Page 4 of 6 FirstFirst ... 2 3 4 5 6 LastLast
Results 31 to 40 of 55

Thread: New performance bars...

  1. #31
    Basic Member altair's Avatar
    Join Date
    Dec 2011
    Location
    Greece
    Posts
    490
    Like we don't already knew our w/l hero ratio with the old bars. Useless is useless..

  2. #32
    Basic Member
    Join Date
    May 2012
    Posts
    54
    This is a joke...

    When I started playing, I really liked the idea of performance bars (that was year ago), they seemed to be okish

    Then they changed them to... something. At first I thought they are broken, because my best hero was death prophet and bloodseeker, and I didn't play bloodseeker that good.

    Then dotabuff released DBR. Finally a way to see my skill!

    Then Valve forced dotabuff to get rid of DBR.

    Ok then, back to hero bars! I started to like them, I think i was really good at death prophet and slark, and I saw it in the bars. There were some mistakes (PA 1.5, Im sure it was too low), but after all, I started to accept them.

    AND NOW THEY CHANGED IT to something completely useless, based ONLY on W/L ratio, they are not even trying to hide it. Ive got the same bar for different heroes that I have W/L 5/4. I mean why? Previous system wasn't perfect, but you could somehow believe that they are not taking W/L into account, but personal stats for each hero. Actually, I believe it was written, that the bar is not influenced by W/L.

    And that's why they changed it to something purely based on W/L.

    Volvo logic

  3. #33
    Basic Member Raging's Avatar
    Join Date
    Sep 2012
    Posts
    212
    Is Valve is trolling us, because DBR was also W/L based? I don't like this.

  4. #34
    Basic Member
    Join Date
    Dec 2011
    Location
    Brasil
    Posts
    872
    The old performance bars made more sense than this, at least if you played a lot of the same hero. It semed to use "end result" stats (total damage, total healing, death avoidance, etc), so statistically, it did make some sense. If someone plays chen 1k times and he is good at him, he will (in the long run, considering good and bad teams) be evaluated correctly. This winrate thing is totally non-hero specific.
    Old performance bars took a while to "kick in", so it was understandable that people with 10 matches with a hero didn't like the evaluation, but it was eye opening for people who spammed heroes. I played a lot of bounty hunter, pudge and chen and it really showed how much room for improvement I had with those heroes.

  5. #35
    Basic Member
    Join Date
    Apr 2012
    Posts
    1
    regardless of what is said, i feel the metric being used is more accurate.

    The old system was pretty terrible in that it only measured your performance based on your k:d ratio compared to your team. It didn't take winning into account at all.
    Omni being my best hero, with a 2:1 win rate, was rated 1/5 bars, worse with wisp.
    Now omni is far better rated at 8.5/10 bars, and wisp with 6/0 record, a perfect 10.

    The old system quite simply rewarded playing carrys, sniping kills, and fountain camping when the team was losing (preventing your death counter increasing), or fountain camping when your team was winning (and sniping kills increasing your counter), Pudge fountain camping was the worst.



    The old system measured based on your skill in avoiding deaths and securing kills.
    The new system measures based on your ability to deliver a victory


    I feel the biggest issue people have is that heroes with few games played can easily have a very incorrect rating, but the old system was worse by far in that it had that same fault, never worked for supports, and rewarded bad sportsmanship.

  6. #36
    I have only 7 heroes without performance bars and I agree, it seems far disproportionately impacted by win/loss. There are a few discrepancies here and there where a hero may have a little more or a little less, but the rule of thumb is: the more wins and fewer losses the higher. I believe the previous system was flawed but more accurate than the new one; I've played DotA since Eul's and I have a good grip on what heroes I'm most skilled with, which was more accurately reflected before than now, where heroes are rated higher for win rate actual performance in those games notwithstanding.

    I have more heroes now than before with 9-10 bars, if I was to be biased I would argue for the new system; I am not.

    As is, it's redundant, simply correlating with the numerical values of win/losses already displayed. Assuming there is some equation calculating the performance bars, win/loss should be less heavily weighed.

  7. #37
    Basic Member
    Join Date
    Feb 2013
    Posts
    4
    You guys are missing a point about this update. Although the actual bars are determined by W/L "supposedly" the order from top to bottom remains the same(Top is best played following down the list to worst played). I think they just changed how the bars work and that is all. Which isn't a bad update I don't believe because now there is 2 variables displayed. Your best performance is still the top hero on your list.

  8. #38
    Basic Member
    Join Date
    Aug 2012
    Posts
    75
    It seems like now it only counts your win percentage.

    I'm a master (full bar) with Shadow Shaman with just 3-1, but I'm just like 75% with one of my top heroes, Furion with 43-25...

    No comments.

  9. #39
    Basic Member
    Join Date
    Jul 2012
    Posts
    360
    If the impact of one player on the game is around 20 % of their team, I'm not sure if it's a good to base the performance on that only. So why not make performance bars a bit more complex and make them consist of around 50 % of KDA-ratio (perhaps weigh assists more for supports and kills for carries) and 50 % of W/L ratio. Also make another tab comparing your performance to other players of the same skill level. That actually tells something that isn't obvious.

  10. #40
    Basic Member
    Join Date
    Dec 2011
    Posts
    11,187
    Quote Originally Posted by altair View Post
    Like we don't already knew our w/l hero ratio with the old bars. Useless is useless..
    exactly. a simple new column for winrate would have sufficed.

    no clue what the order of the rows is based on.

    Quote Originally Posted by Thesiobhan View Post
    TL: DR It's stupid and pointless, go back to the other thing, it wasn't perfect but it was 3 steps above this.
    old performance bars. additional row for winrate. sortable by winrate, gamecount, wincount and performance (maybe also favorite heroes).
    and try to refine the old algorithm to properly reflect your effectiveness (ex: my 70% winrate, 60 game venge was below 2 bars, that is just flat out wrong).
    Last edited by blash365; 02-01-2013 at 08:39 AM.
    Make sure to read the Forum Rules as well as the stickied Threads of the Forum Section you are posting in.

    Contributions i'd like to highlight:
    My Suggestion: Coaching System
    My Sticky: Intended Changes List
    My Challenge: Completely Fixed Hero Challenge: Skywrath Mage

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •