[Search] Search   [Recent Topics] Recent   [Hottest Topics] Hottest   [Login] Login
rated user maps, not for long
Forum Index » New Feature Request
[Avatar]
yemrot

Messages: 27,
Joined: Nov 11, 2011,
Offline

[Avatar]
yemrot

Messages: 27,
Joined: Nov 11, 2011,
Offline
I love the fact user made maps can become rated if enough positive feedback is received, but i have noticed that they do not last long. It seems to me the reason for this is because some players will give a thumbs down if they lose, regardless of how fair/fun the map is. While I cant say I have the perfect solution to this, something should be done to give these maps a chance, otherwise there will never be a new map that lasts for more than a few days.

I was thinking,

1. maybe once a map becomes rated, only players at the 2000 and above level can vote on them (because players at this level have a better understanding of game balance and map fairness)

2. When a map becomes rated, it stays that way for a set amount of time (say a month or so)

3. Only the winning player can vote on map.

I realize these aren't perfect solutions and may not be fair to everyone, but I think we should do something different. Post some suggestions if you have any.
[Avatar]
droidfreak36

Messages: 183,
Joined: Jan 23, 2012,
Offline

[Avatar]
droidfreak36

Messages: 183,
Joined: Jan 23, 2012,
Offline
One major issue is that maps are rated based on fun factor, which isn't a good measure of balance. If a player loses, they usually didn't have much fun. The ideal solution would be for the balance rating to be based off who won/lost in a large sample of games, that way the opinion of the users is not factored in. Therefore to be rated a game should:

1) Have been played many times. I won't speculate as to the the exact number, but it should be enough to make the balance data significant. If it has only been played a few times, the data is probably not valid because it could just be due to the player matchups or other factors.

2) Have a roughly one win/loss ratio for each race. For instance, Khraleans would have won matches 10 and lost 11 matches, Titans would have won 7 lost 8, Sapiens would have won 17 lost 15, indicating that all races have a fair chance at winning. Again, the exact percentage error allowed is up to the developers. Maybe this should be adjusted for other factors such as:
A) Which race(s) they were playing against (Titans vs Saps, Saps vs Khrals, etc), though this would increase the number of plays needed to make the data significant.
B) The skill level (rating) of the players. This could be based on difference in rating (A 2000 player defeating a 1500 player would be considered less significant than a 1500 player beating a 2000 player. In the later case balance is probably way off.) and/or level of rating (A 2000 player beating another 2000 player is more significant than a 1500 player beating another 1500 player, because higher level players win/lose based on map balance more often than lower level players who usually win/lose based on skill difference or luck).

This message was edited 1 time. Last update was at Mar 01, 2012 19:03


DroidFreak, the roboticist who plays as Khrals.

(Or Titans now that I realize how legit they are)
[WWW]
[Avatar]
StarryBlink

Messages: 184,
Joined: Jan 04, 2012,
Offline

[Avatar]
StarryBlink

Messages: 184,
Joined: Jan 04, 2012,
Offline
But I really like using fun factor voted by players.
When you ask players whether they like the map, you're asking them to use their unique human quality to inform you.
It's a precious information you can't get it from any supercomputers.
So, please use it, with a few adjustments to the following human natures :
1. People tend to feel fun if they win, and feel bad if they lose.
2. Feeling from players who play with skill-approximated opponent can better reflect the real quality
than feeling from players who play with skill-differed opponent.



Here is my idea.
1.
Let players vote the map.

2.
If loser vote thumb-up , this positive vote is multiplied by 3.
If winner vote thumb-down , this negative vote is multiplied by 2.

3.
Then multiply each vote again by the skill-approximated factor ( x ).
If the player played with opponent which rating difference less than 100, then x = 2.
If the player played with opponent which rating difference is between 100-200 , then x = 1.
If the player played with opponent which rating difference is between 200-300 , then x = 0.5 .
If the player played with opponent which rating difference is more than 300 , then x = 0.25 .
[Avatar]
droidfreak36

Messages: 183,
Joined: Jan 23, 2012,
Offline

[Avatar]
droidfreak36

Messages: 183,
Joined: Jan 23, 2012,
Offline
I'm not suggesting that the fun factor should be done away with, but I am saying that it should not determine which maps are rated. If you are playing just for fun, then you should be able to sort maps by fun factor to find fun maps, but if a map if fun and unbalanced it should not be rated, because that is unfair. Some quirky user maps are a lot of fun to play, but I wouldn't want to play rated games on them unless I know that they are fair towards all races.

Humans can rate whether a map is fun or not better than a computer, but what I care about for a rated map is whether or not it is fair. I usually have less fun when I get creamed due to an unbalanced map, or even when I win due to a map imbalance.

DroidFreak, the roboticist who plays as Khrals.

(Or Titans now that I realize how legit they are)
[WWW]
[Avatar]
simsverd

Messages: 921,
Joined: Dec 02, 2010,
Location: Bergen, Norway
Offline

[Avatar]
simsverd

Messages: 921,
Joined: Dec 02, 2010,
Location: Bergen, Norway
Offline
many good ideas..

There are principally 3 different ways to make usermade maps rated:

1. Manually
Either the devs, and/or a gruop of experienced players decides based on testing among popular maps
Pros: possibly best way to judge if maps are balanced.
Cons: time consuming, not possible to test all maps, kind of elitist..

