1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
  2. Hey Guest, is it this your first time on the forums?

    Visit the Beginner's Box

    Introduce yourself, read some of the ins and outs of the community, access to useful links and information.

    Dismiss Notice

Kill:Death Ratio

Discussion in 'Community Dev Corner' started by Antman, Jun 15, 2012.

Mods: Downburst, Mazey

  1. Oh, in a good server on a crowded map I've reached 5:1 kd as an Archer.
     
  2. FliesLikeABrick

    FliesLikeABrick THD Team THD Team Administrator Global Moderator

    Messages:
    952
    right, K:D will not be tracked by any official statistics - but we can't control what individual server owners do. This thread might be more helpful if we instead started talking about other interesting statistics that you would like to see (blocks placed, arrows fired, etc)
     
  3. Antman

    Antman Base Burner

    Messages:
    344
    Will just kills be tracked?
     
  4. Lord_bugg

    Lord_bugg Smashing Donator

    Messages:
    241
    Thing is though, if clans use kd's to determine who joins it will mean that people who are good players and work as a team but have a bad kd wont be picked over someone who is rubbish at team work but has a high kd.
     
    kl4060 likes this.
  5. FliesLikeABrick

    FliesLikeABrick THD Team THD Team Administrator Global Moderator

    Messages:
    952
    I'm pretty sure we're all agreeing here - nobody wants K:D recorded (generally speaking)
     
    UnnamedPlayer likes this.
  6. i do want everything recorded but with option to make it public or private. those who care will have a chance to show off and those who doesnt wouldnt have to be ashamed
     
  7. Nighthawk

    Nighthawk gaurenteed shitter

    Messages:
    793
    Crowded map. I've played in one too. Knights literally seem to walk into your arrows. It's just a killfest once you have near 20 people on each team.
     
  8. Dett11

    Dett11 Shipwright

    Messages:
    16
    Ok, i did just read the start of the thread an then wrote an entry, sorry my fault.
    </br>--- merged: Sep 4, 2012 8:54 PM ---</br>

    but then again people would ask where are your kd? and what u do then

    it would still be a competetive pressure and could hinder teamplay
     
  9. tell them to fuck off
    i cant see how it would hinder teamplay. i think it would encourage teamplay
     
  10. thebonesauce

    thebonesauce All life begins and ends with Nu Staff Alumni
    1. MOLEing Over Large Estates - [MOLE]
    2. The Ivory Tower of Grammar-Nazis

    Messages:
    2,554
    Personally, I love it when I'm playing RTDM and see those kids who camp and throw bombs in an attempt to beef up their kills, then suddenly, next match, BAM! It bugs out and they're like, "WHICH ADMIN SCREWED WITH MY KD?!"
     
  11. Brilly

    Brilly Haxor

    Messages:
    222
    No, I'm pretty sure there's a lot of people here against censorship.

    Having much more detailed stats would solve the problem of users trying to artificially bump their k/d up, because we can mock their attempts if we just had a little more information.

    For example, knowing the time spent as knight, archer, or builder, stats about position on the course, idle time, time spent falling.

    All of these statistics can be used to improve and find errors in your own gameplay. To not list the reasons why a person is dying is only going to prevent the player from improving his game.


    Here's a stat that might be interesting for knights, (# of sword kills)/(# of bomb kills)

    That number tells me right away information about who is a serious knight, and who isn't.

    Kill streaks also give me useful information about the quality of the knight.


    If you want players acting in a certain way, please show them a global statistic to measure against.


    There are those who fight for glory, and those who fight to win. Finding a way to tell the difference between those two types of people using only statistics should be your goal.

    Please consider it.
     
  12. Downburst

    Downburst Mindblown Global Moderator Forum Moderator Donator Tester

    Messages:
    1,813
    You can't say that's censorship just like you can't say it's censorship not to make some surveillance video public.

    Anyway, stats will never show you who's a decent player, as people will just tend to play on servers with equally skilled people in the end.
     
    Contrary likes this.
  13. FliesLikeABrick

    FliesLikeABrick THD Team THD Team Administrator Global Moderator

    Messages:
    952
    Brilly - yeah I was gonna say... this isn't at all censorship. I'm just saying that we won't be including deaths in the public stats gathered for all servers. Individual server owners can still do it however they want.

    Some of your ideas are similar to what I've been thinking about doing and what I'm looking for, so thanks (like the time spent as each class, that's a major one I want to track).

    At some point soon I'll post a list of stats that I am thinking of.
     
