Jump to content

A Better Elo


44 replies to this topic

#41 Void Angel

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 7,487 posts
  • LocationParanoiaville

Posted 15 January 2014 - 01:30 AM

No, they really don't. Yes, going from bird wings to 'mech arms can be disorienting. But once you learn to pilot each type of 'mech, the skills transfer. I can pilot both kinds just from having learned the Spider and the Jenner, for instance - there's more to piloting a Thunderbuddy than how its arms work, of course, but that one thing correlates directly. The Thunderbolt is an amazing damage spreader, but I learned to spread damage in the first place on my Atlas. The more 'mech styles you learn, the less picking up a new chassis should confuse you. There are differences, yes - that's why you pick up different chassis - but they're not extreme once you've gotten a good mix of 'mechs in your bays.

Then there's all the things you learn by playing that aren't chassis specific - map knowledge, tactical practices, knowing how the Puggles are likely to react. Knowing how hot different weapons and maps are, without having to think about it; how to manage heat when the pressure's on; how to move at different speeds with different movement profiles; how to spread damage. How to shoot, use cover, and be aware of the game world around you instead of just staring downrange... all of these things are transferable skills. Even accepting your claim about first impressions for the sake of argument, a chassis-specific Elo wouldn't have the effect you expect. Quite the opposite, in fact; if you put me in a brand new Highlander and turn me loose on some hapless newbies, I would murder them. I would smash them so badly that they would never again mistake a merely bad game for a true Elo mismatch. I would crush them so completely, they would feel it in their very souls; their children would be born with despair on their lips to the seventh generation. Seriously, hyperbole aside, turning veterans loose on newbies would have the exact opposite effect from "giving every chassis a good impression."

#42 Ghogiel

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • CS 2021 Gold Champ
  • CS 2021 Gold Champ
  • 6,852 posts

Posted 15 January 2014 - 02:25 AM

View PostWintersdark, on 12 January 2014 - 10:12 AM, said:

Sorry, I call ********.

I haven't seen a failed to find match screen on a solo drop in any weight class in months and months.

lol?

I'll get failed to find matches in solo assaults every single day bar right after Tuesday patch and prime times on weekends.

Last time I solo'd in my assualts I was sitting in comstar updating my user name with how many failed to finds, my record that day was 13 failed to find in a row, then it was like 8, then 8 again, etc. The very next day I hit like 6 or 7 then just closed MWO.

#43 Wintersdark

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • 13,375 posts
  • Google+: Link
  • Twitter: Link
  • LocationCalgary, AB

Posted 15 January 2014 - 05:47 AM

What Void said.

I started when open beta first started, one of the firs influx of open beta newbies. I experienced what its like to be tossed into matches with veterans first hand, and that's one of the reason I'm so adamant about that. I had tons of Mechwarrior experience - years of playing all the games, so I understood how it worked and all the keybinds right away (so a huge advantage over most newbies these days)... But I didn't know the maps. I didn't have an intuitive grasp of the heat system. I didn't know where the hitboxes on the enemy mechs where, so it was very difficult to target specific spots effectively.

The close beta vets gently bent me over and... Well, to avoid breaking the censor, suffice to say that they pounded mebinto the dirt, over and over and over. I think it took me almost a month to get my first (legit, non-stolen) kill.

And that's that. Like void said, put me in a match against new players in any mech, whether I know it or not, and I'll do the same to them. The vast majority of player skill is utterly unrelated to the specific chassis. Most of those things you never think about. Its all just instinctive.

And, as you get more experience, the differences from chassis to chassis mean less and less. They play different, feel different, but ultimately there are very few really unique mechs.

#44 Little Details

    Member

  • PipPipPipPipPip
  • Moderate Giver
  • 172 posts
  • LocationSt Louis, MO, USA

Posted 15 January 2014 - 12:30 PM

Then the pure Elo approach isn't the best means as it currently stands if your primary reason is that vets would be grouped with people brand new to the game unless we had a statistically significant number of games in a chassis. I completely agree with your point on newb vs vet - it took me WEEKS to feel comfortable and I was in the same boat as you playing all the previous MWO games and got crushed - I only kept playing because I loved this type of game growing up and I love the franchise and had the baseline understanding of WHY I was doing badly until I could learn the specifics you mentioned above.