2. Voting
Funfactor (like today) and/or a "balance/fairness" vote
Pros: democratic - all can vote
Cons: data not really usable to judge "balance/fairness" unless heavily modified (because of different knowledge of the game based on experience and skill)

3. Statistical data
Based on data from a number of games played with different race combinations.
Pros: automatic process, fair if skill level is also part of the "formula"
Cons: Will probably need many games to get enough data to be relevant (100+?)


I think statistical data is the easiest way to do it, but i also like the idea that you should be able to give feedback in text also, not only up/down or a score..
But i think that there should not be more mandatory choises after a game.. maybe you could comment on maps in the map browser?

Moderator of gamechat and forum. Tourney admin.

Send me a PM here or invite me to a game if you want to ask me something, suggest a map for tourney or just wanna play a game
[Email]
[Avatar]
PurpleLlama

Messages: 31,
Joined: Apr 21, 2013,
Offline

[Avatar]
PurpleLlama

Messages: 31,
Joined: Apr 21, 2013,
Offline
I'm playing the rated map Snowfire1972a. it came in a random game, and the map is freaking rediculous. it's easily favored for titans, with a one way path to the other side and an easy way to protect Walkers. batteries can't get in range, and neither can destroyers. this map is very favoured to titans and is rated for random play! take it off! That crap map cost me 30 points
J.C.

Messages: 208,
Joined: Jul 05, 2011,
Location: Italy
Offline

J.C.

Messages: 208,
Joined: Jul 05, 2011,
Location: Italy
Offline
In that map, i won as sap v titans. But i agree it's not very good and my opponent played really bad.

I wish there could be a mixed way to make maps rated or not:
1. Balance Votes: only 2000+ players (would be better if 2200-2300+);
2. Statistical, as sims said

Fun factor may be left as it is, like an unranked vote

Probably, the best way would be the "elitist" way sims named, but i agree with the cons. Maybe all of them can be doable at once:

1. Balance votes and statistical data to make it rated or not;
2. Group of experienced players play the maps that meet previous requirements and decide whether they can stay rated or not.

Test my maps:
1-on-1
ColdestPlace - Tiny Desert - SkyAndSand

3-on-3
DeadFlower
David

4-on-4
BasesGaloreXL6
BasesGaloreXL7
regard87

Messages: 97,
Joined: Mar 29, 2012,
Offline

regard87

Messages: 97,
Joined: Mar 29, 2012,
Offline
I know this keeps getting discussed but I do wish the rating system was altered.

Just played a wide map, moutains forcing you thru a narrow channel, lots of starting money, 750 per turn.
I'm Kralean and my opponent is Titan. Tried to get pressure on him early, get Underlings across, send some Pinzers in for early pressure, etc etc.
My Pinzers were dead by Walker before crossing the halfway point.
I quit out of frustration. Probably wasnt in the best temperment for a rated game anyway but still annoying.

Creator of Powerplay, Weakside, Strongside and Gap Control.
Also OddManBreak, IceRink, Breakout2v2

Testing FlyingV, MixNmatch, GatherRoundtheWell Give them a try!
[Avatar]
waxoid

Messages: 442,
Joined: Aug 07, 2010,
Location: Seattle, WA
Offline

[Avatar]
waxoid

Messages: 442,
Joined: Aug 07, 2010,
Location: Seattle, WA
Offline
This was random ranked or you joined it? It hasn't been as bad as I thought it might be, playing random ranked it's the minority of the time that the map is unplayable in terms of balance, I find.

Putting that aside, which we can't do much about, for maps like that don't worry about pressuring Titans early, you can't. Make sure you can secure *your* defense through the bottlenecks (make it hard to advance walkers, e.g. wyrms and buried ants to rip up any plasmas that advance, pinzers to block speeders) and then patiently spam creatures, eventually fill ants completely under the enemy, tons of swarmers and wyrms where they are effective, pressure the edges of Titan range as you can to make threats against walkers. If you see an opening, spring swarmer/ants/(wyrm) to destroy stuff in a way that you kill more than you lose, then immediately retreat/rebury to bank some amount of material gain and heal as necessary. Rinse and repeat. Titan main chances in that style of game are with sims and UVs - doing enough damage that you don't come out ahead on material. Usually that can make the first exchange more difficult for khral but UV recharge time means the following waves are more dangerous...
regard87

Messages: 97,
Joined: Mar 29, 2012,
Offline

regard87

Messages: 97,
Joined: Mar 29, 2012,
Offline
It was random rated.

Yeah I could try a more patient method but trying to outwait a Titan sounds like watching paint dry. I also would still give Titans the edge in any long slow play game. We'll see though.

Creator of Powerplay, Weakside, Strongside and Gap Control.
Also OddManBreak, IceRink, Breakout2v2

Testing FlyingV, MixNmatch, GatherRoundtheWell Give them a try!
[Avatar]
darf nader

Messages: 43,
Joined: Sep 13, 2009,
Offline

[Avatar]
darf nader

Messages: 43,
Joined: Sep 13, 2009,
Offline
I think it would be arbitrary to pick a rank of who can rate and who can't. Perhaps the value of said vote can be weighted based on how many games they have played as well as their rank.

Also, the "fun factor" of the game should be an index. At worst, it would just be a ratio, but it could also be a weighted ratio.
Forum Index » New Feature Request
Powered by JForum 2.1.9 © - 2020-04-14 v124 - UniWar website