not solved, start the game as it is.

         Even perfect 1 to 1 skill balance will not lead to every battle being on knife edge. But system I describe will partially reduce roflstomps which are frustrating for losing side and not really rewarding for winners and it is not making MM harder to work, it only requires a second or so of swapping pairs of players after MM already did it’s task.

         • WR is not a good index and it would self-deregulate quickly if introduced. And, value differences are minute compared to skill differences, so even now it would be very inaccurate.

           • Unfortunately the developers consider XVM WN formulas deeply flawed, specially the vaunted “Win Probability Prediction.” So if they create a skill based MM, it will be on their rules and those will differ from XVM. I predict an ocean of whines when a lot of blue and purples suddenly find themselves yellow or orange.

             • When I did an analysis of 2,000+ solo pub battles at different skill levels, I found XVM’s predicted chance of win to be very accurate for average and slightly above average players(players with a WN8 of 1000-1500).

               XVM chance of win greater than 55% and Loss: 9.36%
               XVM chance of win less than 45% and Won: 8.09%
               XVM chance of win greater than 60% and Loss: 3.50%
               XVM chance of win less than 40% and Won: 3.40%

               Looks pretty accurate to me. I’d like to see any conflicting data with a larger data set for average players.

     • Thnx for the insights, good to see that they are looking at this from a lot of angles.
       I also like the idea with the focus-stats-group instead of asking for players opinions in forums or polls.

     • It doesnt have to be this massive change, all it needs is simply 47% and or 2k games in a completely diff group.

       Also just like we should keep the RNG but use hard caps [Like 1 Assault TD`s 2 Camo td`s at most] this would work.

       So no more than 2 platoons on either side.
       No more than 2 60% w/r guys
       No more 2 48% w/r guys

       Then you match it like this….

       For every 49% the team also gets a “green” slot minimum reserved 51%-54%, or if there arent any on it denies the other team a green.

       The object IMO is not to make teams “Even” but rather to make it not a slaughter one way or the other. I am sick of these games with 5 reds + 2 obvious bots and the rest yellows and me. I am also just as sick the other way.

       Goal of the Skill MM is to make the ENTIRE team`s W/R from 49.5% to 52% W/R IMO.

       This means RNG still is fully in effect and the “Ups and Downs” you love is still there but the peaks and velleys ar much less.

       If I get a team with 49% and they get 52% I can stil win with great play, but now I saw a team avg 56% w/r vs our 47% and as you can tell…slaughter.

       Also you could add very cool things, like giving one team T9 Purple player and the other a T10 Green and vary it that way.

       No one wants the exact same team as Great players would technically become not as great statswise from fighting only great players :)

     • Oh really?
       Who is small group here?

       Who is to make such claim. The MM in WOT is the laughing stock of the industry. To have something that does not include player results at all into the consideration decide the game outcome before you even load is hardly a professional programming.

       +-1 MM is bullshit I agree.

       But not to have +- 10 % on player rating as Max spread no matter what rating you use is simply idiotic sorry again wrong just liek with the corridors

   • I really think that some skillbased matchmaking would be good for the game, it shouldn’t be too strict (so that it only tries to match exactly same skilled together), but at least unicorns and tomatoes shouldnt be in the same game…

     That said, i really don’t want WG to implement some bullshit that doesn’t work, because when it comes to balancing sensitive numbers, WG is really horrible at it and they WILL fuck it up.

     Also skillmatchmaking isn’t hard at all when it comes to 1vs1, but in a 15vs15 the factors are quite different and not so easy to balance.

     However they could also just split matchmaking in ranked and normal, so to those that dont want to play ranked can keep it normal.

     • skill based match making will brake the game no if’s then’s or else.

     • Skill-based MM won’t fix anything. It will be more likely it will break a lot more than it would fix.

       Bad players will still see unicorns, and vice versa. And people will still rage.
       And the matchmaker will assemble weirder teams more often than now.

       Current MM is working fine and it should be left as it is.

       • ”Current MM is working fine and it should be left as it is”.

         explain why a true skill based mm will be bad.
         the word you forget is fairness.
         unicum vs unicum, and tomato vs tomato is what i call fair.

         • So getting harder enemies as a reward for playing good and getting worse enemies for sucking at the game is fair? And ppl still are wondering why noone takes the pro-skill-mm guys serious?
           If u play good u get the right to pwn and if u suck u get the “right” to get pwned. Easy as that. And thats what most ppl would call fair: Getting rewarded for you efforts, not punished.

           • This!

             And: Skill-based MM will even out the stats by a large degree – you always get worthy opponents in EVERY match you play, so if you are a unicum and get matched against a unicum in EVERY match then all your stats are going to drop to “a little above average” levels!

         • How do you measure “true skill” ?

         • unicum vs unicum battles would be a completly mirror matches
           cause all players of that level know the key points on each map so we end having a bounce of tanks in the same spot cause it’s key
           and then it will end up in match where both teams have there strong spots on the map and it will be become a match of whoever blinks first it dies

           and Last
           those tomatoes that you are accuse of are the main resource of dmg xp and credits cause most unicums play certain classes of tanks
           but it’s those tomatoes that you will find in maus/e100/or whatever and if you catch them by surprise you can even milk out all of the HP from the them(td’s are the favorite meal of unicums your track them and they become a credits and free XP piniata for you)
           that strictly MM should be avoid it will kill the game and the fun hey dude look at me i did 8k dmg in a full unicum battle the dmg will be spread and the player will start to lose there stats cause of that

       • No ,it isn’t working fine. When you have 6 unicums on one team and 6 tomatoes on the other, why can’t they program it to where it’s 3 unicums and 3 tomatoes on each side, to put a stop to the one sided ROFLstomps we have today almost every game? This is what is meant when speaking of skill based matchmaking. It boils down to MM deciding games, not the players, their skill or lack thereof, because the outcome is predetermined when you stack teams that way.

         • i agree. to me skill based MM is xvm rating the games between 45%-55%. in high school dodgeball, did u put all the good players on one team? no, u balance the teams with good and bad players.

           • Yes exactly why is everyone thinking bloody purple only teams ?
             You match it 4 purple vs 4 purple and same mix of colors for rest of the team.
             You can expect only a lot of green and yellow mixes since there are a lot of those players.
             And also you can introduce a spread balancer so over time you will be the dummy in the team and Top gun of the team the same amount of times.

         • well, but , by reading what ppl are writing about skill-MM etc., the main problem isnt the lack of a skill mm but that the gap between a good and a bad player is way to big and therefore we dont need a solution that just hides the problem(and creates many many new ones like personal skill becoming irrelevant) but a solution that solves the reasons. Speaking of that what ppl really want is a good tutorial and more informations integrated into the game to teach new players how the different systems(Spotting, penetration, moduledmg etc) are working.

           If u want to improve ingame climate, u dont have to hide the problems and/or punish good players for being good/reward bad players for being bad, but to educate bad players and give em reasons to improve (Like reward good play with even more exp/credits or punish bad playing with less exp/credits).

3. ”there are several IS-6 players with 92% win ratio on these tanks, mostly because they spend big on the game, shooting APCR 99% of the time, will suffer if we pair them on their “equal peers”, and we’re going to take the blame for listening to people that “doesn’t know what they want but want it anyways”

   just what i thought. unicorns are THE PLAGUE of this game.

   • this example is very extreme case. Players with stats and playstyle like are very rare. Most of the unicum players just play well, while not shooting gold ammo all the time. Calling 0.0001 percent of the players “THE PLAGUE” is pretty clueless.

     • if ur 90% in an is6 ur a damn good player. spamming gold has very little to do with it. many players spam gold and arent over 50%. this is where WG does not seem to understand the game very well.

       • Especially since gold/prem penetration of 217 is LESS than for instance regular AP pen of IS-3 with is 225.

         So this example, is false.

4. http://i3.kym-cdn.com/photos/images/newsfeed/000/324/256/3af.gif we don’t need skill mm, what we need is the free premium FVs :3

5. Blatant disregard for how XVM and its processes work seems to be a theme. Not surprised, a lot of players get it wrong too.

   • Yeah, the guys who make XVM really should be proud of themselves, I mean, look at all the free advertisment – everywhere! In random matches, in forums, on Blogs or even from internal WG emails!

     • eh i only use xvm so i can see whether the fella i’m fighting with/against are pros or that black, annoying scum sticking in the pot. important since SEA has quite an extreme gap between the few elite, and the mass-produced target-practice tomato whackjob of a professional wanking loser.

       i’d fight EU bots over SEAtards

6. Skill MM is a really bad idea. It will punish good players ad reward bad players. It will keep the average average. It will only increase the queue times and all the people who rage about the mm and team will still be raging about mm and team.

   People forget that 50% if the time the odds are in their favor and 50% they’re not. People only remember the massive losing streaks but noone remembers the massive winning streaks.

   You may think that with skill mm you’ll suddenly be raping face as toptier a lot more or have better teams and stuff but thats not how it works. You’ll still be losing 50% of the time and as a result raging 50% of the time. That’s how online games work.

   • +1.
     I think, only bad players want this, because when they platoon with kindergartenmates they win 20% matches and they don’t want it and they want always someone who is able to win for them.
     On the other hand
     - teams with random distribution of skills – equal chance to win, bad players on daily menu. Easy mode for good players, WTF-happen-i-died-again-screw-this-wg-sh.t-and-haxors for bad ones.
     - bad-vs-bad: randomly moving players and campers shooting each other without taking cover / good-vs-good: attrition war
     I don’t want any of above

     • why don’t you want a good-vs-good match? scared of something?
       low stats incoming, maybe?
       how terrible. you can cry now.

       • Because this matches are not fun. It’s either attrition war or complete roflstomp if one team fails to cover one place on the map and other team can exploit it immediately.
         I agree that WtfE100 going to hill on sandriver assault and staying there beyond rendering distance to nearest enemy is not vital for team (kind of player with 1100wn, 46% winrate)
         Some videos educating players in being effective without using other as meatshields (that contributes to statpadding but is useless for victory) and playing before team dies would be nice – for all tank classes.

   • It’s not about “raping face”, I don’t think you understand the reason people want it. It’s the same reason people are up in arms about bots.

