1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
  2. Hemos actualizado nuestra PolĂ­tica de Privacidad acorde con la nueva RGPD. +Info // We've updated our Privacy Policy to comply with the GDPR. +Info
    Dismiss Notice

A MOment of bitterness

Discussion in 'PanOceania' started by Judge Dredd, Mar 13, 2019.

  1. ObviousGray

    ObviousGray Frenzied Mushroom
    Warcor

    Joined:
    Nov 27, 2017
    Messages:
    1,629
    Likes Received:
    2,768
    Give us the link to that site.
     
    Triumph, inane.imp and Barrogh like this.
  2. Titus

    Titus Varuna Beach Commando

    Joined:
    Nov 23, 2017
    Messages:
    283
    Likes Received:
    349
    It's painful to see these threads every month.

    I personally am leaving MO because I can't make lists I like and the lieutenant situation is annoying. So I'm going for Varuna and NCA for a long time. Maybe in the future I look back to MO with new eyes... Who knows.
     
    eciu and CaptainVenge like this.
  3. inane.imp

    inane.imp Well-Known Member

    Joined:
    Jan 28, 2018
    Messages:
    4,498
    Likes Received:
    5,355
    The ability to review the data in context rather than just cherry picked points.

    From what I can see based purely on what you've posted it appears that MO's performance is significantly more average than VIRD's. Ie you appear to be comparing something that's close to the median performance with an outlier.

    This doesn't tell us anything about MO.
     
    chaos11 and Triumph like this.
  4. inane.imp

    inane.imp Well-Known Member

    Joined:
    Jan 28, 2018
    Messages:
    4,498
    Likes Received:
    5,355
    Get you don't want to discuss it further here, but perhaps you could point us to a location where this could be discussed?
     
  5. Barrogh

    Barrogh Well-Known Member

    Joined:
    Nov 26, 2017
    Messages:
    1,954
    Likes Received:
    1,670
    Now, I'm not familiar with structure of wargame community and I don't know how much of its participants are really competitive and work on refining their gameplay for the sake of playing to win.

    If video game communities are anything to go by, most players don't really go out of their way to win every match they get into. Some people prefer to do what they think is fun during the game, some would stick to what they think is sufficiently good without pushing themselves out of comfort zone provided by stock decisions, cookie-cutter builds and easier to play options.

    As a result, most developers have to discard as many as about 90% of playerbase from evaluation when they have to analyze competitive balance (some very specific issues aside). You can't judge what a character / faction / deck / whatever can or cannot do if its "pilot" doesn't do everything he is given potential to. But even then the realities of modern digital world means that when it comes to vidiya you still has data on every match played, all statistics in the world you will ever want, and then there's the fact most VGs have short gaming sessions you don't have to spend hours traveling to get involved into, you get plenty of data even if you filter out most of your playerbase.

    What I am getting at, seeing this tournament data with single-digit numbers of some army's appearances, and also knowing that on top of that there are a lot of players who treat tournaments as yet another opportunity to play a few beer&pretzels matches with different opponents, and also knowing that sometimes even proven top players of your local meta come to such events with what they think is a gimped, "challenge" list, or just take a list that is less stressful to run (and not because it's good for particular missions) because they will also have to be judges at that event...

    And then there is a massive number of armies that keeps growing and scatters players across factions.

    Meanwhile, Infinity is still a game with a ton of variables which means you need a massive number of entries for your statistical data to mean anything at all...

    All of that makes me think we'll probably never get any meaningful data on an army's performance unless we lower standards for that data significantly. Oh well.
     
  6. Triumph

    Triumph Well-Known Member

    Joined:
    Nov 23, 2017
    Messages:
    2,606
    Likes Received:
    2,843
    You need to provide the people who's opinion you wish to change access to the data, so they can scrutinise it and confirm that you are neither cherry picking or accidentally misinterpreting the data.

    Otherwise you're essentially saying something along the lines of "Science proves X is true!" whilst also refusing to provide information on how and why that is so. Whether or not it is actually true is irrelevant because alot of people straight up just won't believe you because of that.
     
    Barrogh likes this.
  7. CoffeeBadger

    CoffeeBadger Well-Known Member

    Joined:
    Aug 6, 2018
    Messages:
    38
    Likes Received:
    50
    As someone who has next to no clue about the depth of Infinity mechanics but has a superficial grasp of statistics this cannot be stressed enough. A statistic is of very limited reliability if it does not account for systematic biases. For low number statistics like in wargaming small changes of absolute player composition may already lead to large distortions of the observed metrics (cf. that QK player on the t3 site). So if you want to take an objective look at factions you would need to find a representation that is less player-skill dependent.
    That could mean normalizing each player's match data with respect to their global performance across several factions, compare performance of top 5% and bottom 5% players in each faction (skill ceiling vs. skill floor), look at mean vs. median or even some more complex correlation analysis.

    The problem is, that these methods are only available to CB, since it requires keeping track of every match result on a per-player basis. And while I am the last person to pass up on some possibly interesting numbers I think it can be misleading to take these numbers at face value without further context.
     
  8. eciu

    eciu Easter worshiper

    Joined:
    Nov 23, 2017
    Messages:
    3,673
    Likes Received:
    4,309
    No, apparently they are not only.
     
