wookieontheweb
Messages: 485,
Joined: Jan 27, 2016,
Location: Southampton, UK
Offline
|
|
wookieontheweb
Messages: 485,
Joined: Jan 27, 2016,
Location: Southampton, UK
Offline
|
There was a decision recently to reset the map vote counters and only allow votes after someone has played the map.
I think the result of this is we are about to lose all but the built-in maps from ranked random.
If you go to the map browser and select 1000+, ranked maps, then cycle through the list you get (2018-11-05):
I'm not sure of the exact criteria that makes a map unranked, but from experience at least half (probably more) will be un-ranked before they reach 200 up votes.
This will leave us with only built-in, new maps (but only for a short while) and a few from Master Yoda (and a few others). From above that's less than 150 maps and it will fall.
150 sounds like plenty but it isn't. The range of maps you play in ranked random is normally from XS, S & M. So now you're under 100 maps. Given most people run 10+ concurrent games that means you'll tend to see the same maps over and over again. Interspersed with random new maps that even if you like you're unlikely to see again.
Up vs down votes is not good enough as a way of determining the fate of a map. Some common strategies on some maps are flawed and other strategies are much harder to counter. The map may not be unbalanced but it will feel like it when you're on the wrong path. If you revenge and crush your opponent in a similar way then your assumption is the map must be broken. That's potentially 4 down votes for a balanced map that actually requires some original thinking, something that pushes you out of your comfort zone. Don't we want more of these maps, not less?
I think we need a better way of identifying maps that are "good" and then add them to the built-in list so there is no more voting on them.
- Build up a large pool of known "good" maps.
- Show new players what interesting, balanced, good maps look like.
- Give map makers a dream to aim for!
|
Android 9. Samsung Galaxy A50 |
|
The Impaler
Messages: 147,
Joined: Jun 07, 2017,
Location: 221B Baker St.
Offline
|
|
The Impaler
Messages: 147,
Joined: Jun 07, 2017,
Location: 221B Baker St.
Offline
|
I've passed a link to this post onto the other mods, we will see what they think. Good investigative journalism, friend!
|
Member of the balance team
Moderator
Highest Score: 2685
Highest Rank: #4
Current Rank: N/A
Highest Championship Ranking: 4th
Highest Tournament Ranking: 4th
Current Record: 131W, 14D, 4L
Winner of Duaneski's mapmaker challenge
Winner of Angkor's 2v2 Grand mapmaking challenge
Winner of Lkasr's 3-Minute Tournament |
|
simsverd
Messages: 921,
Joined: Dec 02, 2010,
Location: Bergen, Norway
Offline
|
|
simsverd
Messages: 921,
Joined: Dec 02, 2010,
Location: Bergen, Norway
Offline
|
wookieontheweb wrote:There was a decision recently to reset the map vote counters and only allow votes after someone has played the map.
I think the result of this is we are about to lose all but the built-in maps from ranked random.
I'm not sure of the exact criteria that makes a map unranked, but from experience at least half (probably more) will be un-ranked before they reach 200 up votes.
This will leave us with only built-in, new maps (but only for a short while) and a few from Master Yoda (and a few others). From above that's less than 150 maps and it will fall.
Up vs down votes is not good enough as a way of determining the fate of a map. Some common strategies on some maps are flawed and other strategies are much harder to counter. The map may not be unbalanced but it will feel like it when you're on the wrong path. If you revenge and crush your opponent in a similar way then your assumption is the map must be broken. That's potentially 4 down votes for a balanced map that actually requires some original thinking, something that pushes you out of your comfort zone. Don't we want more of these maps, not less?
I think we need a better way of identifying maps that are "good" and then add them to the built-in list so there is no more voting on them.
Hello
Formula for map to get rated is not changed; its still 25+ votes where 70+% positive.
What has been done:
- disregard votes from players when they have not played the map (also for older votes)
- only use the last 200 votes on a map. This will only affect maps played very much.
