Sabotage Matchmaking

124

Comments

  • sashimiaksashimiak Member, Moderator, Robot Entertainment
    Guys, I can understand that passions are running high here and I know you all have the best intentions and want to improve this game with your feedback. But please, refrain from calling out names, accusing each other and starting arguments. If you think a specific player is abusing the system in an unfair manner, please report them here: https://orcsmustdie.com/#!/en/support or try to resolve the issue in private. We all love this awesome game and we want it to be as good as it can be, so let's stick to constructive feedback and sharing ideas. You all have made some excellent points and I really don't want to start censoring or deleting posts. OMD!U has one of the friendliest, most constructive and positive player bases I've experienced in an online game. Let's keep it that way!
    German Community Manager
  • PericlumPericlum Member
    Hello, I am here as a soloQ player.
    I have three questions:

    Have you guys even played soloQ much?

    Do you guys decide who is going to pick Tundra and Ivy before every match? Or is it a rule in your premades that you must have them in every match?

    Have you premades every lost rift points in waves 3, 4, 5, 6, 7, 8? (Except for Swifty)
    Yeah, I used to, before I joined a guild. It's way more fun to play with clever people than randoms. Discord, an actual strategy, coordinated traps... everything helps.

    Nope. Some people have particular heroes they like that synergize well with a specific map, and they'll say so. The other two grab Tundra/Ivy. Some people also enjoy playing Tundra and Ivy (I personally love Tundra). We WILL avoid everyone taking the same traps, though. Three people with barricades is good, three people with BGH Ballista is not. Oh, and we'll make a point of taking specific traps.

    Uh, occasionally. I know I've screwed up barricade placement at least once, and cost us a match that way. But, if everything goes well, that doesn't happen.
  • What about leaderboards wipe? Or at least i want my high rank back. Because it's went down by playing with t1 feeders against premades. I know party bonus gives you infinitive boss spawning cards for each round, smallest and easiest minion waves and a lot of usefull stuff. But it's unfair that i forced to enjoy this trash instead of balanced matchmaking.
  • PericlumPericlum Member
    I know party bonus gives you infinitive boss spawning cards for each round, smallest and easiest minion waves and a lot of usefull stuff. But it's unfair that i forced to enjoy this trash instead of balanced matchmaking.
    Uh. Was something lost in translation here? Having been playing in premade parties for about a week now, I can definitively say that we do NOT have infinite boss spawning cards. Can't definitively say about the other things, but AFAIK there is no actual mechanical bonus for playing in a premade party.
  • StatboyStatboy Member
    can we just ask what is the win rate of premade teams vs teams with at least 1 random, when they play against each other?
  • BeigeMidnightBeigeMidnight Member
    edited June 5
    premade win rate is almost 100% in every game i played. simply waste of time. they usually use and host their 2nd afk accounts together to play 3v2 or even 3v1.
    and as i can see https://forums.orcsmustdie.com/discussion/comment/53565/#Comment_53565 there is no balance by levels. they can easily send you 1-2 of their afk t1 game windows for ez win
  • they usually use and host their 2nd afk accounts together to play 3v2 or even 3v1.
    maybe i got your comment wrong. Did u assume all/most premades are using low smurfs to increase their winchance?

    premade win rate is almost 100% in every game i played. 
    depends on the premade you are playing against. There are to many kinds of premades to generalize it.
    Even hardcore Premades lose matches. 
    Maybe @shadedev can give us a more realistic winrate.

    and as i can see https://forums.orcsmustdie.com/discussion/comment/53565/#Comment_53565 there is no balance by levels.
    thats right. The Matchmaker has one Job. Creating balanced Matches. For this he tries to match teams with similar MMR.
    If this isnt possible , he tries to match teams with similar average MMR. And if this dont work, he matches teams with different average MMR and the longer the q, the bigger is the average mmr difference.
    Acclevel is not involved.

    I dont think the experience will be better if the matchmaker trys to match player with equal acclvl first and forgets to take a look how well/lucky they had played in matches before. It could be that you play with a high lvl player who isnt interested in teamplay oder simply not unterstand the easiest mechanics. Hopefully im wrong.

    Its the same issue that siege had. In Sabotage they tried to balance the problem per Minion scaling which is not such great and most people would agree. LvL 1 Heroes and Traps would be better for balancing this mode, but this would make trapgrinding unnessecery which is important in robots point of view (and for games health). We need a solution which helps to create more balanced matches , but again didnt delete Trap tiers for one mode.
     
  • thats right. The Matchmaker has one Job. Creating balanced Matches. For this he tries to match teams with similar MMR.
    If this isnt possible , he tries to match teams with similar average MMR. And if this dont work, he matches teams with different average MMR and the longer the q, the bigger is the average mmr difference.
    Acclevel is not involved.


