Imrpoved ranking system

[imbaczek]

also, autohosts could gather more detailed rankings and achievements, I believe Licho is planning something like this. any form of is at the very least pointless and quite possibly dangerous as neddie said.

[Gota]

Do not discuss user friendliness in the forums.There are only a few people who will even listen.
Most will just tell you everything is fine and your wrong.
"spring is the most noob friendly game i know".

isn't the ladder broken anyway?
I remember Neddie saying something about him doing some work on it... .... ....

no clan ladders,no user profiles,smurfs everywhere...oh yeah...we are set to go.

How about you suggest a system that might work rather than whine about how bad the current one is? It's not like everyone hates the idea of a new ranking system, but replacing the current system with a new system that will be equally flawed is just pointless.

I have.

The ones that think a ranking system is bullshit are the ones that hardly play..seriously..
This is all BS talk..It is obvious that to take the few first steps of improvment demands no discussion nor Neddie's Walls of text.
#1 fight smurfing.
#2 make player profiles iwht info about the player like most played maps etc..ah..this has all been discussed about in Neddie's thread...
As you can see nothing has been done..
No need to think where you will be in 20 years,lets start by deciding what to do tomorrow.

[ZellSF]

#1 fight smurfing.

Which can only be effectively done by blocking out people playing on the same IP (which can lead to people who aren't smurfs being unable to play), and I guess people will still be able to play using proxies? People without static IPs gets to smurf as much as they want anyway.

#2 make player profiles iwht info about the player like most played maps etc..

Don't see how this will make for a better ranking system.

This is all BS talk..It is obvious that to take the few first steps of improvment demands no discussion nor Neddie's Walls of text.

No it doesn't demand discussion, it demands a description of a superior system that addresses enough flaws to make the change worth the effort.

[Pressure Line]

im not convinced of the need for any ranking system more complicated than:

in #main etc you have the aggregate total of time for all mods as your rank. in a battle window you have your total time for that mod/game as your rank.

if you want some awesomesauce ladder and ranking system that gives the time and date of every match, and records detailed win/loss/map/faction/team info go code it, if its any good it'll be used.

anything other than an indicator of "is the other guy on my team someone who has only been playing spring for 1 afternoon or a pro" isnt really necessary given the size of the current Spring community. as for smurfing, Gota, you need to remember your own history, or i will go dig up that thread from last year, its really not *that* big a deal (although if a decent way of stopping it can be implemented im all for it)

Spring needs players to survive, not a bunch of elitist douchebags running around waving their epeens at everyone. Remember, everyone has a different idea of what constitues an enjoyable gaming experience, otherwise we'd all just be playing XTA on Small Divide

[Tribulexrenamed]

There is no reason to do something without a reason.

Before saying how the system could be improved, please explain to me WHY the system needs to be improved. Because I dont understand what the problem is that needs a solution.

[HyperLethargy]

There is no reason to do something without a reason.

Before saying how the system could be improved, please explain to me WHY the system needs to be improved. Because I dont understand what the problem is that needs a solution.

Right now it doesn't really represent a players actual skill. A lot of us just aren't that good at the game and it's no fun if there's no chance of getting anything near a 1:1 win:loss ratio. There's people that win constantly and people that lose constantly. Even if most of the people that lose constantly are new to the game, it's no fun for them and they leave.

[ZellSF]

But there aren't any replacement systems which can show how skilled a player is because how random team games are...

[Pressure Line]

There is no reason to do something without a reason.

Before saying how the system could be improved, please explain to me WHY the system needs to be improved. Because I dont understand what the problem is that needs a solution.

Right now it doesn't really represent a players actual skill. A lot of us just aren't that good at the game and it's no fun if there's no chance of getting anything near a 1:1 win:loss ratio. There's people that win constantly and people that lose constantly. Even if most of the people that lose constantly are new to the game, it's no fun for them and they leave.

how will a new ranking system fix that? (and i lose a lot, and im still here 3 years later...) Ranking systems do not automatically make people better.