- upvote on loss and downvote on win have 1,5 value
You are wrong that numbers of rated maps will decrease to the "buildt in". The analisys i have done predict more maps in the rated pool than before changes (based on current maps and votes).
Current system might not be perfect.. i dont think you will find that, but its a decent tradeoff between simplicity, practicall and fairness.
Your solution of making good maps "in buildt" will not scale.. as it rely on continous human evaluation of maps.. or the selection will be outdated with changes in balance and meta. In my oipinion the system have to be automated to be viable.
|
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 |
|
wookieontheweb
Messages: 485,
Joined: Jan 27, 2016,
Location: Southampton, UK
Offline
|
|
wookieontheweb
Messages: 485,
Joined: Jan 27, 2016,
Location: Southampton, UK
Offline
|
You're right. Anything that requires constant human evaluation is not sustainable.
Perhaps when a map goes under the 70% mark (and probably over a certain number of games too). It should be automatically flagged for high rank check. That means it will be far higher in the odds of two 2k+ players being given it as their random ranked map.
The configuration they get is the one giving the largest negative score e.g. s p1 vs t p2 (they won't know it's not random). At the end of the match a popup appears asking them to vote on _balance_ as their decision will determine the fate of the map. They can still revenge to recover their rank.
After say 10 such matchups (including revenge) either the pros will concur that the map is broken and it will go. However if they feel it is good then stats are adjusted such that the negatives are reduced so the map rates as 85% positive.
Not sure how many times a map can go through this process before it's deemed unfair on lower ranks. Probably just once.
|
Android 9. Samsung Galaxy A50 |
|
The Impaler
Messages: 147,
Joined: Jun 07, 2017,
Location: 221B Baker St.
Offline
|
|
The Impaler
Messages: 147,
Joined: Jun 07, 2017,
Location: 221B Baker St.
Offline
|
It's hard to say exactly what is the best way for this to work. the real problem is that people dont vote after their games. However, if we try to incentivise voting, it ONLY incentivizes a vote, not a GOOD vote.
|
Member of the balance team
Moderator
Highest Score: 2685
Highest Rank: #4
Current Rank: N/A
Highest Championship Ranking: 4th
Highest Tournament Ranking: 4th
Current Record: 131W, 14D, 4L
Winner of Duaneski's mapmaker challenge
Winner of Angkor's 2v2 Grand mapmaking challenge
Winner of Lkasr's 3-Minute Tournament |
|
wookieontheweb
Messages: 485,
Joined: Jan 27, 2016,
Location: Southampton, UK
Offline
|
|
wookieontheweb
Messages: 485,
Joined: Jan 27, 2016,
Location: Southampton, UK
Offline
|
- And I've just noticed I've been including the built-in maps to those counts which I should have removed, but that's more effort! and makes the voted numbers worse!
Some questions for SH & the balance team:
- Are there really only 205 maps left?
- Given random ranked rarely (never) picks built-in, large or extra large maps are we really only playing with 171 maps? About 75 of those are Master Yoda, and another 25 are Jeffph and about 20 built-in. That leaves 30 odd maps which are probably going to be voted into oblivion.
- Are all of the other 13000 maps really that bad?
- Did we really plough through all 13k maps?
- What was the decision as to whether a map got re-entered? I know none of mine did
What about giving maps that get demoted a 2nd chance but only allow same race match-ups on them?
|
Android 9. Samsung Galaxy A50 |
|
simsverd
Messages: 921,
Joined: Dec 02, 2010,
Location: Bergen, Norway
Offline
|
|
simsverd
Messages: 921,
Joined: Dec 02, 2010,
Location: Bergen, Norway
Offline
|
Those stats are wrong. There are at least 7-800 rated maps (probably more).
(i have a detailed list from pashaka)
We have also seen that the map-draw seems to have less variety than what should be ... we are looking into it.