Someone had an idea about using an avg weight class score until X number of games are played per chassis which has merit. Or, without having specific categorized 'skill levels' of players clearly delineated (which is another bad idea for ways of stratifying the player base), there can be an arbitrary new player vs veteran player flag after X-hundred(s) of games to seperate those that are new to the chassis vs those that are new to the game when the MM first compiles teams.

While your points basically summarize to avg pilots are avg pilots, great pilots are great, etc etc no matter the machine, and I DO understand that and even agree with it to an extent, I just don't think it's that black and white. There are adjustments that can be made on a more individualized basis per player skill in a particular mech that would A) produced a better overall game for both teams and b: allow greater diversity in getting away from FOTM as the end-all-be-all and make those other 60 mechs in our mech bay get out in the sunshine a little more.

Even with 5k games under my belt now, I just bought atlases and while I know the top 3 general builds for each, without having them elited you don't know exactly how they're going to handle, or you fiddle with 4ML vs 2LL... or maybe 2ERLL once you have elite, etc. If I don't know the final loadout I'll use, or the type of tactics I want to end up employing in a mech, there is zero chance a middling player will going from a cic to hbk to shd. Their mind = blown for the staggeringly different ways to play those 3 even though they're all mediums.

Edited by LT Satisfactory, 15 January 2014 - 12:40 PM.


#45 Void Angel

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 7,487 posts
  • LocationParanoiaville

Posted 15 January 2014 - 03:22 PM

Read this Article.

Now tell me what hypothetical system you are proposing to replace Elo.

Now tell me how that system would shield newbies from the upper half of the rating distribution in a way that Elo cannot.

Since you bring a large body of skill to any match which is not chassis specific, any system that allows you to restart at a ground-floor level of rating will fail to shield newbies from you. Sandboxing new players helps, but once they get out of the sandbox, they're going to start running into people like you and me. The difference in a per-chassis Elo rating is that they're going to be doing that sooner rather than later - much sooner. And the game is difficult for the uninitiated; it didn't really feel that way to me, but I'm one of the guys that's played every commercially released Mechwarrior game there ever was. The biggest challenge I had was learning about lag shields and training myself to overcome them. Statistically, though, a lot of people find the game hard going at first.

Someone whose primary gaming background consists of Madden NFL, Call of Duty, and Soul Calibur is simply not going to have that background - they may learn quickly, or they may learn slowly. But the stats and feedback all show that they have a lot to learn. If you sandbox them, that eases the transition, and it's probably a good idea (I think they are doing that with the UI 2.0 rewrite.) But they're learning the game in a near-vacuum with other newbies; when they hit the general player base, it's going to hurt a bit, no matter what. If they have to deal with upper-level players who bought a new 'mech to try along with it, it's going to feel like getting strapped to a table and having their sensitive parts beaten with a baseball bat.

Again, I simply cannot believe that your characterization of chassis differences are accurate. Each chassis is different, but that difference isn't mind-blowing; it's not intense. I found going from a hunchback to a Shadowhawk to be an easy transition; not because I'm the Medium Master, but because there are a lot fewer battlefield roles than there are chassis. Think about it: you can snipe with big, slow-firing weapons; you can use rapid-fire guns that require line of sight; you can use LRMs; you can brawl; and you can play stab-and-fade. That's really it: sniper,dakka,Lurms,brawler,stabber - you can add "light hunter" if you really want, but that's more of a stabber variant, outside of bullying that one weight class. Combine those with the four weight classes (Cicadas are lights for this purpose) and you have all the combinations possible. Players don't have to learn every new chassis from scratch - once they learn how to build and use a sniper, they can use that role in a Raven as well as a Cataphract. Once they learn how to build a Raven, they can use it as any role they know which it supports. Which roles a variant of a chassis does well will become increasingly obvious to them as their general skills improve as well. There are ways they can still fail; bad builds, bad habits, wrong practices... But once they learn the basics, chassis quirks just don't matter as much - sure, the hunchback has a MUCH larger torso yaw than the Shadowhawk, but you can still use them both the same way if you desire.

In the end, while the game is challenging and diverse in player options, it simply does not inflict or exhibit the psyche-shattering complexity and hellish bewilderment you describe.

PS: In any system, you'll have people starting new accounts to take advantage of new players because they are weak players themselves help out their friends who are just starting. But MWO disincentivizes this practice in many ways - the more you play your account, the more you lose if you start over.





2 user(s) are reading this topic

0 members, 2 guests, 0 anonymous users