If you think matchmaking is rigged, read this

False. It is unlikely at best, and certainly not known, that any random number generation is used. Most likely, you are matched against the opponent whose MMR (or rank) is within tolerances AND has been waiting in queue the longest time in milliseconds. (Tolerances widening as time in queue increases.) Neither of these things is random.

Now you might be thinking that time spent in queue is an archetype neutral decision. But it could be more complicated than this. For example, it’s pretty well known among top Arena players (like my former self) that playing during “normie hours” means a significantly softer field with easier wins, and that if you queue up at 3am you’re going to have a rough time of it against some serious players. Arena, however, uses your in-run win record and not a global MMR. What effect does time of day and, correspondingly, the average skill level of players have on Ranked, where the algorithm’s job is to match players according to skill? I don’t know. But it’s probably got some kind of weirdness to it that would make the Hearthstone version of Freakonomics, if there was such a thing.

This is all correct so far.

Let’s back up a bit.

I simulated the effects of a MMR based matchmaking algorithm, on a starting position that was randomly generated paper rock scissors opponents, and the result was that it sorted the players by “deck” choice.

The algorithm does the exact opposite of random. It unrandoms. The more time it’s given to work, the more unrandom the initially random distribution becomes.

If randomness is matter, then a MMR based matchmaking algorithm in a paper rock scissors metagame is antimatter. The algorithm touches randomness and a little bit of that randomness just flipping disappears.

There are apparently two points of view I’m trying to argue against here. One is that algorithm Y is necessary to generate observable results when algorithm X is predicted to generate the exact same results.

The other view is that matchmaking is random. Its function is not to match you with a random opponent, but a worthy one. It is not “restrictive” to say that this goal is the polar cussword opposite of random.

I think a better example you can use is Grinding in games. Grinding is usually a more negative experience where it’s sweeten by a chance of payoff.(rewards, loots, etc)

This type of negative designs have very specific targets. E.g. an item to shorten the grind, instant upgrade packs, etc.

Ooo that’s true as well. I don’t even know what some of this new stuff out is for example Runestones. Don’t have any, don’t plan on purchasing any either lol.

The game is different Reaver, an somehow a lot of the same. One thing that has stayed the same is people still to this day an age claim the game is rigged for example lol.

It is in a technical stand point. But the game is not rigged to make people lose. That’s the best thing to tell these people.

The game is not purely random based on MMR. The game is rigged to some extent for situational specific purposes.

More or less, people with the same rank such as Diamond 5 are considered to be on the same skill level “at that moment”. So, it would be fair if the match making system “randomly” pick any opponents from the same or closest rank in the waiting pool to play against you. If the system does only this, I would call it “fair”, and so do others.

However, the match making system does more than that. It tries to promote some pattern behaviors to encourage players to feel something for business agenda purpose. Sometimes, it gives you easy opponents from the “same” ranking pool to make you feel “fun” and “happy”. Sometimes, it gives you counter opponents consecutively to make you feel that your decks are not good enough. Thus, eventually and statistically, it hopes that you feel the need to pay some money to improve your decks. If you have played the standard mode long enough such as straight 6 hours, you will notice the pattern of winning and losing by yourself. The system does not only just match you with players of the same rank but they also try to keep your win rate as close to 50% as possible.

I have no evidence on my claim as Blizzard has never fully revealed their match making algorithm. However, you can try some experiment if you are willing to, but it would need extraordinary amount of time.

Try quitting Hearthstone for 3 months and then come back. I guarantee that the 1st match you play, you will win, as long as you play as best as you can. The system is happy that you come back so that it wants to make you happy too and hopes you stay with it. :upside_down_face:

1 Like

I advocate presenting one’s believe thru proper data as a basis.

This never ending repeating discussion is always not useful as there is no actual proper data or wrong set of data or wrong analysis applied.

I think the OG poster tried, along with a few others.

If you want data, I don’t have any on this laptop I am using. Example like a decktracker installed.

An then you have the whole debate on what counts as data. People will tell you that you need to show 1,000’s of games or etc.

I know my own experience from playing the game. I understand random can equal crazy things, but the game has something else beyond MMR.

Things I have tried since playing this month on HS is this.

I tried adding individuals from ranked to ask if they had bonus stars still. As I have destroyed my legend rank playing a very, very badly Undead OTK Warlock with the new 8 cost legendary.

An nobody despite the low ranked Diamond players have yet to have bonus stars. Which would have proven a falsehood.

There is something I still find gross from earlier that happened to me, an that’s a legit Blizzard Bot. Not a random bot account, an A.I. from Blizzard themselves. IN RANKED MODE FROM LEGEND.