  14. frink

    frink Shopkeep Stealer

    Messages:
    40
    I just want to pipe in and say I'd love an increased level of detail in the server logs. I'd like a log event when a player selects a new class or chooses a team, touches/drops/captures a flag, builds a workshop ... Etc.
    It may be excessive to record each stab/shot, since it already takes hours to parse a months worth of data. But since I mainly play builder, I find it slightly frustrating that I can't statistically recognize the best builders.

    K/D whores are super annoying and it would be great if I could see when a player performs an honorable action, like joining the losing team. Or switching to builder and building workshops. Or capturing the flag instead of spawn camping.
     
  15. Contrary

    Contrary The Audacious Paramount of Explosive Flight Donator Tester

    Messages:
    2,196
    I think trying to determine "who is the best" or to give a numerical value to a player's proficiency is an inherently flawed goal. Especially with the dominant KAG format being a team-based game mode, the worth of any given player is just too complex to quantify, even if success wasn't so subjective. Global tracking for KD would create a lot of negativity and really impede player growth by discouraging them from attempting new tactics or engaging in actions which disadvantage them to benefit their team.

    Generally speaking, KD is a really poor metric of player skill and is taken way too seriously.

    Though that's not to say that it is necessarily bad always. Maybe KD that only factors in your last week or so of playing? Maybe restricted to more KD oriented servers like RDM or 1v1?
     
    justi01, Lord_bugg and thebonesauce like this.
  16. thebonesauce

    thebonesauce All life begins and ends with Nu Staff Alumni
    1. MOLEing Over Large Estates - [MOLE]
    2. The Ivory Tower of Grammar-Nazis

    Messages:
    2,554
    But not even RTDM since that'll promote people to bomb jump on top of a hill and then slash people as they try to get up. Or immediately pull the "Player is now spectating - Player has joined the red team." bullshit to avoid death. KD is bad enough. Casuals think it's a true measure of skill, while the serious players have to deal with camping douchebags who rely on entombing, running the clock, etc. And I'm not saying "serious" as in, "OMG SO PRO," but when I want to play KAG, I WANNA PLAY. Especially since I barely ever have time anymore. I go on an RTDM server for a quick match before I leave for school in the morning, and I have to deal with all the shit I just referred to. Can you imagine how much worse it would be? I just wanna slay some doods, not sit and let a clock run down because some kid with waaaay too much time on their hands learned to bomb jump to the top of an unclimbable tower instead of using the bomb to fight. nowutimeen?
     
  17. Brilly

    Brilly Haxor

    Messages:
    222
    Stats can tell you a greeeeeat deal. Basic mathematic principles would disagree with you.

    But most of you seem to be playing with many immature kids who only care about K/D.

    What I am saying is to find a better metric to measure the skills of players.


    I'm saying that you should create a situation where someone can say, "Your K/D is decent, but this other stat shows that you are a horrible team player."

    Savvy?
    </br>--- merged: Sep 6, 2012 2:48 AM ---</br>
    Yes, there needs to be a recognition system that can identify the towers or pits builders make, and award points based on how it slows down the enemy players. You can get that from position of the players, and time data.

    I would recommend a sort of fuzzy logic interpreter that takes in knowledge of position of the players, time spent in an area next to the tower, and an ability to recognize commonly used modules that may be present in the tower.

    If a builder places a spike that kills someone, the builder needs to get recognition somehow.

    A spike identifier that records the time it was placed, who placed it, the number of players it has killed, and its position on the map.

    That's the gist anyway.


    And one more thing, putting all the responsibility on individual server owners is a hassle, because like Frink says, it takes many hours to parse this data from just logs, and global variables would ease this processing ever so much.


    Can't you just hide the stats, but make them available to server owners?

    Does that solve your problem?
     
  18. I agree with this. Although spikes are only one weapon in builder's arsenal they are easy to track, so itsa good start.
    that is not the case actually. on modern computers parsing logs from 3 months is a matter of seconds
     
  19. Dett11

    Dett11 Shipwright

    Messages:
    16
    The problems lies within that, tracking kills and deaths and measuring kd is easy. Measuring teamplay on the other hand, that is unbeleavible hard, if not impossible. You could have a score system, which many games have, but usually the one with the most kills has the highest score anyway, so how is that going to measue teamplay?

    Aswell, tracking numerous stats may be hard on the servers, i dont know. Just my opinions.
     
  20. frink

    frink Shopkeep Stealer

    Messages:
    40
    It varies depending on the depth of the analysis. For my statistics, I record the time, killer, killee, and weapon of each kill in a relational database. Which is why it takes hours for me to parse months of logs. I know I'll be able to optimize it further, but I'm only one man and I have many goals.

    Of course, the KAG developers would probably only track aggregated kills rather than each individual event. And even if they tracked each individual event, they would probably find a much more efficient way than my proof of concept.
     
Mods: Downburst, Mazey