     See, the only thing worse than a good, tried and true tomato is a bot, and even tomatoes don’t like bots right? They drag down your team, do nothing and waste a slot a better player could have been in. Same with skill-MM, to a blurple a red player may as well be a bot a good portion of the time, if you can’t do your tank’s own HP in damage you are a huge liability that has to be made up for by the better players. Skill-based MM would match up equally skilled players so you won’t have that constant necessity to carry every. single. fucking. game.

     Being able to rely on team mates, and not worry about dead weight is what every truly wants, why else would they ban bots? Obviously they won’t ever ban terrible players, at least they could minimise their awfulness on players who are used to playing competitively and/or cooperatively in TCs/CWs/Tournies. Wait times would go up? Same with banning bots. Stats would go down? Good! Means I’m being challenged! Same with tomatoes vs. bots honestly. Less bots means reds have to work for their damage.

   • It’s only a bad idea if used to determine who you will face skill level wise. If it’s used to balance the teams ,so one isn’t stacked with unicums, and the other stacked with tomatoes, then the idea works quite well and gets back to the point of what should win games, skill and a bit of luck.

     Let’s face facts, most games these days are one sided ROFLstomps. It’s actually quite rare to see a close game these days, like back when I started in 8.3, because MM can’t figure out it is stacking the teams.

     • yes, the only skill based MM that is doable is 45%-55% per xvm. all reds or all greens on a team is just stupid.

7. Dev 1= WN8 or WN7 (as much as i hate that spastic formula it pretty much measures skill…)

   If i understood all this coding crap i could probs do a way better job than WG do…i mean, make the game 10v10 or 12v12 (lower mm waiting time) use WN8 rating for people with similar amounts of battles, keep normal mm and name it “unranked game” and name skill mm to “ranked game” and a bunch of other shit which involves the crew skill “common sense”
   +2/-2 mm is fucking retarded as some some tanks have much higher pen than others and Damage and pen are the main factors in the game (main obj is to destroy tanks) some t8s can pen some t10s but not all t8s can pen some t10s…etc…etc…

   “the game is half RNG, half skill”= this quote explains what is wrong with the game (for anyone that has played runescape you will understand how BS rng really is)

   ffs WG get yo shit together (look at Robocraft and how they do mm)

   • “Robocraft MM” HAHA, its bullshit MM

   • I am asking myself whether the real problem WoT suffers from are the devs with their desire to prove that they know better than everybody else. Looks like they have to compensating something.

   • Massive tomato oppinion who doesn’t understand how anything works.

8. this conversation is the proof that the player have small impact in WoT.
   everything else are just controlled stats and characteristics.

   • I’ve seen three comments from you so far in this conversation, and they were all tunnelvisioned drivel. If you cannot understand the complex issues involved in running this game at the level of fun (for players) and profit (for those running it) required for it to stay alive, please don’t pollute the conversation and move on to another game if this one does not fit you. Honestly, the more I play this the more I realize the devs are actually right: the huge majority of players have their head up their arse (happened to me from time to time as well), and fail at reason.

     If I had made a game like this, I’d probably be doing exactly what they are right now, and would NEVER plan the future development of WoT in accordance with the wishes of cretins who can’t think past the next bugger they pick out of their nose while playing.

   • If you’re good you won’t have a 70% winrate… You are just one of 15 players on your team! In any game with so many players it’s hard to be decisive in a game. That’s why skill can be seen by the winrate. You wont make a difference in every round but the games you do make a difference in will be reflected in your ingame winrate.

9. WN8 authors are not happy about spotting damage that is not included.
   I can see many things broken – best way to farm WN8 is to take tier 9-10 medium, do as efficient support as you can, but don’t force anything that requires moderate risk and don’t carry a game as long as someone else tries, unless it is not absolutely necessary (you don’t want to lose map control and game by camping behind TDs, be surrounded in minute, blocked by one TD etc.)
   Problem of light tanks – I took my patton on to middle of hidden village with AMX1390 “nearly unicum”. I ended with 4000 assisted dmg (spot), 1700 dmg and 9 enemies spotted. He ended with no enemies spotted, 500assisted dmg, 400dmg. What exactly he could have done better to beat my 485m view range? Suicide by going over the hill?

   • Well observed. The introduction of t8 lights with the power level they now have is the answer to that, or at least the first step in the right direction. I’m guessing the next will be view range reduction/increase depending on the class, and you can bet the mediums will get hit by that nerfbat at least a bit.

     • quite funny that Developer doesnt like XVM because light tanks are measured on damage done, when its the developers fault in the first place, because spotting damage is not included in the API, so its impossible to calculate it differently.

       • ^100 times this.

         Soviet comrade hates XVM because WN8 is limited by the poor programming of the WOT stats API!

         • :D But SS supports wonder how much he gets from WG these days he starts to sound like Quickybaby and his Praise of Severogorsk map rework….

           • Last warning. If you are here only to post bullshit, I’m going to ban you.

   • he could have seen u were there, know u have a better view range and gun depression, and gone somewhere else.

10. Nobody force players play in a random with skill MM, 2 random modes, skill based with bigger incoming and regular random as is now.

    Today MM is very frustrating in the moment you are a good player because you can lose very easy wins because team is full of tomatos with no brain that leave cap when is the only way to win to move to base and decap… sometimes when team do a stupid movement you can win but usually because a player (skilled player) sacrifice to try cover the gap leaved by noobs.

    True is that AW scare a lot WG because guys that PAY prem accounts in WOT are going to be the first in desert from WOT if they dont improve seriously the game and i dont talk about add more tanks… i talk about skill based MM as option, TRUE alternative game modes, better maps (best balanced for all clases and bigger maps) and of course a fuck way to select WHAT MAP player wants play i am really BORED of play in a row of 10 maps 5 times Himmeldsdorf in a scout for example.

    The problem in WOT apart game lacks is the player base for example yesterday in the sacred valley map encounter mode i was in arty, well, in the hill area with an IS and a JT-88 hide behind the house in the middle of the road to castle… well, i ask for the SEVEN guys covering cap blow up the house because i cant hit enemy with house… you think somebody do a single fuck shot to destroy enemy cover??? NO, 5 minutes and they were unable to destroy a house that only can eat 4 hits… the fun thing is that these IDIOTS that are unable to do something so easy as is destroy a house for arty are the first that cry when rush as lemmings move into an area where nobody cant help them for support and call arty noobs because cant help them.

    You could be worst player or better player BUT in a teamgame a solo fighter unable to help a teammate is a true problem because many times these players made lose battle to the team… guys that to dont receive a shot hide, leave enemy rape teammates and in the end they fail to…. you know better suffer 90% lose of HP and survive than be killed for nothing.

    • Well, the same thing you said about AW was said about WT:GF… guess how it went… ;)

      • Well, i fail in my predictio with WT:GF because the play style used in WT:GF is compared with WOT bad… they want made WT:GF so simulator that in the end finish as shit arcade and as shit simulator… at least WOT style provide a good arcade in general terms fun to play… this is why AW start with a great advantage in this area… apart AW based in modern tanks offer something “new” and the idea of dont use nations and use weapons dealers is interesting… you can go for tanks from diferent nations at same time… or at least looks like this.

        More than leave WOT i see that many prem players can decide leave money for AW and play WOT for free… after years i really dont need use prem account even more i dont plan buy another year prem account in december… and this was before know about AW.

      • WT:GF was also implemented poorly from word go, contributing to the problems it already had.

    • “Nobody force players play in a random with skill MM, 2 random modes, skill based with bigger incoming and regular random as is now.”

      Lol, why skill based MM should get bigger incoming? In some MOBA game, both normal and ranked game have the same reward system.

      • Because you face players in your leage… play skill MM made you see better players ergo you receive more to move in “major” leage.

        • You are wrong, play in skill base MM in other MOBA game only made you see other player have the same skill as you.

          In WOT, you’ll have 2 team have same setup, which mean both team have equally share of unicum, blue, green and red.
          Ex: 2 Uni per team, 3 Blue per team. 5 green per team and 5 red per team.

          If you want a setup like 15 Uni vs 15 Uni or 15 Blue vs 15 blue then the waiting time for those match will be very long (will be bad for both player and WG). Or MM will force a battle that 15 Green vs 15 Blue or 15 Blue vs 15 Uni, then the cry about unfair matchmaking will be more than the current now.

          • same skill MM is stupid. if u put unis agains unis before long many will be red and if u put reds against reds before long many reds will be unis.

            • Because that’s how it totally works in LoL, DotA, WoW, etc.

              /rolleyes

    • This guy… is precious. After reading a lot of moronic comments above, he is the first one that compelled me to login and reply.

      See… he is the kind of retard that expects others to win games for him. He SAT there for 5 minutes waiting for someone to destroy the house and expose themselves, while he could destroy that house with a single arty shell and have 4 minutes and 30 seconds to spare to kill the TDs that now his entire team can fire on. Reason? Beyond obvious that he is a pathetic shithead, he is also playing this as a single player where the team exists only to make him play. Selfish, basically. No wonder he is such a red tomato. Pathetic.

      And that’s why anyone who wants skill MM, just like tons of people above aren’t going to get it. They are stupid, they don’t understand how the game works and what everything entails, and most importantly, they think it will make them be better at this game, when in reality nothing will help them. Nothing.

      Worst part? If they do get it, and it ruins the game, they are not going to be the ones playing and suffering through it. They will move on to something else, leaving the wreck of the game behind. They have no commitment to the game. Fuck them. Cant’ fix stupid.

      Also, this guy is so retarded, he thought WT:GF will be a success. How can anyone take him seriously? And the rest of fucking retards who were yelling “WT:GF is going to destroy this game, whaaaa”. Lol, I am still laughing so hard at them. Fucking pathetic.

    • The release of AW would be great in the following regards:-

      1. More choices of gameplay to mix things up again.

      2. Playing with modern tanks.

      3.a. If its successful, it will take (allegedly) a lot of butthurt gentlemen away from WoT.

      3.b. If its unsuccessful, we get to laugh in the face of the these gentlemen.

    • “Nobody force players play in a random with skill MM, 2 random modes, skill based with bigger incoming and regular random as is now.”