     

    Well, he fails often I guess. Maybe it's a time problem, I don't know. But right now, I just got out of a match against a plat 1/1/2 premade

    I play with my gf, I'm for no reason gold 1, she is silver 2.

    But, if I believe the leaderboard, I see currently 81 players between plat 1 and 4 (at least what is shown). 81 is such a low base that any matchmaking system will fail instantly. However, for no reason then, the matchmaker find a match in less than 10 seconds.

    My deduction is we were 6 players to queue for a sabotage at this time..

    The mode is fun, but absolutely not balanced. I would avoid it at all cost if it was for the daily chest..
  • AxonyAxony Member
    The ranking is not the problem...
    The Main Issue is that the same Guilds and Premades exploit the Power Generator.

    And please don't tell me that it is not exploiting of you start wave 2 with 30.000 coins for each teammember
  • PericlumPericlum Member
    Axony said:
    The ranking is not the problem...
    The Main Issue is that the same Guilds and Premades exploit the Power Generator.

    And please don't tell me that it is not exploiting of you start wave 2 with 30.000 coins for each teammember
    Doesn't happen. In a premade, only one (I guess maybe two) guy takes the Power Generator, and at BEST you'll get 1000/round out of each one you place. I haven't actually done any tests, but my guess is that even with rental parts you're rarely going to break 1500 net coins/Generator.

    Though, either way, it's not an exploit. Complaining about using the Power Generator to accumulate coins is like saying using viscous tar to slow enemies is an exploit. That's literally the stated purpose of the trap.
  • AxonyAxony Member
    edited June 6
    U don't get the point.

    They put up 3-4 of them, keep 1 minion alive, and kite it over and over those 4.
    they facetank the minion until 1 drops low and then the next person takes the aggro.

    Now we use basic math:
    4 Coin generators
    1 repair tool to reactivate them

    It's like having a "Penguin" giving him the rank of "Admiral" and giving him "13" soldiers and 2 more people of a guild...
  • and that's when the time limit comes in ;), like most versus game, when a team finish a wave it should send both team to the break, so if team 2 did not finish the first wave, the second wave anyways after the timer runs down.

    It would make sabotage game faster and open up a world of new strat, like gate killing, faster at the start try to overwhelm the opponent but you open yourself to a less efficient killbox in the later wave, but the oppennent might not get a break time to build.

    but now time is not important, so you can build near the rift and bore your opponent, or stall and have a 30 minute game  :s
  • AxonyAxony Member
    the wave timer is way to long, i think it's 2-3minutes after team 1 finished
  • Axony said:
    U don't get the point.

    They put up 3-4 of them, keep 1 minion alive, and kite it over and over those 4.
    they facetank the minion until 1 drops low and then the next person takes the aggro.

    Now we use basic math:
    4 Coin generators
    1 repair tool to reactivate them

    It's like having a "Penguin" giving him the rank of "Admiral" and giving him "13" soldiers and 2 more people of a guild...
    you cannot reset power generator with gnomish repair tools
    i have never seen anybody with 30000 gold on wave 2
  • lielsliels Member
    If matches were fair *per player _OVER A LONG SERIES OF MATCHES_* then the chance for each of us to win would be close to 50%  (slightly higher if we are getting better faster than our peer group, slightly lower otherwise).  What should *not* happen often is that a player should have ... say over 20 matches ...  16 wins or 16 losses (0.5% chance) ... or 17 (0.1% chance) ... or 18 (0.02%) or 19 (0.002%).  Robot has the data; and I suspect that these occurrences happen way more often than could be accounted for by a series of 20 matches that are fair over those matches.  The best way, of course, is to have every match be fair.  But if creating a fair match is not possible the history of the players should be taken into account so that 16/20 wins or losses happen only 0.5% of the time; so if your past 20 matches have 15 wins, that player should be on the short end of unfair matches.
  • StraggyStraggy Member
    liels said:
    If matches were fair *per player _OVER A LONG SERIES OF MATCHES_* then the chance for each of us to win would be close to 50%  (slightly higher if we are getting better faster than our peer group, slightly lower otherwise).  What should *not* happen often is that a player should have ... say over 20 matches ...  16 wins or 16 losses (0.5% chance) ... or 17 (0.1% chance) ... or 18 (0.02%) or 19 (0.002%).  Robot has the data; and I suspect that these occurrences happen way more often than could be accounted for by a series of 20 matches that are fair over those matches.  The best way, of course, is to have every match be fair.  But if creating a fair match is not possible the history of the players should be taken into account so that 16/20 wins or losses happen only 0.5% of the time; so if your past 20 matches have 15 wins, that player should be on the short end of unfair matches.
    Disregarding that 20 got no real statistical significance (hundreds or even thousands are much better), I totally agree that 'in a perfect matchmaking' that should be the case. 
    The real fact is that the matchmaking system is not bad, Omdu simply does not have the player count to always make fair matches.
    There are three ways the devs could react to it:
    Extend the queue time until a fair match is found.
    Break the queue after a specific time when a fair match was not found (for example 10 or 15 minutes).
    Don't change anything and have relatively low queue times and sometimes unfair matches.

