Okay, we agree it’s a problem. Do you have a competing suggestion?
This isn’t a suggestion. This is wishful thinking. Got an idea for how to implement it? This is all I found in the last 120 posts. Go ahead and let me know if I missed one.
I don’t understand what you’re talking about. With unlimited bans there would be no map advantage at all. Both players could decide which map they are OK to play on. On the contrary the current system favors strategic picking of maps to gain an advantage before the match starts. By just banning Arabia a single player can throw off 80% of other people’s gameplay experience and artificially inflate their Elo since they have more practice on hybrid maps. Arabia players are not trying to skew the system to get Elo, they are trying to play Arabia, against other Arabia players, whereas hybrid players want to force Arabia-only players to play hybrid. Who’s the community getting undeserved Elo here?
3 Likes
Neither. Arabia skills and “all other map skills” are not mutually exclusive. But when you have the leverage to force the map of your choice you are naturally playing to your strengths, which gives you an edge. The “Hybrid players” as you call them don’t seem to be forcing the issue, it’s the Arabia crowd that’s doing it. If Hybrid players alt-f4’d every land map it’d be the same problem. It isn’t what’s happening however, and we know that. So right now, a certain sect of the playerbase is abusing the current format of matchmaking to gain an advantage before the map spawns.
This shouldn’t be possible for anyone and this is partially what you’re defending beyond the absurd issue of players, generally not being able to play when they’ve banned Arabia. Which is still the more important aspect of this for me, but eh.
By quit I mean quitting the matchmaking system initially. Depending on the frustration of finding a game, it could eventually lead to quitting the game… In fact, thats what I did about a year ago.
I am one passionate player who is vocal about it on forums. I am sure there are hundred others who have silently left.
I have returned to the game this week to test the “preferred map feature” and after my first 20 TG games I had ONE Arabia game, even though its my preferred. After 3 Steppes in a row, I hopped back on the forums to check if there was any other discussion towards a better way of selecting maps. Going back to lobby is silly when you have a matchmaking system implemented, but I have used it before (you can see that I played several unranked before I quit last year), but unfortunately it isnt always easy to find a good match - TG ELO is not a good parameter, Unranked ELO wasnt a thing at the time, etc.
I could be playing right now, but instead I am here having this conversation because I dont feel like hitting that play button to possibly get Team Islands, Steppe or Yucatan again.
The LoL comparison was made because what you get to choose is Champion and Lane, with Lane being the most important as it has more influence on what the player must do (and its exactly what was out of player control until the changes to role selection). Whislt on AoE you get to choose Civ, Map and Position, though most people do not care about position, and Map tends to be the most important as, again, its what has the largest influence on how the game will play out (and again it is what is out of player control, which can be frustrating).
Dismiss it if you want, but I feel like it explains the root-cause of dodging and how it was addressed.
2 Likes
Now flip it. There’s gonna be players who are vocal about Alt-F4 on the forums and there are going to be hundreds who silently left.
I get your point, I do, I just do not understand why the Hosted games option is not being exercised and promoted better as a way to Arabia-queue. All they need is a rating system and the hosted games would really swell up. This solves both problems, but it’s not being used because Alt-F4 means people don’t need to use hosting as an effective method to get a close, good game on Arabia. So the infrastructure isn’t being built. That’s what I honestly believe.
Brilliant, we can agree on this. So how do we address the problem? Punish dodging, move players back to ranked lobbies? Or do we improve the matchmaking to allow players to choose what they want to play?
I think if we want to change the matchmaking system, for all the reasons I’ve listed to @SwaggyOP, we’d have to split the ELO further, divide it by Open land Map, Closed Land map, Hybrid Map, and Water Map ELO. Then we’d have to change the bans so you can ban categories of maps, but not actual maps. So you’d only be constrained by the genre of map you want to play.
Even though there’s a hardcore contingency of Arabia players I don’t think they’d be infuriated if they were handed Serengeti or Socotra instead. That way, some settings will be expected, some will not, and the ratings will separate to acknowledge this. It also solves the reverse problem that’s happening now, where you reach a ranking, get dealt a map you’re unfamiliar with, then get to play someone at your preferred map level on that player’s preferred map which is a huge disadvantage.
There’s a lot of questions* to answer with this option though. How do you handle Nomadic starts, or Megarandom? I really don’t know. Splitting the queue is complicated.
Part of me worries that the other queues will lack the same support as the open land map queue, and queue times will overall suffer, which is why I simply prefer having the lobby system improved, with rating. Be able to sort it by this same criteria (open land, closed land, hybrid, water) and just join at your leisure. No questions of what settings, no random opponents, overall a better player experience for those who care reasonably about the social aspect of it.