  9. inane.imp

    inane.imp Well-Known Member

    Joined:
    Jan 28, 2018
    Messages:
    4,498
    Likes Received:
    5,355
    Well CB and you. The rest of us still don't have access to the full dataset. Given that this is entirely reasonable we will just take you at your word then.
     
    ambisinister and ObviousGray like this.
  10. Death

    Death Well-Known Member

    Joined:
    Nov 28, 2017
    Messages:
    1,068
    Likes Received:
    1,439
    Funny how through previews the MO rework created so much excitement while the Invinicble Army was declared dead on arrival. After the update though, its completely switch. Quite a lot of people are having fun playing with IA while MO seems to be a sunk ship for many.

    I don't know what to make of this but I find it amusing.
     
  11. inane.imp

    inane.imp Well-Known Member

    Joined:
    Jan 28, 2018
    Messages:
    4,498
    Likes Received:
    5,355
    IA is a new thing with new models and profiles. There are few existing biases and generates a positive feeling of unfamiliarity.

    MO is a repackaged old thing that did not fundamentally change. It lacks the thrill of the unfamiliar and thoughts about it are quickly subsumed by existing biases.
     
  12. ObviousGray

    ObviousGray Frenzied Mushroom
    Warcor

    Joined:
    Nov 27, 2017
    Messages:
    1,629
    Likes Received:
    2,768
    There were people tried to convince people about IA's awesomeness before, but lets just fold it elsewhere. Not all YJ forumers were salty. That I wanted to say.
     
  13. eciu

    eciu Easter worshiper

    Joined:
    Nov 23, 2017
    Messages:
    3,673
    Likes Received:
    4,309
    I can only speak for myself but I've completly missed the fact of Fusilier departure. All my MO lists were starting list building with 3 Fusiliers and then picking up usual stuff (funny enough Fusilier were even more needed than Hosp+Mag link ^^).

    See you on in other factions/sectorials :)

    Again, you are not exactly careful-reader :)
     
  14. inane.imp

    inane.imp Well-Known Member

    Joined:
    Jan 28, 2018
    Messages:
    4,498
    Likes Received:
    5,355
    Clearly I missed you saying that the first time. Usually I'm better at noticing personal attacks on reading comprehension. Perhaps if you were less deliberately obtuse I'd know what you're on about?
     
    ambisinister likes this.
  15. eciu

    eciu Easter worshiper

    Joined:
    Nov 23, 2017
    Messages:
    3,673
    Likes Received:
    4,309
    I find really interesint you would consider rather polite "careful-reader" as a personal attack, but hey we are on internet.

    What I wanted to say is that I've answered the question which you have asked. When I have something more to post I will surely do that :)
     
  16. inane.imp

    inane.imp Well-Known Member

    Joined:
    Jan 28, 2018
    Messages:
    4,498
    Likes Received:
    5,355
    Hahahaha. Just because it's politely phrased does not mean questioning some one's abilities isn't a personal attack. It doesn't phase me, but in particular getting uppity about reading comprehension is a thing on these forums for some reason (there's been several Storm 'interventions' for only slightly more harshly phrased versions of 'you're not a careful reader').

    You really haven't. Which, as I thought I'd rather made clear, was "put up or shut up". Where is your sources?
     
    Judge Dredd likes this.
  17. CoffeeBadger

    CoffeeBadger Well-Known Member

    Joined:
    Aug 6, 2018
    Messages:
    38
    Likes Received:
    50
    Maybe, however that depends on the format of the raw data avaliable. I could not discern this from the sample you provided, since these were already processed data.

    What I am talking about would be a data cube containing: player ID, faction, tournament ID, match ID, match result.
     
  18. Barrogh

    Barrogh Well-Known Member

    Joined:
    Nov 26, 2017
    Messages:
    1,954
    Likes Received:
    1,670
    Being honest, I don't see a point of spending extra effort to make screenshots and past them here, while jumping through an extra hoop of cropping and marking the resulting pic instead of just providing the link, which is easier and giving more information, so that would be a win-win.

    Unless, of course, you're trying to obfuscate something.

    Don't get me wrong, I'm usually supportive of people who take an important (IMHO; no sarcasm) function of being bitter upon themselves, and with me strongly affected by my local meta that characterizes an attempt to design an army mostly around single combat group as "cute" at best, I'm not instinctively buying into idea that modern MO has a good variety of competitive lists (although what do I know). But this kind of argumentation convinces nobody and only makes people suspicious.
     
    Judge Dredd likes this.
  19. fari

    fari CRISTASOL, EL LIQUIDO DE LOS DIOSES

    Joined:
    Feb 21, 2017
    Messages:
    2,586
    Likes Received:
    2,430
    Now we can compare the MO upgrade to the new NCA upgrade.
     
  20. eciu

    eciu Easter worshiper

    Joined:
    Nov 23, 2017
    Messages:
    3,673
    Likes Received:
    4,309
    I think that state of Black Friar is a good summary of both xD


    Well I thought that CB won't dare to put even more power to CSU's.

    I was wrong (again).

    Apparently most armies I play now will have obligatory max AVA of CSU's from now on (tbh. they already had).