Win rate has nothing to do with MMR. They could have been fighting much tougher opponents for most of their games, while you might have been fighting easier opponents.
Playing solo lane or off-lane heroes. KDA does not reflect team contributions such as experience earned or damage dealt.
They might also be the tank. The tank that everyone expects to take all the damage and so ends up dying and reported for feeding or being reported for throwing if they do not die because the game is at a level disadvantage anyway.
30% or lower. 4v5 is strongly in favour of the 5.
Yes but you also get such people as your opponents so it all balanced out in the end.
You wont have high winrate on all your heroes anyway. You will be bad with some heroes while being good with others. Also a reason why your mmr on HP is divided into Role/Hero/Account mmr.
I have close to 3000 uploaded games on HP and checked nearly all of them and only time I saw a huge leap in mmr is when I played againts master rank premades. All other games both teams was very close to eachother.
Too bad your real mmr is hidden so you cant see what your real mmr is so you dont have to use incomplete data from a website. But I guess showing your real mmr would just give quitters more excuses to leave a game cause his team is much lower than him.
True. Most people are also incapable of accurately judging their own skill level. Those who are capable usually donāt make threads complaining about potatoes, because they recognize the weaknesses in their own play.
I agree that itās incomplete, but Iād say the HP algorithm is 97-98% accurate.
A few times I was bored and checked in-game ranks, then compared them to the HP ranks. They were all accurate within a single division or so.
The only time it jumped 3-5 divisions was if that player had taken long breaks and the site didnāt have much recent games uploaded from that person.
If the game still had lots of players, and most of the games would get uploaded to the site, itād be close to 100% accuracy.
Since when 1 player in the match uploads, it updates stats for all the 9 other players as well whether they upload or not.
But ya Iād like them to do the same thing they do in SC2: Just show the MMR number to us directly so we donāt need outside tools for it.
Also, I donāt think it happens that much. While I do meet virtually any rank, most of them play around my level. QM can consider hero MMR while SL can be anything. (My skill spread per hero is so much that my rank is entirely meaningless. Feels the same about everyone. Hardly anyone can flex anymore, so matches are completely random. At least with QM you can prepare for your hero mentally and there is no tilt because had to fill. At worst, you get unwinnable matchup.)
My wish to see MMR isnāt really about validating the system but more about understanding what may have happened and observing how am I doing compared to earlier (dropping, climbing). Mostly because Iāve written off Slave League so Iām playing Prepick Autodraft League instead.
āHPā is not run by Blizzard? So their MMR values are estimates at best, made up at worst.
As far as I am aware ranked has 1 MMR value which is always used, irrespective of your hero choices. Ranked MMR determines what league you are in so no guesses are required. QM has some MMR flexibility that varies with heroes and also has MMR leeway with matching opponents so you can get significant MMR variance within and between teams. QM is not a serious game mode as people often use it to practice or try out things, as such only stats in Ranked matter.
Your ranked MMR is not hidden. It is your rank points. There may be a slight, temporary, adjustment for leaver penalty, but outside that it is very much visible and well defined.
QM MMR is still hidden but no one really cares since the game mode is MMR flexible anyway.
Itās probably not. The matchmaker will try to match people of similar MMR. But, when thereās a limited pool of players, the matchmaker will just throw whoever onto the team to get a game started.
I mean, itās more complex than just win rate for some of the reasons you state. But at 40%, it means that the matchmaker canāt find matches for them to win roughly half the time, which means theyāre probably really bad.
What you fail to realize is: if the game doesnāt have enough people at your āmmrā or skill level, and it only put you against equal skill level playersā¦ you will sit in queue indefinitely.
Iād rather play with a stinker or 2 on my team than not play at all. But you canāt complain about a game not giving you equally skilled matches, when thereās what like 150 people playing this game? The pool of players is too small.
They have to be on someones team BUT what is worse is normally when this happens to me I get smurfs on the other team.
In all honesty as long as theyāre trying and not actively throwing Iām happy to have one in my team if there are no smurfs (which never happens these days).
Personally Iām over a month without a game and havenāt been tempted to play (except for missing my friends, we still talk just not the same)
You can see other playerās MMR by what rank they are at. If they are high gold 5 or low gold 5 does not really make a significant skill difference. In my experience most ranked games were within +/- 1 division so skill variance is kept low.
If this is QM then yes, it will within reason throw whoever together for matching speed. However QM is not a serious competitive game mode, QM is not ranked.
Or they were somehow placed above their skill level so are slowly falling down to the correct MMR.
Agreed, not everyone who has a very low winrate is a troll. I have a friend who is over account level 3000, he plays the game almost daily and has an overall QM WR of around 46%. Heās not really a terrible player, but heās incurious when it comes to seeking out information to improve, you wonāt find him here on the forums complaining or reading them.
More to the point is he just likes to play the heroes he enjoys. Heās very skilled with some tanks and could easily improve his overall WR if he only played tanks. The rub is he doesnāt enjoy playing tanks. Heās a QM main and plays heroes he enjoys such as his most played Genji. Heās has a 42% WR with Genji over thousands of games, but the important takeaway is he has fun playing Genji, even if he loses more often with that hero as winning or losing isnāt his primary motivation for playing HOTS.
I have encountered many players with the same attitude, but it doesnāt mean they are trolling, rather they are dedicated casuals without a competitive mindset.
Of course he is a terrible pepega who you donāt want to have in the team from a competitive perspective.
Because if you have played several thousand games and still sit around 46% you are either just bad or donāt try to win.
=> If you know that you canāt handle Genji and sit at 42% you are literally sabotaging the game for 4 other persons.
Exactly, for many players winning isnāt tied to their personal enjoyment of the game. I know some people who play Hots mostly to role play as their favorite NPC characters from other games.
The majority of these players are QM, ARAM or VS AI mains, so they arenāt going to hurt games in ranked, which one could argue is the only mode people should really take āseriously.ā
Well and for the 4 others persons itās unfun to play with a 42% WR Genji that canāt handle the hero even after several thousand games. To argue with personal fun, while ruining the match for 4 other persons is utterly selfish. Your friend is master piece of potato example.
I said as much without the name calling. Yes heās a potato, but heās not hurting anyone as he is a QM and ARAM main, no points are on the line in those modes that arenāt classed as ācompetitive.ā
He hurts the other 4 players in the team. Just try to understand what it means to play a team based MOBA game. The fact that he is your friend doesnāt make anything better. He wouldnāt hurt anyone in AI games, but QM is the spare time of 10 humans.
Perhaps. What should be well known to you is that if Hots had a large enough playerbase, he would only play games with potato players like himself. We donāt live in such a perfect world, so QM uses MMR averaging, ARAM has no MMR.
Itās not designed to be competitive, it isnāt really meant like that even by players like me, but Storm League requires such a broad knowledge of heroes, flexibility and willingness to tank 113% of your matches, that it has a potential for wild results.
I totally expected myself to rebound from Silver 5 but no, got my butt flattened so bad yesterday, while playing Auriel after a 2:1 QM result (also not stellar rated, but still). So clearly, I am like bronze 18 with like half the pool that I play regularly. Well, regular means one try a week. Iām like 14:24 this season Was Gold 5 last season but well, times change.
BUT what is worse is normally when this happens to me I get smurfs on the other team.
Was Gold 5 last season but well, times change.