Skip to content

Antipode82#49495

← Customer Feedback for Wizards of the Coast

My feedback

1 result found

  1. Algorithm Improvement

    908 votes

    We're glad you're here

    Please sign in to leave feedback
    Signed in as (Sign out)
    Close
    Close

    We’ll send you updates on this idea

    1051 comments  ·  MTG Arena Bugs & Product Suggestions » Product Suggestion  ·  Admin →
    How important is this to you?

    We're glad you're here

    Please sign in to leave feedback
    Signed in as (Sign out)
    Close
    Close
    An error occurred while saving the comment
    Antipode82#49495 commented  · 

    It's pretty clear from observation that the Arena matchmaker favors specific types of matchups and prevents others.

    This makes sense if it were ensuring similar power level decks were paired together, but in practice it leads to a surprisingly high amount of repetitive matches, either mirrors or counter-decks.

    To illustrtate this, you can see the odds of being matched against a Gruul Adventures mirror or a Rogues player, of facing a Yorion player while doing BR Midrange, or facing RDW while playing slower decks with higher CMC blue, black or white cards. I've also noticed analogue cases in less competitive archetypes: you have much higher odds of being paired against mono-blue mill while playing UR Spells, for instance. I'm also suspecting this could be due to similar card rarity distributions between decklists, but it's hard to tell without the matchup algorithm or a much larger data set.

    I can understand if the spirit behind the matchmaker design is to prevent players with a wide range in power gap from facing each other and having a bad time. However, this is having unintended consequences both in quality of life and deck brewing data gathering.

    It's no fun for either player if they get constantly paired against opponents with the same or wildly similar decks, especially if the matchup leans heavily towards one end of the scale or the other, such as Rakdos vs Yorion archetypes, where the BR player has virtually no tools to interact with their opponent's board, or if matches are frequently decided by card sequencing rather than skillful interaction, such as most mirrors between players of similar skill levels.

    Moreover, having a matchup frequency bias is probably discouraging deckbuilding experimentation. Players who constantly face their worst matchup in random play will be disinclined to try new options that might improve the overall win rate against a more diverse field. You can see an example of this in the underrepresentation of Underworld Breach in the field, which is an amazing card for BR and UR strategies but a significant liability in their worst matchups.

    This also has major implications for seasonal ladder progression: players who rely on random matches to level up will be much more inclined to play decks that have a better overall game against the decks the MATCHMAKER pairs them against, while those who favored matchmaker hard counters will be left behind even though their decks and skills are much higher tier.

    I'm not sure what's the best response to this is, and like I said before, I can understand the spirit behind the algorithm, but considering that Arena is one of the few large-community play options for the foreseeable future, I'm worried about the playtesting implications for the community in the long run.

    Antipode82#49495 supported this idea  · 

Feedback and Knowledge Base

(thinking…)