Where are you getting the stats that it is the most popular? I would be interested to see a link.
1 Like
While I agree with you that laning provides another level of complexity that you donât get with ARAM, the main reason I like ARAM is that it gets to the 5v5 action quicker and more often than QM/Ranked.
And while laning often helps get the win in QM/ranked, the most exciting part of the game is the 5v5 encounters. Objectives are great, but they arenât always 5v5 (depending on map) and Less often is the laning phase the most exciting.
While I mainly play ARAM, QM is still in the rotation.
Oh its just theory based off my observations, I obviously donât work at Blizz and able to provide hard facts. But I can tell you this:
- ARAM queues are about 3-5 seconds for me
- After ARAM ends, you can auto queue and rarely ever see the same people you just played with (it happens on rare occasion, yes).
- QM queues are about 30sec-1min on avg. And before you say there is role matchmaking, I just donât feel this is a major factor if the available pool is that big enough to begin with.
- Ranked is 1-5+ mins depending on what time of day you log in. Forget trying to do rank during off hours.
And there was also the article/dev interview explaining why they added ARAM in the first place (being wildly popular).
All antidotal, but just my guess based off observation.
So, completely unreliable. OK, here is a slightly more rigorous way to look at things. Heroes Profile has a tool that shows you a whole bunch of stats. Granted, you have to take it with a grain of salt, and granted, people who play ARAM are probably the least likely to upload things, but it at least has verifiable numbers rather than anecdotes.
One of the tools it provides that is relevant to this discussion is a breakdown of team compositions:
Just going off hero picks is tricky, because ARAM allowing things like 4 Azmos skews trying to get a game count based on hero counts. Every match, no matter the mode, can only have 2 team comps, so this would be more representative of how popular the mode is in general.
What I did is ran three filters for the latest patch, and added up the total numbers of comps with more than 1000 recorded matches, just because of diminishing returns and, well, laziness. Here are the totals for the three most played modes.
SL: 169,779
ARAM: 143,172
QM: 207,254
By this metric, ARAM is the least popular!
So this is even more evidence that we can easily do away with ARAM, and focus on balance for the other two, as already the fewest number of matches are played there, and those people can just go run it down mid in QM. Which most of them seem to do anyway.
Problem solved! Everyoneâs happy!
1 Like
Completely agree. Which makes your conclusion just as anecdotal as mine.
1 Like
No, it just means I would have to calculate error bars. Regardless, QM players are equally âcasualâ, and the QM play rates obliterate ARAM ones, even when hefty error bars are brought into play.
Removing ARAM is clearly the solution to removing public complaints!
Appreciate your scientific analysis here.
Anyways, how about the fact that 10 ARAM games pop, in the amount of time it take 1 QM to pop?
Iâll just go ahead and press the big red button and nuke the whole thing now.
I am sure your time estimates are based on accurately recording both types of queues with identical parties at identical times rather than going off feelings, right?
Regardless, do you know what queue pops even faster? AI matches. Does that mean they are 5 times more popular?
I know you dismissed the likely true reason simply because it conflicts with your biases, but it really is down to matchmaking. Remember, the QM matchmaker tries to balance not just MMR, but party size, experience, roles, and unlike ARAM, you canât have more than 2 of the same hero in a match.
But of course, once ARAM is removed, you can always find people in game chat who will be open to creating custom games where you can brawl to your heartâs content. Just like it was before they made it a mode. So see, even your minority will still have options, while the remaining developer can focus on the popular modes.
2 Likes
Youâre forgetting Vs AI here, I feel confident that players in that mode are the least likely to upload to HP.
While we now only have unreliable stats from HP regarding mode popularity, back in the day, the then Devs posted (somewhere, here or Reddit), that the most popular modes were; Vs AI, QM, ARAM, Ranked, URD.
That was a long time ago, probably around 2018, Iâm sure that ranking order has changed (except for URD), and my feeling is ARAM would be the most popular PVP mode, simply because queue times are instant and matches are shorter.
What I do know for certain, regardless of the modeâs popularity, is those who main ARAM, seem to be very passionate about it, so perhaps kid gloves are in order when undertaking any significant changes to this mode.
1 Like
HP doesnât accept AI replays, so we will never know if AI is more popular than ARAM or not. And as I discussed above, queue times are not really a reliable measure of popularity when the matchmaking algorithms are wildly different for every mode. While the number of matches recorded on HP isnât rock solid, it is the best that is publicly available, and can at least be the basis for a reasonable statistical analysis.
It would also be interesting to know how many people truly only play ARAM vs people who play a mix of game modes, but that is something that I donât think is possible to accurately calculate for a non-Blizzard employee.