      Really? Why should there be different incomes? I understand you do not see further than you feeling you deserve more creds/CP, but the reasoning is weak.

      “Today MM is very frustrating in the moment you are a good player because you can lose very easy wins because team is full of tomatos with no brain that leave cap when is the only way to win to move to base and decap… sometimes when team do a stupid movement you can win but usually because a player (skilled player) sacrifice to try cover the gap leaved by noobs.”

      And just as often you win for the same reasons. Plus, if you are skilled, these so called tomatoes are where you get the easy damage.

      “True is that AW scare a lot WG because guys that PAY prem accounts in WOT are going to be the first in desert from WOT if they dont improve seriously the game and i dont talk about add more tanks… i talk about skill based MM as option, TRUE alternative game modes, better maps (best balanced for all clases and bigger maps) and of course a fuck way to select WHAT MAP player wants play i am really BORED of play in a row of 10 maps 5 times Himmeldsdorf in a scout for example.”

      Again you have not thought it through. Skill based MM -> everyone trends towards the same win-rate. If two equally matched teams meet all the time then the win-rate will be 50% (ish). So there will after a time be very difficult to separate the good from the bad -> more whine.

      As for the map selection by layer, 15 v 15 scouts on steppes, or 15 v 15 heavies in Himmelsdorf, or 15 v 15 TD on Sand river will be _so_ much fun….

      “The problem in WOT apart game lacks is the player base for example yesterday in the sacred valley map encounter mode i was in arty, well, in the hill area with an IS and a JT-88 hide behind the house in the middle of the road to castle… well, i ask for the SEVEN guys covering cap blow up the house because i cant hit enemy with house… you think somebody do a single fuck shot to destroy enemy cover??? NO, 5 minutes and they were unable to destroy a house that only can eat 4 hits… the fun thing is that these IDIOTS that are unable to do something so easy as is destroy a house for arty are the first that cry when rush as lemmings move into an area where nobody cant help them for support and call arty noobs because cant help them.”

      You are in an arty and cannot destroy a house in 5 minutes? Sure you want skill MM?

      “You could be worst player or better player BUT in a teamgame a solo fighter unable to help a teammate is a true problem because many times these players made lose battle to the team… guys that to dont receive a shot hide, leave enemy rape teammates and in the end they fail to…. you know better suffer 90% lose of HP and survive than be killed for nothing.”

      This goes both ways. Yes, 15:2 roflstomps are no fun for either party, but they tend to even out over time.

11. the game is a lost cause. maybe they will change something, only when they’ll begin to lose players.

    • i think, as do most, that the game is, FANTASTIC!!!!

      I think MM is fair and only needs tweaked a bit.

      My major gripe is arty being in the game, but i can live with them as is.

12. Where is the POLL?
    I want cast my vote!

    • I got it via email. There was a link leading to the portal.

    • some people got it and some people didnt, it came last week for me

13. Omg – a bunch of apes trying to figure out a fusion reactor.

    1) Usually its not the “unicums” that cry for “skill” MM. Looking at the EU-Forums its avg Joe, noobs and bots, because it would make their life easier and they live under the false assumption that skill MM is the solution to steam rolls (15:1 matches). While the true reason for the increasing number of these kind of matches are game design decision from the last ~2 years.

    2) What they are discussing there, is mirror MM it seems. Thats the worst possible kind of skill MM, because in the long run it drags everyone to 49% WR – retarded.
    The only kind of skill MM where I see a chance of being accepted by players in the long run is an opt in/out thingy – you choose if you want to play with skill MM – with some kind of ELO rating and increasing rewards.
    That ofc will never for two reasons – its much more complicated to program than this nonsense mirror MM and because they wont get the balance right between effort and rewards.

    3) They have absolutly no clue how XVM or WN work it seems and judging by the comments regarding ratings and light tanks, they dont even understand how to drive a light tank and win more than 52-53%.

    That aside – you really think people who come up with that retarded WG-Rating, have an idea how to measure skill?

    • Omg – a bunch of apes trying to figure out a fusion reactor.

      HAHAHAHHAHAHA Best comment ever!

    • Unicums wouldn’t since it would make their lives harder.

      Tada.

      • while having no benefits probably.

    • Hmm, never thought about calling it mirror MM for the idea. Might have to steal that name.

14. So they think that a player with 92% win ratio has it just because he shoots APCR all the time? They are kidding.

    • this is the truth. all unicums are shooting gold. thats why they don’t want a real skill based mm, like ELO.
      they fear stats collapse.

      • the tomato has spoken…

        typicall tomato comment:”All ppl with worse stats than mine are bots and all with better must be hackers/cheaters/goldnoobs/statspadders. And if someone somehow got the same stats as me he was just lucky with teams and in reality is one of the above mentioned.”

        • the gold noob is mad. you can go and cry now.
          don’t be ashamed.

          • mhh ur tears taste good. Do I smell some butthurt with this vegie?

          • pubbeh is scared lol

          • The gold ammo is available for credits for one reason… to use it.

            • to use it not to abuse it by idiots.

              • So… if you whine it will prevent them from using the gold ammo? NOPE. They will annoy you even more.

                • i see. this is a gold noobs reunion.

      • Funny, most good players I see are shooting reg, and the scrubs are shooting premium.

        • real question remain.
          why is a uni-CUM scared of ranked/leagues, skilled based mm?

          • Because it will fuck up all the statistics he worked for, starting with win-rate and going down to every other stats.
            Everyone will reach an average “state”.

            And to talk about your thinking that all unicums “spam” gold ammo.Do you have 7 GB free HDD space and some months of time to spare and watch all my replays and point out how many times I, even, used gold ammo?

            • those stats are there because unicums were fighting against weaker players.
              don’t forget that.
              want to play random? fine. let random as it is.
              but at the same time i want a league system/battle mode. everybody is happy.

              • That’s what differenciate them from those “weaker” players they fight, for fucking sake.
                Fuck logic…
                Of course they have better stats…they simply are better then those tomatoes.
                If unicums will fight against unicums and tomatoes against tomatoes you’ll reach an overall average state with nothing to differenciate.

                Edit:I see you added a point.
                Yeah, if it get’s added as a separate mode I won’t mind.

                • He fought against tomatoes and couldn’t beat them, so he joined ‘em…

                  *KurtHetfield
                  Don’t forget that you’re fighting against weaker players everyday, maybe weaker then you. Think before you speak.

15. I think i know which is the main problem here. Those idiots don’t understand the concept of a ranked system. If you have some sort of ladder system, the main point is where you are placed in the ladder; not your winrate. Look at the League of Legends system with different leagues. Besides, in Lol you have normal mode or ranked mode so players are not forced to play something they don’t want.
    The next problem they have to implement this, and i think it is their main concern is the rng and how poorly balanced the game is.

    • they even don’t realise that they are idiots.

    • Bingo!

      All those smart asses here who haven’t really thought about this issue at all should start to stop looking at win rate as the holy grail of player rating.

16. I doubt any of these “developers” are even real. And, if by some long shot they are real, I can guarantee not a single knows dick about statistical analysis.

    • they are real, but in most cases are useless.

    • Oh they are. I know who they are after all and yes, they do work for WG.

17. All they have to do is balance the teams. It doesn’t matter if there are good and bad players together, as long as the teams are somewhat balanced. Not all good on one side and all bad on the other, like it happens very often right now.

    • yep, xvm 45%-55%, with tanks balanced and players balanced. it seems easy to do and ques should not be much longer. some say WG likes current system of more lopsided games to keep most players above 48%. it is very rare for games to end with less than 5 tanks on winning team.

18. Well, I have a nice heretical idea :)

    Remove WR alltogether, just do it, don’t display it anywhere. Don’t rate players by a metric that depends largely on others. I know, it will even out given large number of battles, but short time-scale downward fluctuations will make players unhappy.

    Just display the personal rating (or variant of it). You can have glorious defeats that would increase your rating, not lower it (unlike WR) and vice-versa – on a victory that you had nothing to do with.

    I know, it is just a number, but psychology wins in this case :)

    • some one will make a mod for it :P

      • whats a personal rating?

19. Engine upgrade? Interesting.

    • They will add one more communist horsepower to the “engine” for more fail and whine. In russian upgrade means downgrade… so don’t expect miracles.

    • Engine upgrade=increasing the maximum number of birds that a map can have.

20. I’ve said it 10 times, can repeat one more, because clearly again people have no idea, what they want. And when they will get it, then they will regret, that they wanted it in the first place..

    Skill MM will not work, not in WOT, never!

    Why? Maybe I am the only one with logical thinking, but I think it’s easy as hell:
    Let’s say, WG implements Skill MM based on their Personal Rating (there could be any other rating, it doesn’t matter). So we have the starting situation, when we have unicums, average Joes and tomatoes.

    Scenario 1 – Skill MM with leagues, as is – unicums play against unicums, baddies with baddies, etc. They play against each other for a month, maybe a year and for all of them winrate is starting to level to those magical WG perfect 50%, because in every f**ing game you will encounter the same skill level players as you. And in long term you cannot beat them, because statistically they have the same skill and knowledge as you.
    The same will happen with red players – they will hack and slash or suicide rush one another until the point, where everyone will come to that 50% winrate (and most other stats will equalize too, because as I said, you will have your mirror player against you and behind you, who will do the same exact things as you).
    In this scenario however, there’s one other HUGE flaw, which I haven’t mentioned before. Do you know, why good players are so good? Because they don’t do at all or do stupid things rarely. Unicums usually don’t do suicide rushes, they don’t drive in front of 5 TD’s, and they don’t go alone in one flank, when all other team members go to other.
    And what does it give to our League Skill MM game with good players? Yes, you got it right – enormous amounts of camping. In certain maps and with certain tank setups no one will drive in open field to scout, no one will be the first one to peek out of corner and no one will go to a flank, if there aren’t at least 3 or 4 “trustworthy” team members to count on. Because they have experience and knowledge of what will happen, when you try to be a hero against good players.
    Get it? But oh, you say – what about the other type of Skill MM, the one that balances teams by skill after bunch of random players are picked for a battle.

