Skill levels system

(rookie1) #1

what I would like is if the game had a skills evaluating system :
eg: before first game play an evaluation of your skill level can be done with bots base on your aiming movements etc…
Says the system evaluate me an level 3 skill and servers are designated with skill number ,I wont go on a server skill 10 .
Ill try to join a server of my skill or use an auto server joining system for my skill.
Say after X hours of play the system could suggest me to redo the evalution (if I didn’t choice to do it manually previously) to reevaluted my skill.
mb some brainstorming this could extend the possibilities like at level 3 playing with a bot level 15 …ect…

For sure if a guy go ingame and he pwd by all he wont come back in a live server but give them a chance to improve at his pace and you might keep him for long time

(spookify) #2

That is what SD’s rank system is for… However if you disallow some one to join a server because they are to good that is just criminal…

I hate the idea of ranking, splitting or segregating players based on skill. IMO this is how ET died at least for me… There was only like 2 USA servers that had ET Pro 2.60b and once those died people like me went looking for other server possible not with ET Pro… As soon as we got to these other servers and found a smaller one that was 8 v 8 we are almost insta banned for “hacking”… Soooo Then you try the larger 32 or 64 man servers… Again BANNED! It was complete bs and drove away hundreds of players.

This biggest this to remember is the bigger you make the game and the more play options there are the more narrow the game gets.

COD Black Ops 2 could have 2 Million people playing at one time however there are 50 different game options and I only like 3 of them… I would wait to Q over an hour for Capture the Flag or Check point… It took so long I quit… Millions of people playing and 3 game type were completely useless after 3 month of game release…

Give people options is nice but dont give them to many… Get them into the game and if they suck give them a class or roll or something on the map that will keep them busy.

(Protekt1) #3

That could have been a networking issue.

(Mustang) #4

We discussed an automatic skill algorithm based on things like player stats a while back, I think this is a better approach than having to retake a test everytime.

(amazinglarry) #5

Is a player’s skill really reliably quantifiable in a way that we can break individual players into different levels? There’s no doubt that every player has a certain skill level, but it’s largely subjective, isn’t it?

I mean sure, we can knock out a large percentage of the measurable skill based on weapon accuracy and kdr, but when we get to the finer details there’s a hell of a lot more to consider.

I’m just gonna list some examples in bullet-ish form:

  • Level of competition on the opposing team
  • Level of competition on your allied team
  • How often one is an offensive / defensive player on certain maps (and what classes they favor)
  • Damage dealt vs kills or assists
  • Damage taken vs deaths (which should include a bunch of factors like how often do you self-heal / take advantage of your character skils / successfully escape the situation / find a medic
  • Where damage is taken or given (which should indicate situational awareness)
  • Objective attempts:completion ratio
  • Objective attempts:total play time ratio
  • blah blah blah

I’m probably over-complicating this a bit in the sake of this thread, but I wouldn’t want to see players who may be lacking in a few of the more tangible areas (accuracy / kdr) but who are great at running support or objectives et al to be placed completely out of some skill tier.

I’ve found that usually to get better in anything you obviously need practice but you need to be exposed to greater competition as well.

(Mustang) #6

Of course, that’s why you’d include things like only increasing skill level if you beat (either individual kill, or team win, or both) a player/team higher skilled that yourself. And conversely only decreasing if you lose to a less skilled team/player.

(amazinglarry) #7

Well yeah, but I don’t want to be arbitrarily moved up a skill tier because I managed to back-rage the best player on the other team by pure luck a few times in a match.

Naturally I’d be put back in my place the next match, after I get face-rolled I guess. I just meant to point out that there’s a billion variables potentially involved.

(Humate) #8

I’m just gonna list some examples in bullet-ish form:

  • blah blah blah

I’ll add two more:
*how do they play without voip?
*how much actual effort are they putting in a game on average?

That aside, the rating only needs to be a rough approximation so these things are pretty insignificant.

(rookie1) #9

it doesn’t have to be a perfect science but a guide .If the level 3 want to join a level 10 server its up to him .
yes he can be level 3 only but excel in other aeras eg . on a certain weapons or what ever .
designated server 10 can allow lower level players ,but he will know how skill the server players average is or the host want his server be …
Doesn’t mean he cant pwd few of them .
I would see skill evalution on basic competence like aiming ,movement .k/d …etc …not going with all the variables that define skills .
skill evaluation is better than hours played make a better player .

(amazinglarry) #10

Fair enough - like I said I was probably over-complicating the idea of this thread. I just think that if there is a ‘system’ like this in place, it needs to be clear that it’s something that’s just a general barometer rather than a full ‘skill’ evaluation. :slight_smile:

(INF3RN0) #11

Anything that does a better job of balancing teams is good. Matchmaking doesn’t always mean high levels can’t play with low levels, but it will at least split them up evenly if they are in the same game.

(Mustang) #12

Yea this would be ideal, that way us low skilled players can still play with you good guys and get better, without teams getting stacked.

(rookie1) #13

agreed…

(INF3RN0) #14

NA server was full up tonight, but then died because they couldn’t kill a few veterans. Teams were balanced as well :rolleyes:

(shaftz0r) #15

im sure it died because the server went through one rotation of the maps :stuck_out_tongue:

(INF3RN0) #16

The frustration was vocalized heh. I had a pretty interesting moment of getting 6-7 kills in one clip within a few seconds too lol. Moving around is underrated.

Also tested the shotgun, and it apparently does still kill in 1-2 hits still so no clue what people were on about. Had one fellow at my throat about it and I couldn’t blame em.

(shaftz0r) #17

it isnt “people” just one person (lol) and hes mad because he cant backrage an entire team without getting fragged. i can see some of the point. the low hp is rough, but before you couldnt kill proxi 70% of the time if she was anywhere near you. i think she’s relatively balanced. you would think shotgun whores would just use the engi with the grnades errr sorry “mines”

i wish i had decided to launch tonight. i feel like a lot of population issues would be solved by an external way to see the servers.

(RasteRayzeR) #18

Im not gonna respond about Proxi, I’ve voiced my concerns already.

The skill level system is hard to create because a player that does all the objectives but has a low k/d ratio is gonna be put into the noobs server. Killing is a part of the skill measurement, but you can’t make a ranking system only on that. I guess this will be another wall to bash your head on to find the right balance and features to observe when someone is playing.

I would simply add global statistics about hit rate, k/d ratio, longest spree, avg life time, and each gun’s accuracy + relative k/d ratio and put all that in a quick link menu in-game like in W:ET

Blocking a pro from going to a noob server is called kick vote. Blocking a noob from going to a pro server is unfair if he/she wants to train with the best (you don’t learn much otherwise). And finally, just offer the possibility to get a private server for teams and comp players who don’t want to be disturbed.

Anyway, that’s maybe old-fashioned. Good luck to make a skill evaluation grid !

(INF3RN0) #19

The obj player excuse is obsolete in xT with agnostic objs. I’ve done just as many objs as killing, but in the end win consistency is the most important stat. KPM/DPM should be weighted over KDR though.

(RasteRayzeR) #20

Problem is you take only your own play style into consideration. We don’t all get 10 d/k ratio while we disarm three C4 and drink a beer at the bar at the same time. Try to consider the majority of players.

And win consistency is not an option : you can play the objectives nicely yet still lose because of the team mates you are with, as well as win thanks to them if they can carry you. Consistency in wins is not a real indicator of skill.