The current one works as a basic indicator of general game knowledge, it allows you to see who is likely to build windmills on Comet Catcher (lol no wind) and who is likely to be a useful team member.

I have not seen any convincing reason as to why it should be changed, ingame time as as good an indication of general skill as any other.

[Wisse]

I find this sort of rankings perfectly fine. There's no way someone could make fair ranking system. You can't just go and log win/lose statistics couse than I'll never ever play rANDY and such again. That would just ruin my statistics although I could theoretically be 2nd best BA player on server.

Ppl know ppl here in spring so we actually know what to expect from ur opponent. Current rankings also give you quite good idea what to expect from those who's names you don't know (unless they're smurfs).

My solution to this would be implementing better support for clans, because it's easier to remember 1 clan name than 10 players names. With better support clans would care more about their average skills which would allow other players to rank clans in their mind like we rank each individual person atm.

I'm not sure how many guys out there don't wanna play in clan at all and want to stay loners joining public mix games. I just know that everyone was going after good clan in Wolfenstein:ET and with good clan came known tag and reputation/skill expectancy. Community was sorting itself out. When you wanted a match you just used mIRC, joined #6on6.et shouted that u need "mid" skill opponent and you actually got to play someone who was in ur skill category. When you got better you requested mid+ and so on. High skilled clans were all more or less known too, so you didn't get raped without choosing to. Plus they always told you they find themselves part of high skilled bunch.

And to be honest good players were always searching for good allys which meant they formed/joined clan at some point. All you're left than are solo players who's skill can be judged by their current rank.

[HyperLethargy]

There is no reason to do something without a reason.

Before saying how the system could be improved, please explain to me WHY the system needs to be improved. Because I dont understand what the problem is that needs a solution.

Right now it doesn't really represent a players actual skill. A lot of us just aren't that good at the game and it's no fun if there's no chance of getting anything near a 1:1 win:loss ratio. There's people that win constantly and people that lose constantly. Even if most of the people that lose constantly are new to the game, it's no fun for them and they leave.

how will a new ranking system fix that? (and i lose a lot, and im still here 3 years later...) Ranking systems do not automatically make people better.

The current one works as a basic indicator of general game knowledge, it allows you to see who is likely to build windmills on Comet Catcher (lol no wind) and who is likely to be a useful team member.

I have not seen any convincing reason as to why it should be changed, ingame time as as good an indication of general skill as any other.

it doesn't make people better. the goal is to find people, or games that are most likely to result in a draw, kind of like console matchmaking. bad players fight bad players and good players fight good players.

I find this sort of rankings perfectly fine. There's no way someone could make fair ranking system. You can't just go and log win/lose statistics couse than I'll never ever play rANDY and such again. That would just ruin my statistics although I could theoretically be 2nd best BA player on server.

It's pretty easy to make a fair ranking system, and it wouldn't be forced or anonymous matchmaking. we could just use traffic lights next to games in the lobby: green light for 90%+ expectations of a draw (if draws were possible), amber for 75%-89% and red for less than 75%. if numbers weren't shown it would cut down on smurfing and make bullying impossible.

But there aren't any replacement systems which can show how skilled a player is because how random team games are...

this might be a problem as team games are made fair by mixing teams. maybe a seperate score could be kept for team games, or have the rank aggregate over several matches calculated to come up with a true rank. if scores are calculated from the players last 5 games patterns based on skill will emerge (being on the winning side 3 or 4 out of 5 games probably does have a little to do with skill, whereas it's hard to tell with 2 out of 3)

[imbaczek]

every ranking system can be gamed. the current has an advantage that everybody knows it sucks, so nobody expects anything from it. when per mod/per map rankings get implemented, it'll be better, but still next to useless, so no point in gaming it.

[hoijui]

please lock!
it is starts to get on my nerves.

[ZellSF]

being on the winning side 3 or 4 out of 5 games probably does have a little to do with skill, whereas it's hard to tell with 2 out of 3

In team games, one person usually can't make that much of a difference, so your method doesn't work. Unless you're talking about very small team games.