In my eyes, why does rating not exist for hosted games anyway? That just seems like they’re trying to funnel players into the matchmaking system whether or not they want to be there. Like hosted games aren’t “the official” way to play in spite of it being the method of engagement for two decades.
That’s all my honest opinion on the subject. I’m not against the division of queues as long as it’s done generally and the rating system supports it. I’m in favor of improving hosted games and allocating public rankings so that players can easier find the game of their choosing without sifting through the queue if one’s around. How I see it, these two options aren’t exclusive to each other, but for certain, If we aren’t changing anything yet, and we’re not planning to change anything in the near future, a timeout on Alt-f4 is a simple solution that prevents the abuse of the system currently implemented, and it’s the easiest to introduce.
I believe allowing up to 8 bans is even faster than implementing a timeout.
Splitting the queue before allowing more bans is trying to solve a problem that does not exist before sorting out a problem that is currently happening, so its an unnecessary barrier at this stage. It may become a problem? Sure. It certainly happens now on a small scale and it would be amplified in an year or two… I am sure solutions can be developed along the way.
Adding a timeout only adds frustration that, as I stated earlier, will only push players away from the game (or at least matchmaking); or towards griefing/other bad behaviour (insta-resign, trolling, smurfing). Just like altf4 is pushing players away. You solve one issue and create another.
Matchmaking is a way more efficient system and its why all games push towards that. It gets you in a game fairly quickly, in a server that is usually the best for the players involved and with players all within similar skill level, from opening up the game to clicking your TC in 5-10 minutes. It reduces the chance of points being traded, players being singled out by the community, social phobia, server / lag issues (currently you have to dismiss your lobby to change server, losing all players), etc.
You don’t think there’s a problem with getting a map you have literally no idea what you’re doing on and lose as a result against someone competent on the map? Because that happens. The only reason it doesn’t happen continuously is because of Alt-f4 abuse.
8 bans means ELO is absolutely, totally worthless. You don’t know whether that guy is 2100 playing Arabia, Arena, or Islands. Once you make it so people can select what map they can play and ascribe a general rating to specific maps that rating becomes absolutely worthless.
If all players that would queue on Open Land Queue move to Lobby, the effect will be exactly the same for the “other queues”, no?
Allow more bans and let the algorithm do its thing to find you a game without worrying about creating a lobby, kicking players that are not within the rating you want, kicking players that are from servers on the other side of the world or spending 40 minutes waiting for someone to pick your lobby (like in the Voobly days).
Alternatively, if banning up to 8 maps is not acceptable for some reason, allow the first 10 minutes of Search to be for other players with the same Preferred Map. After 10 minutes it broadens to included non-banned maps on your pool… Player then has the option to cancel and requeue. Again, I am not in favor of splitting queue. Playing a hybrid map every now and then is fun too.
There is a good portion of the community playing Mega Random, Nomad, Black Forest, Arena, etc. Its not like there wont be any games.
In fact, it will improve the chance they get to play those maps, because you wont have Arabia and +3 Arabia variations voted every month (check this month’s 1v1 map pool for evidence).
Unfortunately, nowadays, I vote for whatever is an Open Land map so I can play what I find fun in most games.
If you can control what map you’re playing on, you must split the elo brackets. You can’t reasonably compare the skill of 2100 islands player compared to a 2100 arena player. Without splitting the brackets, you can have absolutely no clue of the skill of any player, regardless of rating, on any map, because who knows what he played to get there.
This is a problem with the current system and it’s only getting worse the longer we have preferred maps and map bans, this is inevitable, and it needs to be changed.
I am not sure I follow. If you suck at the map, you should lose, no? If you ban said map from 1000 to 1800 ELO and then decides to give it a go at 1800, you should lose… But you made that choice. With limited bans, you dont get to make that choice and thats why people altf4.
Sorry, your point seem to validate what I have been saying. Perhaps I misunderstood.
Let’s say you played every game for a year on nothing but islands. nothing but islands. Even if you don’t climb the rankings dramatically (let’s say there’s five or six island regulars you can’t beat) you’re going to be a monster on Islands after a year. Your overall rating does not reflect your skill on maps generally, and you might lose to players on another map and utterly dominate people rated far higher than you on Islands.