    So here comes Scenario 2 – Skill MM which puts in one battle good players and tomatoes, but balances them equally for both teams.
    In general it’s the same sh*t only in different bag. The outcome after, let’s say 6 months will be the same – when you always will have 50% chance to win, statistically in long term you will win those 50% of times, because as good will be you as a player, so good will be a player in enemy team. Either you will fight him on battle field or you will fight enemy noobs, and that good player on enemy team will fight your team noobs. And if you think, that you will be the one, who can carry those games, then there will always be the same player on opposite team, who will think and perform the same.. And there we go, stats for all players go or close to same 50% winrate.

    So what does happen with our precious Skill MM after a year? Right, it does not work anymore, because in both scenarios players got their statistics mostly leveled, because they were encountering their mirror players and teams all the time. Only now, statistics show, that a previously red players, which you now encounter and have in your team, now have green winrate and stats, because they could win more games against their tomato mirrors, but it does not show the real situation and the real abilities of that perspective player.

    These are my 50 cents for those, who have not heard of this aspect or think, that Skill MM will reolve all their problems..

    • Have two modes.
      Random – everything stays the same, keep your worthless Battle Rating (# of games = skill, yo!), your precious 67% win rate is untouched.
      Rated – has its own rating and stats, play your way up/down tiers like every other successful rated system in multiplayer games.

      It’s not that fucking difficult.

    • ranked/leagues type is the only way for a mmo.
      look at LoL and how successful it is.

      • LoL is a different game. Maybe you don’t get where the real problem (the one that kills everything else) lies. Simply put: wait times.
        Let’s take LoL as an example: wait time is in the range of 1-2 minutes but may go up to 5-9 minutes for good players. In LoL, this is fine and acceptable, because after that time spent in queue you’ll spend the next half an hour or more playing the actual game.
        Now, let’s analize WoT. In WoT, 5-9 minutes wait time is pure BS. You’re actually spending more time in queue than in game. And let’s not forget, teams are larger in WoT than in LoL and server population may not be able to support it, and players are dispersed over the 10 tiers we have. All of this either stacking up even longer queue times, or forcing the MM to ignore some rules in order to actually get people playing.
        Is this scenario any acceptable for you?

    • Don’t use win rate as skill indicator!

      Omfg, is that so hard to deduct?

      I am getting sick and tired of this bullshit being repeated over and over, because all you wannabe Einsteins cannot think further than the tip of your fucking noses!

      • How should we determine skill then? Personly I look at indivdual tank stat and here I look at average dmg, hit ratio, spotting, xp and then WR (last important) also I look if the players is in a clan or not. If the player is in a clan I rate his stats lower than a solo player. Tank specific stats is whats matter, not overall stats so much. Most times, players that does very good average dmg and xp on tanks and have good hit ratio are in my opinion much better than players who has higher WR but low avr damage in tanks. To win you need to make damage and play active as well as take the right decisions in able to make that damage. So no I dont rate a player as good if he has 60% WR and 1500 avr dmg in a tier 8 heavy, but I do rate a player that has 55% WR and 2k avr dmg in a tier 8 heavy because I know that person has not been carried by platoon mates or anything.

        This is also why I dont think many of these unicum streamers playing tier 10 mediums provide accurate stats because they have half rounds gold = penning objects they would never do with AP.

    • You aren’t the only one thinking that way evi7.
      Skill MM will never work in WoT.

      But, as Bokuden said, let those that don’t know what they want have Skill MM in a separate mode.
      I’ll take my popcorn and laugh my ass off when they’ll all return to the standard mode once they see it’s not working.

      And @Kurt
      Hell, no, MMO exists since forever and only few games, THAT CANNOT WORK WITHOUT SKILL MM, have it implemented.
      Give me a list with all the MMO’s that have skill MM and I’ll tell you why they have it and, on top of that, give you a list with at least a part of MMO’s that don’t have it.

      Look at LoL and the differences between ladder gameplay and unranked matches gameplay.Being the only good player in a team of kids and playing against a team of competent players will leave you with no hair left…and compared to WoT, you waste 40-70 minutes of your life, not 7.

      • Will “never” work? Bollox.

        It will work perfectly and be the best thing since sliced bread. You just don’t get it.

        • Get skill MM implemented in a separate mode and just watch and see.
          I’m already preparing my popcorn to see how all crybabes will scream for a “rollback”

          And what I don’t get?
          I simply don’t agree with your opinion while I have my own opinion which I described lots of time already and I won’t fill another wall of text for that.
          WoT isn’t made for a skill MM and it works perfectly fine without it…if it wasn’t then it would’nt had become what it is today.
          And again, I already described my point of view in several threads on the same topic.

        • Nah you just don’t get it.

          • Time will tell everything ;)

    • Only if you use WR as skill rating. In other cases, it will still work.

      • MaterielDefender and nixxxie
        As I wrote in the original post, it does not matter, if you balance players by winrate, average damage or houses destroyed. In most of the times, you will face an opponent, who will try to keep up and will be able to do the same average damage/exp/whatever other stats like you, and after hundreds or thousends of games, you will slowly level with those players anyway.
        Maybe there will be a margin, which will stay, but in general all players will have much more close stats to one another, than they have now.

        • Nope, with team reballancer WR flattens for everyone but good player still gets his portion of noobs to farm damage/spot/track/whatever else is needed to maintain his rating. Only some statpadders will find that they dropped severely ;) .
          With league scenario it’s of course a different matter and all indexes go bust in due time – unless you add some league weight multiplier, and even they they would be hard to compare.

    • You are correct. By the law of numbers if you play enough games % will even up.

      Just prohibit XVM, ALL mods, ALL modifications, just vanilla game and you will have very interesting scenario.
      Will be very interesting Not to see any ratings, stats, damage done, damage received. Not to see who is Red, who is Purple.

      • why play?

        yes, game is fun, but u play for stats. u play to be good, to get medals, to win games.

        if u remove stats i would not play as often and not play as well nor would i play seriously nor would i spend ANY money on game AT ALL.

        thats why i like the test server. u can play and do stupid stuff and not have it affect win rate or WN.

21. It can take thrill out of this game. Most of “epic” moments came when your team is bad and one have to carry the game. Almost all Ace tankers, Top guns an so happens in those situations when your teammates are unable to kill/make damage/spot and you have to do it all. We are still playing pubs for those moment, not for OK games. For everything else is clan.

    • Don’t worry mate, if the crybabies get their “skill balancer MM” then every match will have “epic” moments.

      If you’re even half good at the game you will be saddled with a bunch of tomatoes for you alone to carry. Don’t worry though, just gives you a chance to have epic moments more often!!!!

      Not my idea of fun.

22. >Developer 4:
    >Why 16-20 matches? PS: i don’t like XVMmod because I’m less than happy on how >light tanks are measured on “skill equals damage dealt” equation

    Srly?

    If i remeber correctly; scout damage is not included in the API data.
    So maybe give us seperated scout and detrack dmg?

    So xvm/WN8 can rate scouts correctly?

    • Spotting damage IS included in the API, but it was added in 8.8…you had nothing before that, thus, you can’t include it in the rating for older players.

23. Please WG make a parallel WoT where there is total skill matching because I want to see the whiney fucks come crawling back because their stats took a nose dive. Mind you all the low ranking players would be over there happy that they were all close to the 50% mark.

    Under Kommissar: Comrade Kommissar, our tank commanders are doing their best but half of them are below average.
    Kommissar: Comrade Under Kommissar, all the tank commanders whose performance is below average clearly lack the will to victory. Send them too the gulag.

    Later………….
    Under Kommissar: Comrade Kommissar, our tank commanders are doing their best but half of them are below average.
    Kommissar: Comrade Under Kommissar, all the tank commanders whose performance is below average clearly lack the will to victory. Send them too the gulag.

    Later………….
    Under Kommissar: Comrade Kommissar, our tank commanders are doing their best but half of them are below average.
    Kommissar: Comrade Under Kommissar, all the tank commanders whose performance is below average clearly lack the will to victory. Send them too the gulag.

    Much later………….
    Under Kommissar: Comrade Kommissar, our only remaining tank commander, who used to be our very best tank commander, now is only average.
    Kommissar: Comrade Under Kommissar, clearly he has lost the will to victory. Send him too the gulag.

    • Haha!

      Makes a perfect point. You only get *good* games by beeing able to exploit the mistakes of the other side. For HT class inpenetrable angling/hugging a spot, for MD getting into flanking positions and for lights getting into the optimal spotting positions. The moves of both sides would be mirrored… If skill lvl where to be equalized in a match then only RNG would be left to decide the outcome… That would be a lot more frustrating then having the occasional *bad* team that is roflstomped.

      I had an avesome not long ago with my AMX 1390 with almost 10 k spotting dmg on Siegfried line standard north spawn. I saw in the beginning that most of the oppenents where fast mediums while our team had many TDs.Thus i anticipated a fast rush on the field and that our TDs would be hugging the ridge close to cap. I went as fast as i could to a bush in E3 area (i think…) and kept quiet. Soon enough the meds started showing up in a more or less orderly fashion and they got blasted into oblivion by my teams TDs that remained unspotted due to the distance.
      Now here is the point. If those med drivers where equally skilled to me they would have known to retreat and not to go up any further when their avantgarde got shot to pieces by unseen TDs. Instead they kept on coming like the soviet army with commisars pointing a gun at their backs and got ripped to shreds one by one untill all their meds & lights (7-9) where gone … My spotting would have been nullified if they understood that their avantgarde of 2-3 tanks that just got shredded had entered a killzone.

      • So what’s the problem? It would mean that you’d actually need to get creative and find ways to solve deadlocks. I find it absolutely pathetic if “better” players claim getting fail teams to roflstomp them would be their birth right. That doesn’t make you a good player. Let’s see how good you really are by duking it out with your like! Actually that would require people to keep improving, instead of relying on RNG to give them enough teams they can roflstomp so that they can keep wanking off over their win rate.

        • Better players who have good crew skill, equipment and perform way over average has in deed the fucking right to kill baddies and stomp weaker teams. What is the point even trying to become better if you can use your advantage? And no, no unicum would have their current stats if they would face better players more often because nobody has gotten their stats this way. Skill MM is a nerf to good players or those wanting to improve, they dont have anything to gain simply.