I agree, but considering queue times for ARAM are almost instant, and ARAM games on average are faster than QM/Ranked, ARAM should be the most played mode. There is no waiting 2â3 minutes for QM, even longer for Ranked, where there is also draft lobby that makes those games last longer than ARAM.
1 Like
I do agree that Ranked isnât the most played, but it actually wouldnât surprise me if it is closer in popularity to ARAM than one would expect from Reddit/forum chatter. While people who play ARAM, and nothing but ARAM, tend to be very vocal, it does tend to be a fairly small group of people who play it passionately. When I do play ARAM, I often see the same people from match to match, unlike QM. Hero balance is about as good as it has ever been, so the people who are dedicated to SL (which seems to be most of my friends list), donât have all that much to complain about, other than in the lower ranks where smurfs/trolls are relatively abundant.
The whole âshort match/queue timeâ can be very misleading, as since there is minimal matchmaking, a fairly small pool of players can make for short queue times, even if the total number of games are low. For example, 100 people getting insta-queues and 13 minute matches is still a lower total number of matches than 1000 people getting longer queues/matches any given night. That is an exaggeration, but you can see what I mean.
I canât deny this, as this is also my experience playing in ARAM. Many times I check the profile of such people, and they have been playing ARAM for almost 12 hours straight!
Youâre right too, those who love ARAM, are very vocal and passionate about it, which I why I think it might be the best to not touch any fundamental aspects of ARAM, as patches to fix any problems are infrequent.
1 Like
Game playerbase isnât made up an average of players playing all the time; the population has drastic spikes were most of the players only play for a limited time in a particular window.
https://steamcharts.com/app/570
Thatâs a graph projection for DotA 2. Off-peak hours list ~230k players while peak 630k. Other games might have more drastic spikes, but the concept applies to many multiplayer games.
Aram would take a smaller amount of players to cycle shorter games, while QM would take a significantly higher playerbase to out-play aram. When you look at other activity levels â such as complaints or shared moments â itâs not unlikely that QM has more players than ARAM.
Keep in mind that many chronic complainers here project the remaining playerbase to be significantly smaller than it can be demonstrated to be due to HP replay rates. People that refuse to consider otherwise tend to refuse to see otherwise, but thatâs also why their complaints try to disqualify players instead of better rationalizing that most players do not share their personal experience. ie, there are more [ÂŹthem] than there are [them].
Observation influences understanding, and understanding influences observation.
One of the key âsellingâ points of HotS & QM is picking a specific hero; players that really want to play that specific hero can be similarly passionate about their games to the aram vocals.
We could also think of it like stores: I can get in and out of a dollar store faster, but Iâm also only going to buy a few token things; when I go to Costco, I get more stuff at a higher price point, so which store is going to have more profits from my anecdotal evidence?
This is a fair assessment, which is why I agreed in my last reply to Hoku.
Speculation aside, I would love it if someone at Blizz could release the actual numbers for these modes. When they released the figures in the past, Vs AI was the most played mode. This was before they âimprovedâ the AI, so I kinda doubt it would still be in the number one slot.
1 Like
In my opinion people should just keep playing ARAM including mirrors if that is what they like and ignore those who hate mirrors. Its clear that those who hate mirrors dont want a civil discussion about it and just want to force thier toxic opinions into others. Too bad people who like it have to suffer playing outnumbered or get into leaver que becasue of some loud minority that cant accept change.
Minky pretty much nailed it here. But wouldnât you assume that if 5-10 ARAM games are happening in the time that it takes one QM to happen, then logically it is a 1:7 ratio? (my estimate).
The algorithm for QM matchmaking goes against your argument, so not sure why you would bring it up? It adds to the fact that QMs happen less often.
Also as someone else mentioned, ARAM are generally shorter. I imagine an ARAM player gets in 5 games vs QM getting in 3 games in the same timeframe.
Lastly, you bring up AI popping instantly. With full AI games there is no user pool to pull from so they happen almost instantly as soon an a server opens. If you choose the Coop vs AI you are going to wait for a very long time, nobody does this mode. So this is a week argument entirely.
One game mode that has a matchmaking to care about vs a game mode with no matchmaking hence the fast ques. Your post make no sense.
The question was about what mode was more popular. Not what your feelings on matchmaking were. Faster queues (regardless of matchmaking) = more games happening. Thanks for agreeing and sorry you donât understand what the topic at hand.
That doesnât follow. If you have 20 people rematching against each other for an hour, with instant queues and 15 minute matches, that is 8 matches total, where 200 people with a 1 minute queue and 19 minute matches would be 60 matches total.
Once again, you are conflating speed with popularity only because it supports your belief, not because data supports your claim.