This season on my main account, I’m at 54% WR.
On my other account where I haven’t played Aram as much, I’m at 84% WR. Lucky winning streak? Possibly. The games feel very easy nonetheless. (solo queue mostly on both accounts)
Lots of smurfs could confirm this reality. Create an account and level it to 50. If you’re a master player with good experience on lots of heroes, you’ll likely get a 80% WR.
On heroesprofile, both teams have most of the time the same MMR average in a ±100 difference, which is around 1 league apart.
The only big exception are created by 4-5 stacks in high mmr or with smurfs, which don’t represent most case (maybe 15%).
Queue times are way shorter than QM since the MM isn’t restricted by roles and mirrored heroes. So, to say that ARAM doesn’t have MMR because there’s almost no waiting time compared to QM is not a good reason.
Arguments for no:
You tell me.
We can have a bronze player with a master player, but it’s no different than QM MM.
the consequence of third party tracking is that it’s going to eventually fall into parity because it assumes that matches are made with the assumption that they should be made. If the match is made, then the mmr must therefore be in a range to have the match, so the numbers will be adjusted enough over time to reflect that.
when people look at mmr ratings in Qm they either pull from third party sites, or they use ranked matching and ranked doesn’t have any bearing on QM mmr. that’s why a lot of mmr complaint topics tend to boil down to confirmation bias.
I would guess that it formally does, simply because it’s easy to fit it into the same matchmaking workflow- you just use the same workflow for QM, ARAM, Ranked, whatever. They all function the same (because why would you create a separate algorithm, anyways). But at the same time, given how many times I see a mix of people that a way too far apart in skill while the queue time was short, my guess it just expands the search like after 3 seconds or something similar. So if you are playing during peak times then it may find a decent game almost immediately, but then if you are like me then you will see a lot of completely rainbow matches despite the very short queue time.
Assumptions and estimates will only get us so far, but how enough is it to prove a point?
With a limited amount of data, we “proved” that solo queue players win 49.0% of the time while 5 stacks are at 53.6% in SL with healthy confidence intervals.
I wish I could find it, but yes it was said officially that ARAM used no MMR. As you say, if you queue late at night you can be matched with AI and it’s not uncommon to see a level 2 account put agasint a Master player.
i’ve played over 7000 arams and the other day i had a lv 4 on my team who’d only played 5 matches.
so i’m gonna guess there isn’t MMR
i also quite often see one team getting a combined team level of a few hundred
and the other team getting a combined team level of 5000+
experts who’ve spent hundreds of hours in the game and know every skill and talent,
vs a team of people who don’t even know how to play their own hero.
Was the lvl 4 in a duo?
QM is no different in these situations and it uses MMR.
Was it during peak hours?
lvl +5000 doesn’t mean anything.
Count on your heroesprofile how many games were severely unfair (±150 MMR) and post your results.
I’m putting my guess to less than 15%, similar to QM.
Also, a master player from SL starts at 1,800 MMR from his first ARAM game.
Back when Brawls were released, I had almost never seen a bronze player in QM except in extremely rare cases with groups. It happened quite commonly in brawls. I didn’t truly understand how bad people could be until ARAM and why I always laugh at people who say people in bronze are just as good as masters they’ve played with.
You do understand if ARAM had matchmaking, you could get MMR by beating bots?
There isn’t any.
I’ve played against plenty of bots in ARAM, even matches where the enemy had 4 AI’s and 1 human.
Not disconnected players. Real AI’s that show up as AI’s in the loading screen.
There are also level 0 people who have never played a single game in their life before that match.
I had the same thing. 4 AI’s on my side vs 1 human with 4 AI’s on the enemy team. So it can’t use MMR. I remember too when it was introduced the blues saying there was no MMR.
if it had any mm like storm league, the top aram games would be hellishly hard to play and those at the bottom would be just like bronze 5 games
but since it has no such ladder you can get completely random folk on each side or even robots
in a mode like storm league you cant que up as a gm and randomly go into a match with a bunch of ais and level 12 bronzes, the lowest youll get is maybe a diamond 1 or master
my main issue with aram is when it pops up that message about speeding up the que with robots, i usually cancel at that point because why bother
I’m not comparing SL MM to ARAM MM, but QM MM to ARAM MM.
I’m now at my 39th ARAM games and games are getting harder, which is fine and what I’ve expected. I no longer have team on average Bronze 4 MMR, but gold/platinum.
It’s getting closer to games on my main.
I can’t believe that I’m alone in this.
Easy way to find out is to play cross region starting at lvl 1 on your main and look how your team mates perform.
seems completely random to me, i get 4 very hard games in a row, then out of nowhere i get a level 1 on our side when everyone is solo, who barely knows how to cast abilities
the lowest i ever get in qm (with acc lv 1300) is maybe a level 600, or if its an extremely high mmr smurf maybe a level 150, but never a level 1
another weird thing in aram you never see in qm: if you whisper a lot of the players when its over, a lot of them have only played ai mode, especially those under level 50
i never see low lvl ai mains randomly dabbing into qm and being placed on my side
I can’t say I see any difference after having played a lot of ARAM. Most teammates are around gold, which is to be expected as that is the average rank.
What I sometimes see is a day to day difference. I play solo only and some times I get very very low levels, and the other days I get hard games with higher levels. I do make it artificially more difficult for myself as I usually always lock healer if available, which doesn’t always have optimal carry ability.
But it’s pretty obvious there is no matchmaker at work since some of the players it gives you have just joined HotS even if my winrate is like 70% after a ton of games. I have never had fresh players on my team in Quick Match except like 2015 when I joined.
Do you mind sharing how many games have you played in ARAM since 2020 season 4? It’s when heroesprofile started estimating ARAM MMR.
Let’s say this:
Each team’s players MMR: 1700, 2000, 2300, 2600, 2900.
I’m the 1700 player on a new account, but since my skill is master, it’s much easier to win.
2300 on average, but replace the 1700 by 2900 and you’ll have a true value of 2540, which results normally in a stomp when both team are supposedly fair.
It’s just an estimate and not necessarily truth. But, I do believe that it exists to make games more balanced when a player has over 100 games.
I may go see if I can find something explicit, but I distinctly remember that when it was first announced, it was meant to be a mode where it literally treated every player the same, regardless of age of account, hero pool, etc. The closest thing I can find to such a statement was from the original trailer.
Here it is implied that there is no MMR, as anyone can be matched with anyone else. “Everyone enters Heroes Brawl on equal footing.”