          Its the average to below average players who should not get everything handed to them easily. If you make alot of average dmg and kills and play agressive you WR will eventualy increase with this current system. WR alone is not enough I agree, but overall stats will vary alot depending on tanks you play and how much you play them. For example 50% of all my games are in tiger II and ferdinand and rest of that are various grinds. Are my overall stats a good indication of how im performing, when I only play one tank and have done it more or less for a year?

          You cant rely on RNG to carry you, everybody get the same RNG. But good performance and the current MM will increase your stats above average, while bad performance and current MM will punish you. With skill MM a tomato can be a 49-50% er because he will most likely meet more bad players and hence level out the WR in a positvie way for him.

        • @MaterielDefender

          You are missing the point it seems. The kind of team you will get is random, if it is bad/good/standard is the same for all which means no matter what stats you have you will still get plenty enough so called *bad teams* on our side where you really need to give 100% + gods of RNG on your side so the team does not get roflstomped or… sacrifice yourself for the win by taking your light and smoke out an enemy spotter from the uber spotting bush or TD platoon hiding behind several bushes or distracting enemy HTs by getting on their behinds and thus giving your HTs a small opening to get in and put in effective shots…. This is also when newcomers or “bad’ players have a chance to watch and learn.That is when your stats improve over the 50% average. Not by roflstomping bad opponents. The maps really are not large enough to do a lot of “creative” manouvers as you called it, There are a limited amount of sensible places to go if you want to have an effect on the outcome of the match. Those places are known to the players that constantly improve their gameplay and of course with that knowledge comes an advantage.

    • GG, Herman, GG
      :))

    • Win rate is not a good skill indicator. Please *think* thoroughly about that, and about possible alternatives, instead of following a beaten path.

      • How about you think about alternatives since you keep bleating on about it?

        • win% and WN r best ways to judge players based in tanks they play per tier. good stats at tier 10 means ur a better player than tier 5 with same stats.

24. 1) Skill based MM is a bad idea. They are right about it.

    Unicorns spend a lot of time with thinling about the game, improving they skills and SPENDING a lot of money on the game – Gold ammo, prem time, free XP, you name it.

    Putting them against each other will see their stats drop. THE ONE THING they have their mind on. No more WN8 padding, no more 60+ Winrate. Welcome to beein average (statswise). You WOULD INSTANTLY generate a shitstorm on the forums.

    2) Not the ~49% win players are the problem, unicorns are. They want the game exactly as they think it would be great, but more often than not blatantly abuse their team to max out their WN8 score, Save a teammate or get the kill after he dies? You know the answer…. And, they are the most vocal on Chat, making the game even more toxic. EFFE/EXNOM anyone?
    So, maybe pooling them up and let them battle it out is not a bad idea after all…?

    3) PLatoons are a problem too. Why not make them fight each other exclusivly? Works on other games, and you dont have to put up as a solo player with 3*batchat platoons

    • 1.Yes, Skill MM will never work in a game where even developers agreed that is half skill half RNG.
      If a player worked hard in order to pwn, then let him pwn FFS.Tomatoes have to deal with it.
      2.I think it’s exactly how developers said…most asses aren’t even sure of what they want.
      3.Then you’ll have to wait, literally, hours to make a team full of platoons(OF THE SAME SKILL on top of that!!!)

    • Your problem is that you consider win rate to be the only valid and useful skill indicator. It is not.

      • He didn’t said anything about win rate.If saying “49%” players instead of “average/bad” players makes you think that he’s refeering only to that simple stat, you’re wrong.

    • Playing to increase your stats, and not to have truly thrilling, challenging matches? How sad.

      • Having only thrilling matches will become tiresome and losing in those matches will make you smash your head on the walls.
        Just play ranked LoL or something close to it and see how “cool” it is.

      • Material: who says playing in current random is not thrilling? Try maintain 1.9k avr dmg in a tiner 8 solo for 4k games and come back and tell me it is not “thrilling”.

    • fuck unicorns. they are the cancer of this game.

      • You’re just mad cuz you’re an average tomato and can’t accept the truth. Jealous bitch… In WoT you’re an easy frag. Deal with it.
        Now, get the fuck out!

      • Damn…such dissapointment…
        I always thought I was a crab, not a cancer…
        :sadface:

25. I actually like the thought process of Developer 8, but only him.
    He doesn’t think a full ladder system is the answer, at least I don’t get that picture and he seems not really hanged up on only the %wr as a main statistic. Which imho should be dumped as it is only a consequential stats that flows from how skilled one actually is.

26. Fuck me.. it sounds like some of the Devs actually have a brain… who’d have thought that???

    Seriously though.. there are many good and varied reasons to NOT implement skilled MM in any shape or form… every single “version” has major flaws which could break the game (game=Playerbase) within weeks of release…

    The current MM, with;-
    +/-2 spread
    varied nations (ie abilities)
    mixed LT/MT/HT/REMFs
    25% RNG (a bit high IMHO.. 15% would be better).

    …is perfectly serviceable… yeah it has brainfarts, like another 5v5 arty REMFS yesterday, but we all have them as well…

    Skilled MM will (more than likely) break the game and kill the Golden Goose…

    @Gkirmathal.. WR% being “only a consequential stats that flows from how skilled one actually is” is the whole damn point of it… winning it the fking aim of the game…

27. WG got it right. The MOBAs with skill MM are hate generators of the Internet. Look at the countless memes spawned from impotent rage of 13yr old LOL kiddies!

28. So main problem of a skill MM is: how to evaluete skill?
    You can imagine any method, win ratio, damage ratio, etc. that will be bad. Lets give you an exemple, I’ve fine the PERFECT formula that give me: this one is better than him by X%, etc.
    So, now, I make a skill MM. What happened. Bad players play with bad players, average with average and good with good. Well. But what will happen? The less bad of the worst will perform better and the less good of the best will perform worse simply because there opponents are now of the same skill… So everyone will tend to be average. So, you will have the exactly same situation as now: everyone in same battles without segregation. Why? Just because every rating formula won’t evaluate skill, but performance. And performance can only reflects skill if everyone have the same reference, and this actual reference is the random players gived by the MM. If you change the reference between players, you will fail to compare them.

    • 2nd problem, bad players will progress faster in the tree and best players will slower. Is it fair?

    • I believe that good players will always achieve above average damage and kill ratios. Remember: Even will skill based MM, there will always only be a few good players in each team. These will have an edge over the less skilled enemy players and thus achieve a good performance.

      • Skill MM will still be a boost to fucking bad players, why is that fair? Why is it even fair to make it harder for good players, its not like their stats have come free.

29. All I’m asking is to divide the bad/good players equally. Both teams should have average WR of roughly equal. This would balance the games and make those steamroll games less likely to happen.

    I hate it when I solo and enemy team is made of greens/blues with 55% WR and my team is entirely deep red with average WR of 46%.

    So if three player unicum platoon with 65% WR goes to havok, to balance them MM should give their team plenty of 44% tomatoes, and put more greens/blues/unicums to other team to have roughly same average WR.

    Forget the definition of skill and how to measure, just use the WR of the players, its accurate enough to do this kind of balancing. Yes I know platooning can bloat your WR, but it’s still accurate enough to do the job I described.

    If this kind of system can be implemented succesfully, it would mean unicums WR would sink, and red potatoes WR would rise.

    • and thus the skill-mm would be broken because the wr would be over the course of 1 patch equalized and the x-tra dimension of stat-rivalry and acceptance treshold into most clans would be useless too.

      Skill MM wil only reinforce the bad aspects of the game that it is supposed to alleviate. There is no way to do it and have positive results. Give me a scenario when you think it would be good and i can counter it the negative aspects of the *solution* and how it would make things worse.

    • LOOK

      If I wanted to spend my time looking after infants I’d get a job at a fucking nursery. What you are proposing just means that every good player has to babysit a bunch of fucking children every single game. That’s not fun, it’s WORK.

      • Give this man a medal.

      • One of the best comment here.

30. “if they want difficult matches where coordination & cooperation is everything, there’s Clan Wars or Platoons.”

    are they even playing their own game? going into a battle of 15 vs 15 with a platoon of 3 changes nothing!

    • Lol Wut?

      Sure if you and your tomato friends toon up, it may end in disaster.

      If three unicorns team up, they can rape the other team alone -ask the three Type 59, batchat, t57 platoons about that

      • It’s “unicum”, not “unicorn”.

        A unicum is a very rare instance of something (actually the only instance). A unicorn is a fable animal.

        • Don’t use the dictionary when talking about terms used in WoT.

          Unicorn is considered to mean the exact same thing a Unicum means, an extremly good player.

          The same way a Tomato has the same meaning as a Deer or Potato…a Noob.

    • I have to ask you the same thing, do you even play the game?

      Doing randoms in a platoon of unicums will severly boost your win chances.
      Knowing that you have at least 2 other exceptional players to back you up in case you fail(as shit happens) is far better than watching how your team melts down once you fuck up.

      Same shit goes with a 3 man platoon of useless players.
      Playing 12vs15 will severly decrease the chances of winning a match.

31. I’m starting to like more and more the way developers think.

32. Surely its not too hard to deduce a reliable “skill rating” for each player?

    Even the current XVM stats are a pretty good indicator of all-round player ability – or who has been botting, or stat-padding, or re-rolled, etc.

    I think the real worry that WG would have about skill matched MM using an XVM-type algorithm is that it would reveal too much about the game management itself….

    • Its rather had to “stat pad” WN8 outside of a few specific tanks.

33. First of all they would have to fix personal rating.

    • It seems that WoT PR is just WN7 *5.25. At least that gives you a very close estimate of the WoT PR for 95% of the players. The correlation coefficient of WN7 and WoT PR is very close to 1.0.

      • Nope its not, as the Wot PR doesnt even care what tiers u play. For Wot PR doing 1k dmg in tier 1s is the same as 1k in tier 10s whats clear bs.

        • Yet if you take over 95% of the WoT players and divide their WoT PR by 5.25 you get a number that is very very close to their WN7 rating. The correlation coefficient is very high. The formula that is used now must be very close to WN7.

        • The correlation of the WoT Personal Rating/5.25 and WN7 rating on large groups of players is between 0.96 and 0.97. That’s a rather high correlation. The formulas must use the same data in very similar ways.