This is the system working properly, as is, and it’s inevitable. The polarizing nature of overwhelmingly biasing map selection will lead to players with vastly different skillsets represented by the same overall rating that will effectively tell you nothing. If we aren’t going to prevent people from circumventing random map selection with either biasing the sample or dodging maps they don’t want to play, this deterioration of the rating system is going to progress.
The only options are to split the ELO ratings and maps by category and reset everything, or remove map preference and punish Alt-F4. This is the problem, those are our options. Pick your favorite.
Sidenote. This is why Jonslow is, and is known to be, an Arena Specialist. He still had that high rating on Voobly, but since he only played arena games (and you could see said distribution) people know what was up. Your average everyday rando does not have that information on your other everyday rando.
Ah, I see. 2100 Island may be a much better or worse player than a 2100 Arena if you see them in isolation, I agree, because you are considering them isolated in two different playerbases.
I lack the data to ascertain if the above would indeed happen, but considering how there is a significant portion of the playerbase that would continue to play more than just one map, the comparison wouldnt be between two completely different playerbases, but actually 50-80% similar which would level the ELO. I may be wrong.
I mean, we’re certain to be in that phase already because of the number of players that almost certainly play exclusively Arabia or Arena and skip any other map. We don’t see the downsides of it besides alt-f4 because Alt-F4 is still an option. But you could easily come up with a scenario where this massive disparity of overall skill represents a problem in practice.
If effective ELO was ever used as a prerequisite for tournament qualifiers, or leagues, etc, a player can be seated with a very good map and no experience whatsoever, and a relatively solid player can be denied entry on the base of having 17 less elo than the next guy even though that player is 87% arena. That’s just the simplest example (what good is rating if it can’t be used to seed or gateway a bracket for an event properly) of what a good elo system should be able to do and in the current system we have absolutely no way to confirm that this is how it is working, or if it is working.
Practically we seed by meritocracy, but if we want to run a big tournament you’ve either gotta have a lot of time to burn or draw the cutoff line somewhere.
Actually, we can see one specific example of where this comes into play currently, and that’s in those variety players you mentioned. Say you play everything. That will automatically come with it’s downsides in a more polarized, further progressed Alt-F4 or “ban everything else” map selection world. Simply put, if you get slated on Arabia or Arena, you’re far more likely to find someone who’s a map specialist who will naturally have the upper hand in the matchup simply on the basis of playing it more than you. This will be the case for any popular map that gets it’s own following of “ban everything else” players.
Good discussion. I think I understood your point of view and perhaps we may have to agree on disagreeing.
I am not too concerned about the ELO per se, or the effects of not splitting the ladder may create where a few maps are easier to climb than others. I may be simplifying it, but I reckon the ELO variation wouldnt be over 100 points between them (considering shared playerbases) and perhaps except for the top level, it wouldnt significantly affect 95% of the playerbase.
Forced maps and altf4 seems to affect everyone and simply punishing altf4 wont solve the issue I am afraid.
Funnily enough I was about to join a Team Islands game but someone altf4. Good. Imagine being stuck with someone that doesnt want to be on a Team Islands game for the next 20-100 minutes. Back to queue.
Land the enemy and #Yolo. What could possibly be the problem?
just look at the winrates here: AoE2.net
almost nobody is finding fair games (45-55% win rate). it’s pretty much all >60% winrate for the top 500. this is the sign of a system that doesn’t work. the absolute best players should win more because nobody can match them, but the amount of players getting easy games is comical
rating system needs to be fixed. matchmaking needs to be fixed. letting people alt+f4 needs to be fixed.
and most importantly, people need to be able to actually play the game instead of sitting in queues forever.
1 Like
TG ELO is flawed and inflated. It is mostly useless. You can find people with 40% winrate at 2500 ELO that has a 1000 ELO 1v1, simply because the more you play, higher you climb as you get more points than you lose (something about it comparing to the highest on the enemy team to award points, if I am not mistaken).
I just played 2 TGs, lost one, won one. Net gain of +10 points. This happens to everyone, so bad players will climb the ladder nonetheless and we will see this continuous inflation of the ELO points. As a reference, back in June last year the highest TG rating was about 2700 and now its 3300. Same players.
Ignore it and use the 1v1 to measure ELO, specially after the player has over 50 games.
1v1 rating is somewhat better, but the fact that if you have a TG ELO before playing 1v1 also inflates your 1v1 ELO and you start near 1700 instead of 1000 ELO. You can find people with 10% winrate in 10 games at 1500 ELO because of this.
1 Like
It was one of my teammates that altf4 I suppose. His name disappeared from the waiting room list before we dropped out of the game and back to queue.