Now we could spin this like “Omg guys, look blizzard has A.I. they can match you with an force you to lose”! Or we could be adults an say hmm… That’s so odd you have destroyed your MMR that legend players an diamond players arn’t my random opponent. Why would they do this? Queue times? The queue times are nothing compared to how awful wild was years ago.

I don’t know Reaver. It’s no different then when I play Warlock with over 1,000 dubs an try a class like DH with less than 20 wins an the game has this odd matchmaking take place for a game or two.

I think the best I can come up with is that the game was designed years ago to allow new players to fight new players. We both know this was an update to the game. Does it factor in how many wins with a class before you get meta? Does it have anything in place for a returning player?

I just dunno. It’s a lot easier to ask questions than to have answers.
As far as proof goes I atm only have screenshots I can show you about the confirmed Blizzard A.I. Since I took 3 of them.

That doesn’t make the game rigged. It does make the game quite sad to think someone like myself can be in legend an have a candycane win if you will be gifted to me while others lower ranked struggle.

What is your stance on the matter? Someone like you who wants proper data must have formed an opinion on this subject by now right?

What stance do you mean specifically?

For MMR, the above had describe it quite well. (But need to dig thru the long list of posts)

As I have learnt from my previous rigged thread, it takes more than the 1,000 games to be reliable.

Yet, for just discussion purposes on this forum, just a 100 games of raw data would be a good start point.

My believe is that if there is rigging involve, then it is easily captured when we look thru those raw data.

It would only be relevant within the same game. You’re smarter than this…

Ol’ Sc0tieMcB, too pompous for a TL;DR. :stuck_out_tongue_closed_eyes:

Busy right now, bookmarked and will read soon. I’m sure it’s an interesting take, per usual.

Also, iPhone autocorrect is a joke.

This could be a explanation for whenever you bought packs and crafted a new deck, that this deck will rock a few days and after the “honeymoon phase” you cant win anymore untill , you buy craft something new.

Found this in the Blizzard Matchmaking Engine Patent:

" * .

  • [0173]

In an implementation, when a player makes a game-related purchase, microtransaction engine 128 may encourage future purchases by matching the player (e.g., using matchmaking described herein) in a gameplay session that will utilize the game-related purchase. Doing so may enhance a level of enjoyment by the player for the game-related purchase, which may encourage future purchases. For example, if the player purchased a particular weapon, microtransaction engine 128 may match the player in a gameplay session in which the particular weapon is highly effective, giving the player an impression that the particular weapon was a good purchase. This may encourage the player to make future purchases to achieve similar gameplay results.

In other words you could be matched against favourable matchups, where your new cards shine and if the system thinks you need to buy new stuff it matches you against hardcounters with better shiny legendaries, till you give up and buy more stuff out of frustration.

So if you are on a loosing streak just buy some packs and craft an legendary and you will be good for a few days :slight_smile: so you can use the engine in your favour.

Not to be snarky, although, that’s a patent a) for Call of Duty, and b) has never been proven to have been used. I understand the impulse, but the logic does not support that this patent (or anything like it) is in-game. It’s most appropriate to remain agnostic here until further evidence (if there is any) presents itself.

Besides, this isn’t a topic about matchmaking.

You think it doesnt work for Heartstone aswell? It doesnt matter if there is proove, all that counts is they have it and potentially use it.

And for sure they use it, theres no need for prove, if you can use it you do so.

What i learned from the patent is:

Loosing Streak = buy packs and craft and the game will reward you for your ingame purchases.

Loosing Streak: Stop playing for a day or two and the game will reward you that you log back in and gives you easy opponents / good matchups.

This is a pattern i experienced myself.

Oh the i have misunderstood this, my apologies.

My apologies, I think I thought I was in another topic altogether. My bad, man.

1 Like

Holy cow 11k posts, no wonder you maybe got confused in which thread your :slight_smile:

I had been overactive for some time. As of late, my presence both here and ingame have been dwindling alas.

11k posts and… I can’t even post links nor images.

You’re still 100% wrong about everything patent related.

I dont know nothing you are right. Like i said before in the end it doesnt bother me if its rigged or not i enjoy this supreme f2p title :slight_smile:

Matchmaking random till win streak win streak reaches (X). Then its time for arranged marriages. :stuck_out_tongue:

Wrong. My son joined the military, was gone for 6 months, came home for a few weeks, got the gifted packs bundle etc for returning, chose beast hunter for his free deck, and I coached him on how to play it. He got demolished on his first 5 games and I’m not some noob player.

Your theory went straight out the window and your “guarantee” of winning idea didn’t work at all.

Looks like you need to go back to the drawing board with a new hypothesis of rigging and how some forced happiness works lol.

2 Likes