50% winnrate bullcrap

Not really.

Good to know, that’s what the matchmaker is in general when it is QM and difficulty depends on the drafting process that was went for in Un/Ranked, whilst fun is a subjective metric so we cannot really specify what is fun and what is not.


Anyways

Do you think using one metric of source for the past 3 years is quite questionable?

1 Like

You are basically asking for less reliable source than this.

Preposterous, I just want more sources than one, is that hard to ask?

Trick question, so who cares?

It’s ur problem u wanted more.

Or… hear me out, it makes your argument better and stronger, but sure pal, whatever suits you lol.

2 Likes

lol ok whatever dude bro mate friend

1 Like

Were are your so called facts cause i cant find them anywere. But just keep on useing the whiteknight card you so dearly love to use casue you cant win an argument without :smiley:

So just because you only see the same people post here means we are all whiteknights. i’m done talking to you. Ignored from now on. OP is a weak minded quitter that think game is purposly dragging him down when he is properly responsible for a lot of his loses himself casue he refuse to get bettter. But you can call us all whiteknight and blizz fanbois as much you want. You no better yourself nor are all the others who defend this weak-minded quitter with sword and shield like the knight in shining armor you are yourself.

1 Like

Several 70+% winrate solo smurfs disagree with your forced 50% theory.

If they have that high winrate then they are properly also master rank smurfs that runs around beating players 4 ranks below him.

It also works the other way. I played a few matches last night with someone who has over 1200 games, and a 45% win rate. Checking their match history, they had managed to lose 3 in a row a few days ago. Against the AI.

This is the part that makes the system feel like there is some kind of weird forced win rate. A ton of people are getting stuck on the fact that the OP mentioned a win streak. Ignore the win streak - the fact of the matter is that a lot of players, including me, are put into the majority of games that are just stompfests one way or the other. It’s not fun to lose by getting completely stomped and it’s not fun to win that way either.

Again, if the system were properly identifying players’ skill ratings and putting them with and against players of similar skill, yes, we would still expect a 50% win rate, but we would also expect the majority of games to feel evenly matched.

50% of the time , his prédictions are correct 100% of the time !

You’re a bit confused… A white knight is defined by blind loyalty, and a willingness to accept what the authority says, at face value-- Even if there is evidence to the contrary. An example of contrary evidence being the frequent MMR discrepancies between teams in QM, which you wouldn’t know about without 3rd party sites like HotSLogs.

An echo chamber is a situation in which beliefs are amplified or reinforced by repetition inside a closed system (these forums) and insulated from rebuttal (shouting down anyone who disagrees).

The two things are not cause and effect

I just find this very ironic considering all you wrote.

Firstly, you could call me a black knight (or w/e the opposite of white knight would be), and I wouldn’t find it insulting… Second, I am not arguing on the OP’s behalf, I am arguing for transparency.

I frankly don’t understand why you have taken my posts to be personal attacks against you. : /

1 Like

The reason everything ends up as personal experiences is firstly due to people being ANGRY enough to post about it, and secondly due to any outside data being dismissed.

Yes, ofc personal experiences are going to have “loop holes” and yes, we will likely only ever hear one side, but that does not make those experiences inherently wrong.

MAYBE there is a language barrier here, but if you are meaning to write as you do-- Then you are writing in terms of absolutes here. And the trouble with that, is that reasonable doubt, or a single example, is all that is required for the entire statement to be wrong…

I have a friend, who at the time of this writing, just achieved a 90% win rate in diamond. The thing is, 2 seasons ago, they had been stuck in low plat-- and I mean stuck in low plat for years. They are a one-trick Chogall, and their next highest win rate (hero-wise) isn’t even 50% atm.

If the system is expected to maintain a 50% balance, unless you belong in a higher league than you are, then it has clearly failed in some cases (such as in the case of a supposedly OP hero). So I would question why said system is not open to criticism… I am speaking broadly ofc.

1 Like

Hotslogs is not a good site for checking mmr disparities.