    I won't argue what's my favourite, but you don't have more options with this low player count.
  • lielsliels Member
    Straggy said:
    liels said:
    If matches were fair *per player _OVER A LONG SERIES OF MATCHES_* then the chance for each of us to win would be close to 50%  (slightly higher if we are getting better faster than our peer group, slightly lower otherwise).  What should *not* happen often is that a player should have ... say over 20 matches ...  16 wins or 16 losses (0.5% chance) ... or 17 (0.1% chance) ... or 18 (0.02%) or 19 (0.002%).  Robot has the data; and I suspect that these occurrences happen way more often than could be accounted for by a series of 20 matches that are fair over those matches.  The best way, of course, is to have every match be fair.  But if creating a fair match is not possible the history of the players should be taken into account so that 16/20 wins or losses happen only 0.5% of the time; so if your past 20 matches have 15 wins, that player should be on the short end of unfair matches.
    Disregarding that 20 got no real statistical significance (hundreds or even thousands are much better), I totally agree that 'in a perfect matchmaking' that should be the case. 
    The real fact is that the matchmaking system is not bad, Omdu simply does not have the player count to always make fair matches.

    There are three ways the devs could react to it:
    Extend the queue time until a fair match is found.
    Break the queue after a specific time when a fair match was not found (for example 10 or 15 minutes).
    Don't change anything and have relatively low queue times and sometimes unfair matches.

    I won't argue what's my favourite, but you don't have more options with this low player count.
    The 20 is the wrong thing for statistical significance.  It's sets of 20 games across the entire player base.  I'm suspecting that there are *way* more 16/4 win/loss over the last 20 games across all players than their should be.  My suggestion is to take that into account when setting up matches.  I also suspect that premade-3 makes this difficult.  Limiting premade to 2 and putting the weak player onto the strong team would balance things.  And probably help the weak player learn.
  • well that depend how weak the player is , because 2 level 100 with a level 2, that put the minion at around 66 level, the level 2 guy is not goiing to have any fun at all.

    we will have to wait for next to see what is the level requirement for sabotage, that should solve quite a few issues
  • I don't know what kind of feedback I can give here, but I do know that I can give some information. I have a Level 80 account. Most of the traps I use are in the blue/purple zone. Yesterday, I was at Silver IV. Lost a match and found myself at Silver V. I'm going to call this my start point, since I do not recall playing any more after that demotion.

    Today, I played about 9 or 10 games of Sabotage. It did not matter which character I played as each and every single one of them were losses. What bothered me the most was the last one I was in. I was in a 3-man team against a 2-man team. I don't know how that happened, but whatever. I also notice the other team didn't use a single barricade. As stated above, that one was a loss, too. I have been demoted to Bronze I now and I will not play Sabotage again until the next patch.

    Before Patch 2.1, I used to have more balanced matches. Many of them went into Sudden Death. I used to have fun in this.
  • PericlumPericlum Member
    I don't know what kind of feedback I can give here, but I do know that I can give some information. I have a Level 80 account. Most of the traps I use are in the blue/purple zone. Yesterday, I was at Silver IV. Lost a match and found myself at Silver V. I'm going to call this my start point, since I do not recall playing any more after that demotion.

    Today, I played about 9 or 10 games of Sabotage. It did not matter which character I played as each and every single one of them were losses. What bothered me the most was the last one I was in. I was in a 3-man team against a 2-man team. I don't know how that happened, but whatever. I also notice the other team didn't use a single barricade. As stated above, that one was a loss, too. I have been demoted to Bronze I now and I will not play Sabotage again until the next patch.

    Before Patch 2.1, I used to have more balanced matches. Many of them went into Sudden Death. I used to have fun in this.
    This doesn't sound like a flaw with the Matchmaking system. It sounds like you're bad at this. Your three man team played an enemy who didn't use barricades, and LOST? I don't even know how that's possible. You should have been able to out-stall them easily.

    What, precisely, do you think was unbalanced in the matches you played?
Sign In or Register to comment.