34. I think the devs are making a grave mistake, and that is considering win rate to be the primary skill measure.

    Another mistake they are making is trying to use a global skill indicator for each player. They should measure player skill per tank he owns and is going to use in the given battle he is going to be matched for.

    I also believe that there is too much RNG in the game. As I previously posted here, it makes it hard for bad players to understand why they failed, or why they were successful. Of course, there are “intangible” elements of skill, like map and tactical awareness, but yet less RNG would mean better feedback for the actions players perform.

    Btw, I believe that WoT is “helping” the RNG in certain situations. Why is it that the IS-6 with its horrible accuracy so often hits when firing “out of the hip” or on the move, while e.g. the T34 as good as always misses when doing so? I am convinced that WG has some extra calculations implemented here. I wrote something about it on the WoT forums. Here’s a brief outline: The most common algorithm calculating a Gauss distributed random value actually always computes *two* values. One may be worse than the other. It would be easy to compute a mixture of both values depending on tank and situation when the tank fired (e.g. on the move, or not fully aimed). I believe that the WoT code is doing something like that. WG is hiding a lot from us players.

    I am also convinced that truly good players will stay good even if regularly matched against other good players. They would still achieve above average damage and kills. It’s only the win rate that would level out, and that is a clear indicator imo that win rate is not a good skill indicator. It’s favored by a RNG based MM often putting good players in one team, but not the other, simply due to the fact that there are comparably few good players, so they have a good chance to not have equally skilled players in a match they enter – there just aren’t enough of them so that RNG would reliably hit some of them too.

    Regarding XVM: XVM cannot do miracles, and if it doesn’t get fed with certain data, it cannot make predictions made on these. The devs should stop laughing because it is *stupid*. Give XVM all the data, and then see what’s going to happen. Btw, I have XVM set up to show me the player win rates with the tanks they are entering the current match with, and that is quite a good indicator of how a match will go, and also does justice to good light tank and scout players.

    • soooo, I can get 17k Wn8 games with my t6 light tank, doing medicore damage (2k) without spotting damage taken into account.
      But getting mastery badge, patrol duty, 6 kills and 1,2k damage on my T3 TD gives me 2,8k WN8

      Yeah, thats balanced. Sorry, WN8 is just inbalanced in itself, taking this thing measure skill is just…..

      First, WG would need to clarify how “skill” is truly to be measured, and I think thats just impossible to do, Winrate is the closest thing to it I think, take avg kills and damage into account as well, and you may get equally talented players face each other. Which is not a good idea. If you always fight against EXACTLY your level of “skill”, you wont stay “good”. Because, 1v1, only 1 can win.

      And to be honest, most unicorns get their stats by feeding on bad players – you wont be able to catch another unicorn offguard and put like 2k damage into him, because he “panicked” or did sth really stupid. Nor will he spam HE on your HT, or miss your weakspots.

      Your stats WILL drop

      • There it is again: The failed concept of win rate as a skill indicator! A good player is a good player even with a losing streak! Extreme example: Good player does 10K damage and kill 9 enemy tanks. He ends up as the only tank of his team, and the remaining three enemy tanks will kill him. Every other player (including those three good enemy players) have achieved less than 1/3 of his damage and kills.

        But they won. Does that make them the better players! NO, IT DOESN’T!

        Win rate mainly rates TEAM performance! That is why RNG MM favors win rates of good players.

        You need to get rid of that concept of win rate being the main and only valid indicator of player skill! As soon as you have a skill rating that honors individual performance more, the way players think about playing a match and being good will change, and the less will they have a problem with skill MM!

        • So, you mean a truly good player will have a losing streak, that affects his global winrate as much as 1%?

          How long is this “losing streak” going to last? 10 games. 100? 1000?
          If you are on a losing streak that extends to more that a couple of games – your are not a truly good player. You are average.

          NO Unicum has a losing streak of like 30 or 40 games. Just impossible for a player that does constantly more damage than everyone else.

          And yes, ON OCCASION your example MAY happen, biut if he kills 9 tanks and does 10k damage, that wont happen even twice in a row.

          S T U P I D example that was. Sorry. not going to happen twice in a row.
          And, do you think he would do 9 kills and 10k damage against palyers of his calibre? Make it 1 kill and 2k damage instead.

          And you know why losing games net you the most WN8? Because you skilled player are the last one to go down, which allows you to finish off all those damaged tanks and do a lot of damage – but thats, on the other hand, not really the damage your team needed. As QB says: Didnt cary enough.

          You need to get your concept right before claiming to know it better – READ what the devs say. Its not JUST winrate

          • FORGET ABOUT WIN RATE!!! It is not a valid skill indicator. It mostly measures team performance. That’s why the current RNG based MM favors good players so much.

            The example was meant to clarify that, and to show what player skill boils down to.

            You have understood NOTHING.

            Skill MM will level out win rates around 50% – but not other, better performance indicators, like avg dmg and kills.

            • Wait… So everyone will win 50% of their games, but some players will still contribute more? Some players will have better average damage than other players, yet still win the same number of games?

              How in the hell is that fair? Work harder, get the same reward?

            • Skill MM will level out all stats. How the fuck you think you will make more xp and avr dmg of you suddenly meet more good players? It wont happen, all stats are related to eachother more or less. You win less by facing better enemier more often and your avr xp and dmg will dcrease ALOT. You only get above average stats by facing majority of worse players (most of the playerbase), do the opposite and you become lower stats wise.

              • And *looking up in the comments* some people still think we don’t get it…
                Leave them buddy, I’ll share my popcorn with you once this shit happens.

          • You simply know nothing about statistics. WR is NOT a valid skill rating if it’s measured for one game only. Yet, it IS VALID if it’s measured hundred times, let alone 1000 or 10000 games.

            Simple reason : Team is random. It’s a random factor, and yourself is not. Random factor is ALWAYS fair under a same treatment, so the only thing which you can control is YOU. So, after 10000 games, the only thing which could control your MM is YOU.

            Show me a 70… no, even 60% WR guy which is NOT skillful, you won’t be able to. Yes, it could be inflated by other things (tank mostly played, platoon), but :

            1. If you have a high WR with sealclubbing, then he is simply skillful at low tier. WG simply need to add average tier to the formula.
            2. If you have a high WR with platooning, then you could also count “forming a good platoon” as a skill variable.

35. OH GODS… This is going to be a major screwup. Why break something that works good?
    DEVS, READ FTR COMMENTS! Much better idea is a REBALLANCER to current MM. Just contact me, we can talk this over ;) .

    >> once again, how do we…. Measure…… skill???

    What matters is RECENT PERFORMANCE on a CERTAIN tank – i don’t care if you are an unicum, whose computer broke and you now have to use 3fps abacus, or if you are 7year old would-be unicum, or your girlfriend just dumped you. And if you spam gold ammo only, i’m ok with that because your influence on battle outcome is already accounted for. From balance point of view only your latest scores matter.

    So, everyone with a basic knowlege of control theory will grasp instantly that this must be self regulating and stable. Pick LAST 20-30battles ON TANK and average them (XP is a good entry point, to rebalance in future), you can weight it tank tier vs battle tier and you get battle proficiency index. All you need to do is reshuffle players horizontally between two already assembled (random!) teams to make battle balanced. Works for platoons too – members are moved together.

    And, you don’t use hated wn8 and xvm ;) , you can keep PR index for other uses without redesigning it, you evade WR statpadding traps (and it won’t deregulate itself because of WR feedback loop). Current (working!) MM stays as it is, waiting times aren’t getting longer.

    Give it a go, at least…

    • PS: there is another advantage to that: it rebalances OP tanks together with players.

    • Using tank recent stats is not workable, since usually it has a very small sample. Statistically I think tank with <30 battles can't be counted with your method.

      It could be used as another factor in the equation, though.

      • 30battles is enough for wot – it’s quite a lot. And it takes into account elite/stock status, crew level and all other deficiencies/boosts. The only flaw can be with premium tanks as you can train multiple crews – but i don’t think that it’s a big problem.

36. First of all, before talking about the so- called “Skill Based MM” the margin of RNG must be reduced! RNG is a too powerful variable to the game imho.

    Because +/- 25% of variation of important parameters to influence the game is too powerful in terms of beeing able to troll the players performance. I even think, that RNG is already there to influence the game accordingly (helping Noobs, reducing Unicums performance silently) we all face this feeling every day we play WoT, no?

    Lowering RNG to +/- 10% variation would make the game “more honest” – meaning its not artificially sabotage your performance or buff your performance without any reason, just by pure dice-roll.

    Why is the inflicted Damage able to vary by 50%? It is absolutely independant on how much armor you penetrated or what distance the Target was you hit. Just pure coincidence, which imho does not make sense at any point. “If” Damage varies that a penetrating Shell did, it should only be dependant on how much armor it has to penetrate (meaning what Energy is left after penetrating the Armor).

    At the end there are so much things in the Game that should be made more concrete before we start to mess around with MM by Players “Skill” and adjust it. Keep in mind that here are many Statpadders aswell – are they skilled? Maybe in picking Tanks, but not general in playing WoT – skill is relative, to many factors.. like the Tank, the Map, your Mindset (Mood) and even Internet Performance at times…

    Just my two cents

    • Further i even think that you get a certain “stack” of RNG parameters prior to every battle by the Server. Your client walks through it “shot-by-shot”, “hit-by-hit” – your performance is decided already to a certain factor before you start the battle.

      This is done to keep as many players happy at the same time as possible. Meaning giving as much as possible people the feeling of success and reward. Unicums get over time a certain (lets call it) “Challenge” of reduced RNG parameters, while Noobs get a certain buff (lets call it) “Help” to compensate their Ingame results by a certain margin.

      This is done that as many players stick to the game as possible to generate the highest possible success with it. Mind, that in “programming” nothing is abandoned to coincidence, so why should that possibility “to do so” just left without attention?

      Many would wave this aside by the word “Conspiracy Theory” but this word is wrong in that consens (no Conspiracy necessary) – cos its just modern “Business” – and Mass Market handling in PvP / free to Play Games. Every developper that would neglect this potential would probably not do his job right – certain disadvantage to the “good Players”.

