Matchmaking does NOT seem Random to me (with explanation)

‘Obviously’. :rofl:

There are plenty of adepts of the cult of so-called ‘Matchmaking’ and ‘MMR’, yet I haven’t seen anyone present something — at least a formula.

Btw, Mercs are a different mode, but there’s at least a ‘MMR’ number presented to players, no matter how horrible the PvP system there might be (and that’s an opinion by one of the most avid Mercs afficionados here) — and the way it works is a total mess, to put it very mildly, that bogus number being indicative of essentially nothing — apart from the amount of time/games spent there (by the way, the top of the ‘rating’ are bots farming it non-stop).

Oh, the ‘player’ factor plays a vanishigly small part here, noticeable only over a large number of games, perhaps. It’s mostly factors like decks (budget/starter vs optimised ‘meta’ etc) and so on.

Dunno what ‘101’ is supposed to mean, but go on, enlighten me about that algorithm. Formula!

Oh, and I know of things such as Elo’s rating and so on, which works very well where it’s applicable. What’s going on in this game is apparently some sloppy mess, hidden behind obscurity, and some mumbo-humbo fancy words about some supposed MMR, which probably measures objectively nothing.

Supposedly not:

Casual mode, Ranked at Legend rank, and non-cooperative Tavern Brawls determine pairings using each player’s matchmaking rating (MMR).[1] [2] Ranked play without a Star Bonus is determined by each player’s rank …

https://hearthstone.wiki.gg/wiki/Matchmaking

So, in the beginning of a season, those ‘pros’ should be separated from beginners by that ‘MMR’ magic.

However, lemme tell you that in the recent years I’ve played from the very bottom all the way to Legend without a star bonus at least twice (in Classic and Standard respectively), having been inactive for so long that my supposedly oh-so-high ‘MMR’ would be erased or irrelevant for the reasons listed above. Gotta tell you: there’s practically no visible difference in terms of opponents’ skill, decks and so on, except for there being a bit more budget, beginner or ‘unconventional’ decks at the very, very bottom (below Gold or such).

The only noticeable difference in ‘skill’ is like that:

Yep, that’s indeed one of the moment where remaining silent is better than revealing one’s ignorance (well, at least I adhere to this opinion), including via giggling at what one can’t comprehend, because:

That’s just factually wrong, for it can be easily disproved by counterexamples, such as this one (yes, posted it already many times, but what of it — it’s one of my favourite ones):

https://www.youtube.com/watch?v=tav_ttqyv9Q

It’s a rather crude bot, playing very sloppily, vs a skilled player.

Only in theory, and that’s how it’s supposed to be.

Even if the actual matchmaking is not rigged, it’d better be :grinning: , since otherwise it’s just a sloppy mess.

Hmm, speaking of ‘normal’ — what’de be approximately normal (literally so) is the distribution of players by win rate (if you’re not familiar with Gaussian, or normal, distribution, I’d suggest reading up if you’re interested). That is far from the ‘all are equal’ picture, i.e. a nearly-rectangular, or ‘supergaussian’, distribution. So, in order to achieve the latter, you’d need to actively meddle in order to flatten the ‘bell curve’, as it is colloquially referred to, of the normal distribution, which would be expected to occur ‘naturally’.

Because it’s more efficient, besides, because they might aready have — having demonstrated at least two bits of technology with potential applications for rigging, I’m inclined to assume they’d explore a third one as well — seems more in line with the course of action they’ve taken.

Only that person can specify what he or she meant, I can’t answer for them, but I can provide my reasoning or my answers to your question, since I’ve found it interesting.

Yes, probably so — as said, I’ve correlated it with their advances in technology (functions like ‘Zephrys’ or ‘Optimotron’) and AI (for example, at some point the one used for single-player bosse was improved substantially, see, for example: Heroic Karazhan --- Free Medivh! in August 2022). In any case, I’ve described the way I’d probably approach and implement (in practice, not just with general concepts) rigging the game if I were tasked with it.

Just because some theories about the game being rigged are amateurish or wrong, it doesn’t mean that so is the notion in general.

In brief, I’ve touched three different aspects here.

  1. Is it likely that the game is rigged?

My answer is yes for roughly the reasons that the topic-starter listed.

  1. Is there a motive, an incentive for them to do so?

My answer is yes — see the part about ‘mass appeal’.

  1. Have they got the means?

My answer: yes, with their advances in AI and some technology with such applications already having been implemented and demonstrated.

1 Like