  1. Their mmr starting point and rate of change per game doesn’t coincide with what HotS actually uses (k-factor in elo math) and there are obvious differences that can be seen between the two.
  2. Hotslogs make seasonal adjustments in modes that don’t do seasons (ie, quickmatch) If you look at the mmr change before a season, and after, without games played, the mmr value will be changed because of the transition of seasons.
  3. The stability of hotlsogs is based on people uploading their own data. While the idea follows that enough games played will eventually approach parity with in-game numbers, when the site was reset and lost all their old data, that’s context they won’t get back because their k-factor is assuming all the data they get from that point on is from ‘new’ accounts.

3a. When ranked transitioned to storm league, some sites assume players are ‘starting over’ instead of transitioning their mmr. So they can project bronze-ranked mmr for players in diamond, which is a very big disparity.
4. Hotslogs only knows that a match was made, it didn’t make the match. So it doesn’t know ‘why’ players were matched at certain ranges, it just tries to overcompensate for the match, thus making its k-factor projections even more skewed. Since players can form parties without mmr limits, players are allowed to do things the system cannot “expect” and force how that game is matched. Similarly, hotslogs doesn’t know how long the game took to match, so the parameters loosening does affect stuff the site can’t show, or calculate.

While you may had said “oh, third party site information gets throw out cuz it’s not the real information” there is an important distinction on ‘why’ the information isn’t useful in where it is trying to be applied. There’s a difference in using “what we have” (third party sites) and using it with knowledge applications of how that information works.

When threads have used hotslog mmr values for show ‘forced matching’ they don’t acknowledge or account for the flaws in the system: they assume/declare it has to be true, and don’t account for bias in the stats – there are tests for that sort of thing – and because they think the system is ‘forced’, they’re not checking if it’s ‘true’, they’re just looking for anything that resembles confirmation of the claim.

So they don’t care about obvious flaws or try to improve what they have to use, they just declare it as “it’s all we have” and… it’s not. That’s part of the issue of the narrow-perspective stuff I wrote above: a lot of posters don’t even consider stuff, so of course the consequences of things they don’t know doesn’t occur to them.

2 Likes

I achieved 75% win rate the moment i entered master rank up until 1.3k points and that’s all games I played in Storm League when I achieved that number.

And currently its downscaling to more reasonable win rate, inevitably anyone would have climbed will get a chain of losses not because the game wants to, but simply as put your climb and winrate streak was so hard that the game and how is played changes dramatically, and depending on how you can handle such environment lies on you and your teammates, while of course the many times i specified: no one tilts, no one griefs, perfect game (for the most part).

If MM is unable to balance people based on their performance then the game would be considered unbalanced, right?

But that is also because humans are just a learning creatures, so if someone hard stuck in silver learns the basics and now is a diamond is due the fact they taken the step to learn and improve on themselves, and thus a win rate spike happens as a result due physical (and mental) improvements so MM has now to consider this new knowledge and try to balance the player out with the other competing players in that tier.

With no context we really can’t say much how someone can achieve for being hardstuck platinum a 90% win rate to diamond, you should elaborate more over this.

For my own case, I just simply climbed because the skill set and mentality in such ranks are working in my favor, and they are quite lackluster… seriously I just solo carried myself out of lower tiers from just playing 4 heroes, Lucio, Lunara, Valla or Sylvanas, followed by manly going Sylvanas in Diamond until I reached Master from huge win streak.

Tbh, I am having trouble understanding what you are getting at… HotsLogs = bad because it’s inaccurate, and for people to claim “it’s all we have,” as far as data goes, is just those people being lazy and not wanting to improve? Do I have that right?

You expect them all to write their own programs to calculate not only their MMR, but that of all the people they are playing with over thousands of games? … I’m not meaning to make fun of you when I write that, but I’m not sure what you are expecting. Let me ask you this, what does Blizzard’s data say on the subject? … … …

Not to “defend” HotsLogs, but it gives you a range, and while that range may not use the same metric that Blizzard does, the range itself is not inaccurate in terms of an independent gauge.

1 Like

To be fair HotsLog has been in a downward spiral the moment they decided to shutdown the site and placed “under construction”, everyone now uses HeroesProfile as a data source and the amount of replays sent in comparison to HotSLog shows.

You want better context? I could give you the perfect replay to illustrate everything-- But I am afraid some overzealous forum mod would view any link as “naming and shaming” given the way we have been talking about this… Do you want to give me an email or something I could send it to? Its 0.99mb, not sure I could send that over discord.

1 Like