      But anyway i think Skill is still visible over Time quite truly – but it can vary by a lot in the particular battle – your particular Tank on a particular Map (with or without Platoon). And therefore, i see no benefit in this “Skill based MM”

37. It’s like these developers think they made the first computer game in history, that skill MM is a completely new idea never used before. While it is true that skill is practically impossible to quantify exactly, there are ‘good enough’ approximations possible (just like how tanks are balanced now, particularly tier 8).

    Balancing teams by vehicle type and tier is a given, but balancing by player ability is deeply controversial. *facepalm* Might as well go back to the ultra wide tier spread by that logic.

38. First I would like a skill MM but on my own terms.

    1. Battles fought. 1-5k, 5k-10k, 10k-20k, 20k – end (whaterver numbers may vary)
    I often watch guys with 1k battles buying a premium tank platoon with another nub who did the same thing and ruining the game. Many players don’t realize but they don’t really know how to play untill 10k battles and minimum 400 battles with a tank to truly know that tank, untill they join a clan, etc (good clan that is)

    2. WG rating match rating… atfer you pair them by the number of the battles try to pair them by WG rating with a 2000 Wg rating difference.

    3. Ban Xvm or any mod that shows in game stats. the players who think they are good… will ALWAYS bash the red players and hide in the corner of the map. Lots of players will drop the game because they are constantly bashed, insulted by “better players”

    • Well… the truth is… you can bash bad players (aka reds) without any XVM or such mods. Just take a look on what they are doing/not doing.

39. Jesus. Just balance the baddies. There are oceans of reds playing and a few good players, but MM tends to stack them together, all the good players to one side and all the retards to the other, hence the astonishing number of 15-0 battles every day.
    Just put the same amount of retards for both teams, and try to minimize having a tomato be the top tier for one team and a unicum for the other. The example of the IS8 could show that. If the IS8 on one of the teams is a retard that suicides in the first minute, the comet won’t have a choice but to deal the enemy IS8.
    Also, one of the devs are clueless about light tanks. The good light tank players happens to be the ones using them as medium tanks, not scouts. They killed the scouting role and wanna blame XVM and WN8 for that? And it is their fault spotting damage isn’t in the API.

    And damn, who said anything about removing RNG? But making a game where RNG is half of it is equally, if not more stupid. Tone RNG down to 15% or 10%, so a match doesn’t get decided because your 750 dmg gun dealt 530 leaving the other with 1HP, who in return shoots you for 900.

40. Why do they need to divide the players into “leagues”? Just have the MM balance the skill on each team so you don’t get those battles with one team of all blue and purples against a team of all orange and red. No more rotflstomps and everyone has fun.

    • Because that would FAIL to resolve one of the key issues at hand, which is the frustration of better players at having a team full of tomatoes. By doing what you’re asking for, you would be OBLIGING good players to HAVE to carry tomatoes every single game. And where would be the incentive to improve to bad players, when they’re always being carried?

41. I really don’t understand the issues here and certainly don’t understand why the developers are thinking of such a complex solution.

    From first principles, the key objectives are:
    - Create a league where better players can avoid the frustration of ineffective teammates, whether they’re AFK, bots, “play for fun”, or just inexperienced
    - Create a league where inexperienced players have the opportunity to learn and practice their skills, without being instantly owned by unicums and sealclubbers

    To me, the easy solution is:
    - Create x2 new leagues, an “upper” and a “lower”, just logically split the playerbase in half based upon their PR rating
    - Make this an OPTIONAL mode. Some people want to play normal randoms (e.g. sealclubbers). Let them do it
    - The EU has two datacentres. Allocate one for normal random games, the other for the leagues. By only having 2 leagues, you get a big skill base, so wait times shouldn’t be too affected
    - No change required to MM, better players who chose to play league game, go into the upper league and tanks are allocated into games as per the present. Note that it’s a wide skillset, everything from average to unicum. Thus saves cost for updating MM
    - By having defined leagues, the expected impact upon better players for any form of skillbased MM can be visualised. Just collect stats for a players performance in EACH league. Thus if he’s always playing upper league and his rating is lower than a player who only plays randoms, he can explain why. For genuinely good players, it’ll be a mark of skill that he can show he only plays upper league games. Chances are that good clans would require it. It would mean a cost to update the DBs to add a variable to cover the league, but it’s probably going to be MUCH cheaper than updating MM.
    - Promote between leagues one a week (e.g. on a Monday morning), based upon their rating for the previous 4 weeks averaged. Top and bottom 5% of each league get passed to the other league

    Can’t the developers PLEASE look at this and debate it.

42. What would be funny to see and then maybe shut up the mounths of some whinners
    is WG gets fed by the request of skill MM and just throws skill MM in the game for a small period of time
    and then we will see how these people who glorifie the skill MM would react
    it may costs players and money to WG but it would be fun feasting on the tears of those who are saying Skill MM or get the fuck out would react
    And those who say look at LoL
    Does the brain actually works??They are 2 diffrent games!!!
    LoL is a moba of a 5v5 in a small map with 3 lanes and certain build ups for each character
    WoT is a MMO with 15vs15 in bigger map with TANKS not walking characters
    it doesnt have builds all tanks have a final form and it depens on wheather the RNGesus smiles on you and on your SKILL
    also the economy of wot is based on tomatoes getting raped by good players if you take them away then you are strongly influence the gains of players
    also Randoms is Random and should remain Random with only small changes into the MM nothing more

43. Ban the XVM, problem (the whiners and qqers who cry when they see unicorns in the enemy team) solved.

    • Why, will all the tomatoes also leave and/or magically become enhanced and actually start doing SOMETHING USEFUL?

44. quote: “let’s assume for a moment Mr. 63% wins gets into the ultra-unicorn league”

    WG already failed – winrate is not a measurement of skill

    with a skill based MM the winrate for everyone will stabilize at ~50% after the 1st 200 battles or so
    skill is determined by his personal rating, not by his winrate

    • No, you actually failed at reading.

      Further down is well explained why the concept wouldn’t work with that measurement, and devs are brainstorming ideas and problems.

      Nothing wrong with having lines of thoughts. and either way, i dont think Skill MM will come, as it stands, a grand majority of players dislike the concept, and so does the devs, so expect it to be shelved.

      • I didn’t failed at reading, after I read that shit I stopped reading

        what’s the point?! they clearly are set up to demolish skill MM with random generated excuses – like I said in the other topic, there are games out there with a skill based MM and they’re going strong
        if WG doesn’t want to do skill MM then stop with this BS, work on the fucking game rather than disproving the utility of it

        bunch of morons …

        case and point, they didn’t even tested it: quote “I can’t see it working correctly for now until we stress test.”
        and they already know the outcome ?!?! what the fuck!

        • also why is XVM “involved” in this?!

          XVM uses a Nth iteration of a half-baked formula some people wrote based on data WG allows access to
          there’s a lot more data WG doesn’t make available so the formula would be accurate
          please don’t tell me WG actually thought of the skill MM based on XVM ….

          • Well by common rule, if you don’t read, you don’t get to have an opinion because you dont know what you re talking about.

            and reading about your iterations of XVM, now i know you truly don’t know how to read because the devs mentioned none of that, but used XVM as a flawed measuring model, and perhaps, something they dont want to use anytime.. completely contrary of what you just said.

            Reading is damn hard, if you re not gonna read and comprehend, why bother posting jackshit?

            • I browsed it, read the most important bits
              and no where do they talk ELO rating .. nowhere
              why I’m bringing about XVM, it’s becasue XVM is the only “tool” used in-game to differentiate skill, they even talk on how it misrepresents light tanks in the formula

              maybe you need to read it again

              • at one point one sais: quote” Players with exceptionally good personal ratings (think of +9000) ”

                say what now?! what rating is that?! there aren’t such numbers in ELO

                how MWO did it: http://mwomercs.com/forums/topic/66079-matchmaking/

45. I’ve seen some folks that are essentially hinting at this but I’d like to propose a (relatively) simple tweak.

    Leave the core MM alone. Simply add a second pass after the players/tanks have been selected. This would essentially be a team balancing pass. Since this is just a subtle tweak any stat would essentially work for determining someones potential contributions. Let just use WR for this example.

    If team 1 has an overall worse WR that team 2, check to see if there are comparable tanks that could be team swapped. I.e. each team has a T30. Team 1′s WR is lower but Team 2′s T30 has a higher WR. Therefore the T30′s are swapped. This could be done multiple times to help balance out the games where you have all of the tomatoes against all of the UNIs.

    I know there is some whine about not wanting to play with baddies, but I think the majority of folks just want to have matches with a fighting chance to win.

    A second pass on MM, should only add a couple seconds to the loading of the battle, the team would already be picked. This would not be expected to produce perfectly 50/50 chances for each game but I think it would significantly cut down on the outliers with a significant skill difference between the teams.

    • I’ve seen that suggested by others before. However win rate can be padded by platooning. What might work well instead of balance the teams by average damage and average spotting damage in the vehicles they are playing.

    • Nope, WR is very bad for that, i explained previously:
      1. it will totally deregulate the whole system very quickly
      2. it is not precise enough, small values spread and big differences.

46. They could make meaningfull skill match making but essentially they don’t want to cause its a lot of work since there are so many variables.

47. Interestingly enough, this change will affect mostly the great/superb players because they will start getting a knife-end match every now and then commonly, and thus their status will suffer, and more whine will come, while the regular and good player may get a long term benefit by farming (sucking dry) the super players’ stats, is this intended?

    ^^^ This is what the Unicums are scared of…. If wargaming is smart they will improve the system with a Skill Based MM, for the MAJORITY of their player population. He just told you that:

    ; “about 7% of the estimated server population on the RU cluster has a below 47% while a staggering 77% has a ratio of acceptable terms between 48%-52%),,,

    It’s apparent that they are hurting at some point somewhere. Players are tired (Those of use in the middle) of getting Hurt from both ends…. All of us (the Majority) have been getting the short end of the stick, from the UPPER % players, and the LOWER % Players. WE the MAJORITY, who are the 48%-52%, are tired and feed up with the EXTREME LOW% & EXEREME HIGH% players hitting us all where it hurts.