[Pressure Line]

unless you are the most useless person on the team, who builds 3 mexes, 2 solars, a kbot lab and starts pumping out fleas. in which case if the team loses its probably your fault, and if the team wins its probably not because of your outstanding contribution ^_^

[aegis]

hey, I've won games with fleas (and bladewings)

[REVENGE]

hey, I've won games with fleas (and bladewings)

Mmmm <3 Fleas and Bladeys.

[ginekolog]

meh current system is quite ok as stars and goldstart are usually good in all mods once they master one. Even if Randy have no clue about not and he plays first time and i tell him spam storm/thud + dont guard factores he ll win most players :)

Per mod stat would be necessary if we had more mods that have similar popularity and would be VERY different. This is not a case in spring as mods are similar .

I trust goldstar playes even with my AK's life ffs

[Pressure Line]

I trust goldstar playes even with my AK's life ffs

But would you trust a player who had a gold star but had only ever played SW:IW, S44 or E&E?

Rankings separated by mod are a "nice to have" not an "omg we need this now its ruining our games."

Anything else (win-loss rankings, ladder-style rankings etc) needs to be opt-in, and really imo doesnt need lobby integration more than what is already present (gold/silver/bronze trophies for the top 3 of each ladder)

[Ashnal]

I think the hidden rankings system with green/yellow/red is the way to go. Some adaptation of this would allow better matchmaking with less epeen waving.

Since its a color system based on ranking hidden from the users, without a great deal of effort it would be very hard for the average player to figure out where he stands overall in the community rankings, due to his ranking being relative to all others and vice versa. It would result in better matching of players without out giving players statistics to wave their epeens in front of.

[Error323]

Hi everyone,

Introduction
Having read this thread and being aware of the issues involved with ranking systems, I would still like to share my proposal for one. Here we go:

Proposal
Within the Spring engine, after a game has ended the statistics (units killed, damage dealt etc.) would be posted to a webserver via e.g. libcurl by the Spring server and each Spring client. These statistics will be matched against eachother to prevent tampering with the source by a single player.

The post data will be dumped in a mysql database and the ranking is done on the webserver (not in the Spring source) so that the precise ranking formula is unknown to the main public.

Since just raw data is posted by the Spring engine, all kinds of cool statistics can be drawn from it on the webserver like "Ecowhore of the month" "Highest potential RSI patient of the month" . Even each game could be tracked, the demofile could be posted aswell and we could have graphical stats of the games played and assign star rankings to certain demo's. Also clan-based statistics per mod could be relatively easy tracked.

All this should be done for each mod available, and perhaps a trustee for each mod could help deciding on a ranking formula since this may differ per mod. And I'm only experienced in XTA.

Now the coolest part would be that the system can be fully automated for players. If I can get access to the Spring lobby server so that ppl don't need a new account, I can build something really cool.

Rank tampering prevention
As someone else stated already you can always tamper with the ranking system. The ways of this proposal to prevent this are:

• Submission of statistics per game by each Spring client and the server.
• Rankingsystem not in Spring src, but the webserver.
• Ranking adjusts relative to opponent(s) ranking(s).
• Only submit games lasting longer than time T, where T is defined per mod.

This leaves (afaik) 1: the smurfing issue, which I don't really see as a problem since a smurfer would just take longer to gain ranking then his/her actual useraccount. 2: The play against self issue, which can be `solved' by also storing the ipaddress (gained by webserver, not in the Spring source) and using the unitactivity stats.

Actual ranking formula
Given all the stats by Spring and designing a few new ones, one can design a pretty robust ranking formula. This part will need some further research on my end, I will let you know what I come up with when the time comes. But first I would like to hear your thoughts on the proposal so far.

Conclusion
I think this could really work, and be even more than just a ranking system. And since the engine would just send raw data to the webserver a lot more stats can be addressed by being creative :). If you find more ways on howto tamper with it, besides posting the issue(s), please also try to think of the solution(s).