Regarding map to "re-enter".. just read what i wrote above.. it explains all we have done and why some maps fell out out the ranked pool, and why some others got in.
|
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 |
|
wookieontheweb
Messages: 485,
Joined: Jan 27, 2016,
Location: Southampton, UK
Offline
|
|
wookieontheweb
Messages: 485,
Joined: Jan 27, 2016,
Location: Southampton, UK
Offline
|
Thanks Sims. That's a relief that there are more than 200 maps remaining. Looks like there is another bug in the map browser.
I'll stop worrying about it. Perhaps this is the same issue giving the unexpected variety. It probably uses the same algorithm.
|
Android 9. Samsung Galaxy A50 |
|
svart drake
Messages: 2,
Joined: Feb 03, 2019,
Offline
|
|
svart drake
Messages: 2,
Joined: Feb 03, 2019,
Offline
|
Hi,
I think this thread is important. I wonder why the developpers didn't ask themselves why Uniwar is dying out whereas it had some really outstanding features vs. other available games. In my opinion, THE most important was the map creativity supported by community vote to ensure a diversified pool of good maps and renew interest in the game. (Clue : guess why I almost stopped playing...)
Alas, the vetting process is flawed, as according to the map browser system there are indeed +/- 2% of maps that are currently rated (where does Simsverd get his figures from?), among which too many "boring ones" that may be balanced but do not allow tactical skills (think of the numerous maps where you have marauders vs speeder match ups and where you can only count on your luck (think of the supposedly randomized algorithm, the second flaw of the game) to win).
Maybe (good) map making should be incentivized: how about giving out say 50 unicoins if one of your maps reach the rated system & 250 games, 100 unicoins if your ranked map reaches 500 games etc. (clue: as you can see from my account I"m not speaking in my own interest as I have plenty of useless coins)?
Or, what about giving automatically "built-in" rank to thos maps which reached a 1,000 games without being delisted? After all it's hard to believe a map could resist the analysis of so many players without being kicked out if it were really bad. Or maybe to acquire "built-in" status, the map should be certified on top by say at least 5 games played by so-called pros?
Oh, and last but not least with no new units / no new terrain in more than a year, if there are no more good maps is it really a surprise if the pool of players is dwindling?
|
|
The Impaler
Messages: 147,
Joined: Jun 07, 2017,
Location: 221B Baker St.
Offline
|
|
The Impaler
Messages: 147,
Joined: Jun 07, 2017,
Location: 221B Baker St.
Offline
|
Every era. of every game ever has had people saying the game is dying. unless you have numbers for daily active users it just sounds like fear mongering.
We need a system that works automatically because we cant vet every single map, and it also has to not incentivise people from gaming the system and voting in bad maps for coins etc.
This message was edited 1 time. Last update was at Feb 04, 2019 11:01
|
Member of the balance team
Moderator
Highest Score: 2685
Highest Rank: #4
Current Rank: N/A
Highest Championship Ranking: 4th
Highest Tournament Ranking: 4th
Current Record: 131W, 14D, 4L
Winner of Duaneski's mapmaker challenge
Winner of Angkor's 2v2 Grand mapmaking challenge
Winner of Lkasr's 3-Minute Tournament |
|
svart drake
Messages: 2,
Joined: Feb 03, 2019,
Offline
|
|
svart drake
Messages: 2,
Joined: Feb 03, 2019,
Offline
|
Well, Mr Impaler, let's look at figures: I can see roughly 200 active players at any given time: does that reflect a growing pool that reaches out in the whole world? Let's ask for the attrition rate also (maybe you can get the numbers I can't).
I have made several proposals, that invite to discussion. You have made as far as I can see none for this subject. Obviously criticism is easy but creativity to come with new solutions is much harder
So, how can you construe my analysis, backed with practical ideas as "fear mongering"?
Isn't the danger rather your blind faith in this game?
|
|
|
|