    Wargaming cannot keep the Majority Suffering much longer just to make the HIGH % players Exclusively happy with their Win Rates, at the expense of everyone else. And the Low End players the same.

    We Majority Players in the Middle are tired of it,…. AND We want a Skill Based Match Maker. Tired of the disparity in player skill, Lopsided defeats, Eradict game STOMPS that are just ridiculous….

    • So the 48% ers will magicly become 50% and the above 50% will suffer. That is bullshit. If you are good you have the right to use that advantage towards others.

      • that’s just because you measure skill only by WR – it’s wrong

        • Skill MM will even out all stats eventualy, as well as average xp and damage. Because for good players if you win less due to skill MM, one can also assume that you make less average dmg (because facing equal opponents) and hence earns less XP.

          In order to make more damage you need to get the right opportunities, and facing more equal players it will be harder making as much damage as you do now by facing random players (who are mostly average to below average).

48. Every time I see one of these I lose a little more faith in WG. If the game is 50% skill and 59% RNG there is no reason to waste time playing it.

    I honestly think people don’t care about skill MM. All they want is better matches. Do wit with skill MM, do it without, if we get good matches we’re good.

    But WG is completely clueless

    • Well, I kinda know what you mean. I sorta always think this in the back of my mind, how RNG, and the MM controls the out come of a game etc. Along with other variables in the game etc. And they just even spelled it out here that a portion of it is not skill at all, actually.

      And even funnier is when you try to even say this in a truthful light over on the main forums, your immediately, “Swarmed” and ran out of town…For even thinking it.

      You know its almost like, they are saying ( a few of them there ), “Why bother, its not controlled by skill anyway, its controlled by the RNG, threshold settings in the software, or that, or this, etc….

      I’ve had days when I’ve gotten teams that were SO Lopsided, (like 9-10 games in a row) where I would get top Damage on about 7/10 games.. (all were losses) And, no I did not do it suddenly at the end, like most of the Damage Farmer tanks, camping in a heavy until the end, but the entire teams were just player that could not play at all.

      How does anyone explain those kind of matches? When the team is just “Rolled over, obliterated” in about 1.8 mins. in that game? Were they all Bots?
      Asleep? or could not play?,,,, but 10x in a row… Kinda takes the fun out of the game. And then you say, “Why do I even bother to keep playing anymore?”

49. One thing that many topics of skill based matchmaking skip over is getting rid of the landslide/steamroll battles and the battle with predetermined outcomes due to very large skill differences.

    In an analysis of over 2,000 solo pub battles, this is what i found:
    Players lose less than 3% of the battles with a greater than 60% XVM chance to win.
    Players win less than 3% of the battles with a less than 40% XVM chance to win.
    the battles outside of the 40-60% chance to win range account for about 30% of the average player’s battles.
    So basically 30% of the average player’s battles have a predetermined result due to the difference in skill between the teams.

    If we could get rid of the battles outside of the XVM 40-60% chance to win range the game play would be more dependent on layer skill and teamwork. Yet it wouldn’t force everyone to have the same win rate.

    • About steamroll battle, as i see it, there are 2 common scenarios:

      1st: One team make a fail rush,or enemy make a successful rush, after that the number is just too difference for a come back to take place. It’s the communication problem within the team. No MM could fix that.

      2nd: The top tier tanks get smoked first, because he/she makes mistakes or enemies just being better, after that everyone knows what happen. This is the case which consider “skill” in MM could somehow offset the problem. For this we don’t need a full team Skill-MM, we just need to consider the skill of a few tanks in each team (Properly, 4-5 top tier vehicles). That way we can have a bit fairer match, while the dynamic of the game is still retained.

      But one way or another, I still think skill-base MM is a waste of time. Isn’t that we have team battle? Wasn’t that some kind of skill-base MM? If those people really want skill-base MM that much they can just play team battle.

      • Team battles are very restrictive on what tanks you can use if you want to play using skill based matchmaking. Also, the skill based matchmaking for team battles only works if there are a lot of teams waiting to play. Otherwise it throws in a group of yellows against a group of purples and blues regularly.

50. Interesting part is that separating people by ratings may create even more disparity.
    Say, if we gather all players with 53% WR and let them play amongst themselves, those who win more (for skill or RNG) are prone to “graduate” to the 54%, while their counterparts will “downgrade” to the 52% category.

    It works the same if you use the WG ratings or any WN formula. Those who will will increase their stats, those who lose will lose points, and they’ll eventually drift out from their categories.

    Even the idea of creating thresholds of ratings will keep on causing drift, unless all the better players are bunched in a team and the worse ones are put in the other. But that would be even worse.

    Best option would be to let the players choose if they want to enter Random MM or Ranked MM, and that would eventually lead to people leaving the Ranked category because it’s either unpopulated or filled with unicorns anyway.

51. Skill MM is the dumbest idea I ever heard of, I’ve seen many skilled players whining about noobs here and there, but they forget that with Skill MM their average damage will go real low. Remember, you never get an Ace Tanker Badge when your teammates are really good (let’s say +80% win chance) hope that gives you an idea.

    • If there is skill based matchmaking, you won’t ever get 80% chance of win battles.

52. Skill based MM is THE worst thing that can happen.

    If you only play against equal skilled players battles will be boring, your winrate will close to 50% and once you finally reach upper bracelet you’re just cannon fodder for next 500-1000 games for higher skilled players in your bracelet.

    Not talking about waiting times for game to find 30 people with same RATING and SAME TIERS. Do you really want to wait 1 minute for each ~8minute game? Even if you say that “I rather wait 4 minutes than play with n00bs” it’s going to be hell of a lot more annoying to wait long time for a game to start, then realize that every “dirty” or “secret” trick in your book is worth nothing and you’re stuck there camping with people who just want to win without getting hurt, dying and thus reducing their ratings.

    Seriously, anyone who even suggests “skill” based MM is just not thinking it through. Because here is the reason why skill based MM is bad:

    THERE IS ALWAYS SOMEONE FOR WHO YOU ARE JUST AN EASY KILL

    Only few players in world can say that they’re the best, and they are the best. But even for them Skill-MM would be playing with same ~100 people all day long. Fun? Don’t think so.

    • Of course, randoms are for relaxing. Team battles and clan wars are for people who want a real challange and meet equal players and earn real assets (gold). Not to mention clan wars are serious team play, in randoms its not like that and hence everybody will suffer meeting equal opponents. You simply dont have any advnantage any longer and everything becomes a game or RNG and latency.

      The ones that will benefit from this are the players that are average to below average because all their stats will level out. Their WR will reach 50% and avr dmg and xp will also stagnate and certanly not become any worse = steady xp and income. I just wonder how this skill MM will affect the economy. Baddies will get more consistent income and good players will get a overall lower xp and credit income (due to facing equal players and beeing less efficient than before).

53. Very interesting. Iam not really a fan of a skill based MM, but the dev talk showed some interesting insights. I see no problem in general, but the problems will start with tier differences and Iam afraid that the great and exceptional players aka blue and unicums will suffer the most playing anything lower then tier 10. Not to mention full platoons of ppl with 11 000+ ratings… How to balance that? With giving them 12 red players and in the oposing team 6 blue and rest green players? The queue times for such platoons would be horendous.

54. Letting equal players face eachother is a game of lottery, because then it is all about who has the best RNG roll and latency, similar to drag racing in GT5, to some point it was all about who had the best latency to the server that got the jump of the start.

    Once again making games more equal 50/50% chance of winning is in fucking deed balanced but a “lottery”, because both teams are as good and hence a win is decided by luck/rng. Equal games is not what you want in randoms in terms of player skill. People play randoms/public for a reason.

    Current system encourage players to become better so they can domnatie the larger playerbase (48-52%), with skill MM, becoming better, wont give you any advantage over the players you meet because you will meet equal ones, despite the majority of players are below to average (48-52%).

    And this is not only WR, but stats like avr dmg, xp, kills, spotting, survival, hit ratio etc. All are related to eachother and the players you face. Play againt average to below average players and then againt “equal ones” and you will have a much harder time using your skill efficient, because you simply dont have any skill advantage any longer.

55. Think of skill MM as a “tax”. Now the businesses (players) earn profit by having a competitive advantage over the majority of businesses (below to average players).

    If skill MM will come, all businesses that earns over a certan limit will have to pay higher tax (just because they have a advantage) and because of that added tax, they will profit as much as the others. Now all businesses have the same profit because the ones that earns more has to pay more tax. No matter what the businesses do to compete, the more they earn the more tax they pay. So in the end a low en business (below to average players) and high end businesses (above average players) get the same profit. And when ever the business is trying to compete and perform better it gets punished by having higher tax, just so all businesses should be the “same”.

    Would we accept this in a free society? Hell lets make your parents work their ass off and pay their wage to me, because I am a student and hence nobody should get more money than I have.

    Your cute mother gets a raise on her hard work? Hell pay that fucking money to me, why should she get a profit and hence cause “unfairness” towards others that works less.

56. I don’t care about skill MM.

    I’d rather see an MM that doesn’t give one team 10 TDs and the other team 12 mediums.

57. I just made 10 kills in tiger II and yes both teams were kinda bad I guess and I had luck. Dont take away these epic moments from players by forcing them to play with equal opponents all time. We play for epic random moments, and both noobs like me and unicums need it.

58. Simply make teams equal, no leagues, enough said.

    • Equal is the same shit as leagues, it will make WR equal of all different groups. Having equal teams means having the same amount of ideots, average and pros. No way to progress.

      • Having same amount of reds, yellows and mylittleponies on both teams is better than three ponies troll platoon against all red enemy.

        • No, if you makew equal you ruin the random nature of the game.

          • You may enjoy it the way it is now, but some others like me want more battles based on skill than based on luck.

            And you know that most red players are kids and dads, right?

59. (っ-●益●)っ

60. To me, this is as simple as fuck, try to balance the teams according to the players’ skill in the queue, meaning distributing players evenly to the both teams, according to their skill.

    So don’t put all the good ones in one team and the rest in the other.

    They should also take into account tank modules, stock tanks are no match for a fully equipped one.

61. I like dev #4, he/she knows what’s up with XVM and WNEfficency’s terrible algorithm.